- Email: [email protected]
Doping effects of surface functionalization on graphene with aromatic molecule and organic solvents
Accepted Manuscript Title: Doping Effects of Surface Functionalization on Graphene with Aromatic Molecule and Organic Solvents Authors: Guangfu Wu, Xin Tang, M. Meyyappan, King Wai Chiu Lai PII: DOI: Reference:
S0169-4332(17)32037-8 http://dx.doi.org/doi:10.1016/j.apsusc.2017.07.048 APSUSC 36582
To appear in:
APSUSC
Received date: Revised date: Accepted date:
8-3-2017 26-6-2017 5-7-2017
Please cite this article as: Guangfu Wu, Xin Tang, M.Meyyappan, King Wai Chiu Lai, Doping Effects of Surface Functionalization on Graphene with Aromatic Molecule and Organic Solvents, Applied Surface Sciencehttp://dx.doi.org/10.1016/j.apsusc.2017.07.048 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Doping Effects of Surface Functionalization on Graphene with Aromatic Molecule and Organic Solvents Guangfu Wua, Xin Tanga, M. Meyyappanb, and King Wai Chiu Laia,* a. Department of Mechanical and Biomedical Engineering, City University of Hong Kong, 83 Tat Chee Ave., H.K. SAR, China b. NASA Ames Research Center, Moffett Field, CA, USA
* Corresponding author. Tel: +852-3442-9099; Fax: +852-3442-0172 E-mail address: [email protected]
1
Graphic Abstract:
We have investigated the doping effects of surface functionalization and its influence on the carrier mobility of graphene.
Highlights
For the first time, the doping effects of the surface functionalization on graphene with aromatic molecule and organic solvents were investigated by Raman and electronic properties. Raman spectra showed that both PBASE and these two solvents imposed doping effects on graphene. Electrical measurements further revealed that PBASE imposed a p-doping effect while DMF and CH3OH imposed n-doping effect. CH3OH causes a much smaller reduction in the carrier mobility of G-FETs than that of DMF.
Abstract
Aromatic molecule functionalization plays a key role in the development of graphene field-effect transistors (G-FETs) for bio-detection. We have investigated the doping effects of surface functionalization and its influence on the carrier mobility of graphene. The aromatic
2
molecule (1-pyrenebutanoic acid succinimidyl ester, PBASE), which is widely used as a linker to anchor bio-probes, was employed here to functionalize graphene. Dimethyl formamide (DMF) and methanol (CH3OH) were used as two solvents to dissolve PBASE. Raman spectra showed that both PBASE and these two solvents imposed doping effects on graphene. The PBASE was stably immobilized on the graphene surface, which was confirmed by the new peak at around 1623.5 cm-1 and the disordered D peak at 1350 cm-1. Electrical measurements and Fermi level shift analysis further revealed that PBASE imposes a p-doping effect while DMF and CH3OH impose an n-doping effect. More importantly, CH3OH causes a smaller reduction in the carrier mobility of G-FETs (from 1095.6 cm2/V·s to 802.4 cm2/V·s) than DMF (from 1640.4 cm2/V·s to 5.0 cm2/V·s). Therefore, CH3OH can be regarded as a better solvent for the PBASE functionalization. This careful study on the influence of organic solvents on graphene during PBASE functionalization process provides an effective approach to monitor the surface functionalization of graphene.
Keywords: Raman analysis; Surface functionalization; Graphene; Aromatic; Doping effect
1. Introduction
Graphene exhibits remarkable properties1-2 such as high carrier mobility, large specific surface area and excellent electrical conductivity, which make it a candidate material to develop photodetectors,3-4 capacitors5 and sensors.6 Recently, the field-effect transistor (FET) structure
3
employing graphene as the sensing material has been shown to have potential for sensitive biodetection.7-8 Generally, the graphene surface needs to be functionalized with bio-probes to realize specific detection. In general, the bio-probes are anchored on graphene by linker molecules. Typically covalent and non-covalent methods are employed to functionalize graphene with linker molecules.9 Covalent methods can stably and specifically introduce functional groups on the surface of graphene, but they will inevitably alter the physical properties and native electronic structure of graphene by converting the sp2 carbons to sp3 ones. Therefore, non-covalent approaches are preferable in order to preserve the inherent properties of pristine grapheme; in this case, aromatic molecules such as pyrene and its derivatives have a strong affinity for graphene because of the stable π-π stacking across the basal plane.10-11 1-pyrenebutanoic acid succinimidyl ester (PBASE) is an aromatic molecule which has both an aromatic ring and a reactive tail and has been widely used as a linker molecule in biosensors.12 For example, glucose oxidase is introduced to the surface of graphene and carbon nanotubes by using the linker molecule PBASE.12-13 In these examples, the role of PBASE is to serve as an anchor for introducing recognition elements such as an antibody onto the nanomaterial surface. The attachment of PBASE onto the surface of graphene is based on the stable π-π stacking between graphene and the pyrene group of PBASE, which has been confirmed in previous studies.14 Since graphene can be regarded as an infinite aromatic molecule and pyrene has four π-π binding sites, the interaction between pyrene and graphene adopts parallel-stacking orientation to maximize the PBASE-graphene interaction.15 It was also revealed that the distance of PBASE-graphene stacking was about 2.3 Å.14 This value is smaller than that of C6F6-C6H6 π-π stacking (3.6-3.8 Å), which is ascribed to that the electron withdrawing group in PBASE reduced the repulsive electrostatic interaction between pyrene and graphene, enhancing the π-π interaction.15
4
Aromatic molecules consisting of electron-withdrawing or electron-donating groups could dope carbon nanotubes,16 but similar phenomenon for graphene have been rarely studied. Besides, organic solvents such as dimethyl formamide (DMF) and methanol (CH3OH) are necessary to dissolve the aromatic molecule for graphene functionalization. The effects of these organic solvents during aromatic molecule functionalization process on the graphene properties have not yet been investigated. The graphene field-effect transistor (G-FET) is utilized here to explore the doping effect of aromatic molecule PBASE and organic solvents on graphene. The graphene layer serves as the conducting channel bridging the source and drain electrodes. A solution-gate electrode is usually employed in the G-FET to tune graphene’s Fermi level, and to obtain the transfer characteristics that reveal the doping condition (or Fermi level shift) and the carrier mobility of graphene.17 Besides, Raman spectroscopy is an effective method to investigate the inherent physical properties of carbon nanomaterials.18-19 The characteristic D peak (~1350 cm-1), G peak (~1580 cm-1) and 2D peak (~2685 cm-1) in Raman spectra are able to reveal the number of graphene layers, the defects in graphene and change in carrier concentration (electron or hole).20-21
In this paper, the doping effects of the aromatic molecule PBASE and two different organic solvents (DMF and CH3OH) on graphene are investigated carefully by monitoring the changes in Raman spectrum and the shift of Dirac point voltage of the G-FETs during the chemical functionalization process. The Fermi level shift of graphene is also estimated by these two methods. The results indicate that the PBASE imposes a p-doping effect on graphene because of its electron-withdrawing property while DMF and CH3OH impose an n-doping effect. Besides, CH3OH causes a smaller reduction in carrier mobility of G-FETs than DMF. This work
5
provides a two-in-one method to explore the effect of both the solvent and aromatic molecule on graphene.
2. Materials and methods
2.1 Fabrication and characterizations
2.1.1 G-FETs fabrication process
The fabrication of G-FETs is described in detail in Figure 1. CVD-grown graphene on copper foil was first spin-coated with a thin layer of PMMA. Then, the PMMA/graphene/Cu film was immersed in FeCl3·6H2O solution (1 M) until the Cu was completely etched away. The released PMMA/graphene film was washed by deionized water and transferred onto a SiO2/Si substrate. After removing the PMMA layer with acetone, patterned structures were made on the surface of graphene by photolithography, and the patterned graphene was transferred onto another Si/SiO2 substrate. The source and drain metal electrodes were created at the two ends of graphene film (1 × 1 mm2) by using silver conductive paint. Finally, silicone rubber was attached on top of the substrate to insulate the electrodes (source and drain) and to confine the liquid solution during experiment. The inset represents the optical image of the G-FET.
2.1.2 Characterizations
The as-fabricated G-FETs were functionalized with PBASE in two different organic solvents (DMF and CH3OH). The detailed functionalization approaches can be found in the supporting information. A schematic of the functionalized G-FET structure is shown in Figure 2. The graphene functionalization was characterized by Raman spectroscopy and electrical
6
measurements. All Raman measurements were conducted under room temperature (Renishaw Raman Spectroscopy 1000B) with 514 nm laser excitation. All electrical characterization was conducted using a semiconductor device analyzer coupled with a probe station (Semiconductor Analyzer, B1500, Keysight Co. Ltd., CA, USA) under room temperature. The aromatic molecule functionalized G-FETs were biased at 100 mV for all measurements, and the gate voltage was applied to the G-FETs via an Ag/AgCl electrode that was immersed in PBS buffer solution (0.1 mM, pH 7.2).
2.2. Carrier Mobility
The carrier mobility, which indicates the ability of the carrier (hole and electron) to drift in a material, is the essential feature of electronic devices.22 Capacitive equivalent model, which has been used to describe the capacitive property of the solution-gate G-FET,23 is employed here to estimate the carrier mobility of G-FETs integrated with a solution gate during the functionalization process. The details can be found under Supporting Information. It is assumed that the total gate capacitance (CTG) is a series of electrostatic double layer capacitance (CG_EDL) and quantum capacitance of graphene layer (CQ). This model is expressed by the following equation:17
CTG
CG EDL CQ CG EDL CQ
(1)
where the quantum capacitance of graphene layer CQ was estimated as ~2 µF·cm-2 based on previous work.24 CG_EDL is the capacitance of electrostatic double layer (EDL) that forms between graphene and the solution interface. For a typical back-gate G-FET, in which graphene
7
layer was put on dielectric layer (300 nm SiO2), the back gate capacitance (CG_BG) is simply described by the equation below:25
CG _ BG
SiO 0 2
(2)
d SiO2
where εSiO2 is the relative permittivity of SiO2; ε0 is the vacuum permittivity (ε0 = 8.854 × 10-12 F·m-1); and dSiO2 is the thickness of SiO2. For the solution-gate G-FET, the EDL, which works as a top-gate dielectric layer when a voltage is applied via the gate electrode, can be modeled as a parallel plate capacitor. This assumption has been used to estimate the electronic properties of graphene devices,26 and the thickness of the EDL can be calculated according to the DebyeHückel equation.27-28 In this work, the capacitance of EDL can be expressed as below:26
CG _ EDL
EDL 0 d EDL
(3)
where εEDL and dEDL are the relative permittivity (~78)29 of 0.1 mM PBS solution and the thickness of EDL (~7.7 nm for 0.1 mM PBS).30 The CTG, which is obtained from the capacitive equivalent model, can be used to calculate the carrier mobility of G-FET. The carrier mobility (hole or electron), in the linear region of transfer curve, can be calculated based on the following equation:17
I ds l 1 Vg w CTG Vds
(4)
8
where the length to width ration (l/w) of graphene channel was set to 1 in this work; Ids is source-drain current; Vg is gate voltage; and Vds is source-drain voltage.
2.3 Fermi level shift of graphene
The Fermi level for pristine graphene is located at the Dirac point; for doped (p-doped or n-doped) graphene, the Fermi level will shift away from the Dirac point.31 As shown in Figure 3, the Fermi level shift (∆EF) of the doped graphene is defined as the energy difference between the Fermi level (EF) and the Dirac point.
In this study, the Fermi level shift of graphene was estimated by two methods. One involved the G peak energy (ωG) of graphene in the Raman spectrum. For pristine graphene, the G peak is located at ~1580 cm-1, and both p-doped and n-doped graphene have right-shifted G peaks.21 Therefore, the absolute value of the Fermi level shift (│∆EF│) with respect to the Dirac point, is expressed as below:32
EF
G 1580cm1 42cm1 eV
(5)
The other one involves electrical measurement of the G-FET. The Fermi level shift of graphene in this case is expressed by the following equation:33
EF sgn(n) vF n
(6)
9
where νF is Fermi velocity of graphene; ħ is reduced Plank constant (1.0546 ×10-34 J·s); carrier density n is defined to be positive for holes, and negative for electrons. For solution gated (top gate) G-FETs, the carrier density of graphene is estimated by the following equation:34
n
CTG Vg e
(7)
where CTG is the total gate capacitance as shown in equation (1); e is elementary charge; and Vg is the gate voltage at Dirac point (VDP).
3. Results and discussion
3.1 Raman spectra
Raman spectroscopy was used to characterize five types of graphene: (1) untreated graphene; (2) DMF functionalized graphene; (3) CH3OH functionalized graphene; (4) DMF/PBASE functionalized graphene; (5) CH3OH/PBASE functionalized graphene. The changes in the characteristic bands were carefully examined during the functionalization process with different chemical treatments. In order to minimize the errors due to the variations in graphene samples, the data from 12 sampling points of the graphene sample before and after functionalization were averaged. The graphene sample size was 1 × 1 mm2 as shown in Figure 4.
DMF and CH3OH were chosen as candidate solvents for the functionalization experiments, and the Raman spectra of graphene before and after PBASE functionalization in these two solvents are shown in Figure 5. The black curve represents the Raman spectrum of untreated graphene, blue and red curves represent PBASE functionalized graphene in DMF and
10
CH3OH, respectively. The shift of G and 2D peaks and their intensity ratio (I2D/IG) are the indicators of doping effect on graphene.18
In Figure 5a and 5b, the positions of the G peak and 2D peak of untreated graphene are consistent with reported values.20 The small variations of these two peaks are attributed to the PMMA residues resulting from the fabrication process.35 These are few layer (one or two layers) graphene samples because the initial value of I2D/IG (the intensity ratio of 2D and G peak) is larger than 2.036 as shown in Figure 5. For the DMF and CH3OH-treated graphene, both G and 2D peaks exhibit an obvious left-shift as shown in Figure S3. The left shifts are attributed to the doping effect arising from the solvent molecule due to the existence of N in DMF and O atoms (electron donor) in CH3OH. The decrease in I2D/IG of the DMF and CH3OH-treated graphene further confirms that both DMF and CH3OH impose doping effect on graphene.
For PBASE functionalization in DMF (or in CH3OH), as shown in Figure 5c and 5d, a new peak located at 1623.5 cm-1 (1621.7 cm-1) appears after PBASE functionalization. This peak is attributed to pyrene group resonance,14 which directly proves that PBASE is attached onto the graphene surface. In a previous study, the number of the π-π stacking binding sites in polycyclic aromatic molecules was revealed to affect the Raman intensity significantly when they were adsorbed on the aromatic polysodium styrenesulfonate (PSSS) templates via π-π interactions.37 We believe that the published study will help us to explain the attachment of the aromatic molecule PBASE onto the graphene surface. Since graphene has many π-π binding sites, we think that the aromatic molecule PBASE can be immobilized on the graphene surface stably via π-π stacking. Meanwhile, slight right-shifts in both G and 2D peaks are observed compared with DMF and CH3OH-treated graphene, which are ascribed to the doping effect of PBASE on
11
graphene. The G peak (~1580 cm-1) is corresponding to a doubly degenerate phonon mode at the Γ point. The 2D peak (~2685 cm-1) results from simultaneous excitation of two phonons with wavevector close to the K point in the graphene Brillouin zone. Doping of graphene decreased the probability of excited charge carrier recombination.38 The decreased probability afforded non-adiabatic photon perturbations, which removed the Kohn anomaly and increased the energy of the phonons at the Γ point and K point. The increased phonon energy resulted in the Raman G and 2D peaks with higher frequency and larger wavelength. Therefore, the right-shifts (larger wavelength) in the G peak and 2D peak can be observed in Raman spectrum of PBASE treated graphene. This phenomenon has been discussed in the previous studies.18, 21 The I2D/IG ratio shows an increase after PBASE functionalization in DMF and CH3OH as shown in Figure 6. The right shift in both G and 2D peaks and the decrease in I2D/IG indicate that PBASE imposes an opposite doping effect on graphene which is attributed to the existence of the carbonyl group (electron acceptor) in PBASE. These observations of G and 2D peak shifts are in agreement with previous experimental results.39 Besides, the D peak for untreated graphene at 1350 cm-1 is not obvious, which indicates that few defects exist in graphene samples. The PBASE functionalization mildly increases the disorder in D peak, which is attributed to the orbital hybridization of the PBASE molecule with graphene plane.
3.2 Electronic properties of functionalized G-FETs
Although Raman spectrum can be used to confirm the doping effect on graphene, it is difficult to distinguish whether it is p- or n-doping effect. Moreover, it is difficult to obtain the detailed information on electronic properties such as carrier mobility by Raman spectrum. Therefore, we conducted transfer curve measurements on four groups of functionalized G-FETs:
12
(1) DMF functionalized G-FETs; (2) CH3OH functionalized G-FETs; (3) DMF/PBASE functionalized G-FET; (4) CH3OH/PBASE functionalized G-FET. The functionalization conditions and the experimental measurement conditions were kept the same for the timedependent measurement. The temperature was kept at 25 °C and the humidity was kept at 50%.
First, the transfer curves on DMF (and CH3OH) functionalized G-FETs are discussed. The transfer characteristics of DMF functionalized G-FET after different incubation times are shown in Figure 7. There are two noteworthy features: (1) the left shift of the Dirac point voltage (VDP); (2) the reduction of conductance of DMF functionalized G-FETs. As shown in the inset of Figure 7a, the VDP shifts from 0.24 V to 0.08 V. This provides evidence that DMF imposes ndoping effect on G-FETs.40 The conductance decreases dramatically during the DMF functionalization process, and we extracted the carrier mobility (hole and electron) from the transfer curves using equations (1)-(3). The results are summarized in Figure 7c. The DMF causes a significant reduction in hole mobility from 1640.4 cm2/V·s to 5.0 cm2/V·s (Detailed calculation of hole mobility can be found in Figure S4 and Table S1). The electron mobility shows a similar change with the increase in incubation time. The reduction of carrier mobility could be attributed to the charged impurity scattering caused by adsorption of the DMF molecules.41
Compared with DMF-functionalized G-FET, CH3OH causes a smaller left-shift of VDP from 0.22 V to 0.20 V as shown in Figure 7b. Moreover, in Figure 7d, the reduction of carrier mobility is from 1095.6 cm2/V·s to 802.4 cm2/V·s, which is much smaller than that caused by DMF. The reason could be that DMF has a larger molecular dipole moment (3.82 Debye) and polarizability (7.8 × 10-24 cm3)42 than CH3OH (1.7 Debye; 0.6 × 10-24 cm3).43 For DMF, the
13
larger dipole moment and polarizability contribute to a larger noncovalent DMF-graphene binding energy,43 making DMF on graphene surface more difficult to be removed during the functionalization process. Therefore, the DMF molecule is adsorbed more stably on graphene surface than CH3OH, causing a heavier scattering effect on the graphene device.14 Thus, the reduction in carrier mobility of DMF-functionalized G-FET is much larger than that of the CH3OH case.
We now discuss the doping effect of PBASE functionalization on G-FETs in the above two solvents with the transfer characteristics shown in Figure 8a and 8b. Unlike the continuous left-shift of VDP for DMF-functionalized G-FETs, PBASE (5 mM in DMF) functionalization causes a right-shift of the VDP from 0.22 V to 0.24 V in the first 30 min, and stays stable from 30 min to 60 min. From 60 to 120 min, the VDP shows a left shift (from 0.24 V to 0.18 V). Results discussed earlier showed that DMF imposed n-doping effect on graphene. Therefore, the rightshift in the first 30 min can be attributed to the p-doping effect of PBASE. The p-doping effect of PBASE is also evidenced from the transfer curves of the CH3OH/PBASE-functionalized G-FET.
Figure 8b shows continuous right-shift (from 0.21 V to 0.27 V) of the VDP of CH3OH/PBASE-functionalized G-FETs, which is quite different from that of DMF/PBASEfunctionalized G-FETs. Experimental results discussed earlier confirmed that both DMF and CH3OH imposed n-doping effect on graphene. Therefore, the n-doping effect from these two solvents and the p-doping effect from PBASE are competitive. The n-doping effect of CH3OH is slightly based on the previously measured VDP shift (from 0.22 to 0.20 V). Therefore, the pdoping effect of PBASE dominates the CH3OH/PBASE functionalization process. Besides, when
14
the incubation time increases, the stronger n-doping effect of DMF neutralizes the p-doping effect of PBASE, and dominates the DMF/PBASE functionalization process.
The carrier mobility (hole and electron) of the DMF/PBASE and CH3OH/PBASE functionalized G-FETs is summarized in Figure 8c and 8d. For CH3OH/PBASE functionalized G-FET, the hole mobility decreases from 1377.5 cm2/V·s to 1144.3 cm2/V·s, which is much smaller than that of DMF/PBASE functionalized cases (977.0 cm2/V·s to 76.3 cm2/V·s). These results are consistent with that of DMF and CH3OH functionalized G-FETs, which further suggests that CH3OH has a smaller influence on the carrier mobility of G-FET during PBASE functionalization process.
In order to investigate the influence of PBASE concentration, 10 mM PBASE was used to functionalize graphene devices in DMF. The transfer characteristics and carrier mobility of PBASE (10 mM in DMF) functionalized graphene devices after different incubation times are shown in Figure 9. This functionalization causes an obvious change of VDP, which shows a larger right shift (from 0.22 V to 0.26 V) in the first 30 min. The observed right-shift of VDP further confirms that PBASE imposes p-doping effect on G-FETs which is also observed for PBASE (5 mM in DMF) functionalized G-FETs. From 60 to 120 min, the VDP changes from 0.26 to 0.21 V which is smaller than that of 5 mM PBASE. The observed right-shift of VDP further confirms that PBASE imposes p-doping effect on G-FETs which is also observed for PBASE (5 mM in DMF) functionalized G-FETs. From 60 to 120 min, the VDP changes from 0.26 to 0.21 V which is smaller than that of 5 mM PBASE. Furthermore, the carrier mobility is still reducing after 240 min for 10 mM PBASE, which is different from that of 5 mM PBASE. The performances of Dirac point voltage shift and carrier mobility change indicate that the number of the aromatic
15
molecule PBASE has an influence on graphene during the functionalization process. The volume of PBASE solution was kept same for all functionalization experiments. Larger concentration afforded larger amount of PBASE molecule that took a longer time to make the PBASEgraphene interaction reach to the saturation. In Figure 7 and Figure 8, the carrier mobility became stable after 120 min for 5 mM PBASE because most of the PBASE molecules were attached onto the surface of graphene. Similar phenomenon has been reported in a previous study.44 The adsorption of antibiotics on graphene reached balance or complete absorption due to the low concentration of antibiotics. In Figure 9, 10 mM PBASE was used to functionalize graphene, the larger number of PBASE molecule caused the interaction to reach the saturation in a longer time. Therefore, the carrier mobility is still reducing after 240 min for 10 mM PBASE. This phenomenon is similar with that in other two studies.37, 45 Larger number of FAD caused the shift in the absorbance of PSSS-templated AgNPs in longer time, and larger number of rGO molecule costed a longer time to make the absorbance of PEG200 decrease to the stable state. The attachment of the PBASE to graphene by π-π stacking appears to have no effect on the conductance of the devices.46 Therefore, the reduction of conductance in the experiments could be mainly attributed to the adsorption of DMF.
3.3 Fermi level shift analysis
The Fermi level shift (∆EF) of graphene is regarded as an indicator for chemical doping.31 To further understand the doping effects caused by different dopants, we then discuss the ∆EF of graphene after DMF, CH3OH and PBASE functionalization based on Raman spectroscopy and electrical measurements. In Raman spectrum, functionalization with the above three chemicals causes the changes in G peak, which has been used to estimate the │∆EF│ of graphene.32 Based
16
on equation (5), the │∆EF│ of graphene after DMF, CH3OH and PBASE functionalization is summarized in Table 1. The │∆EF│ of untreated graphene sample used in DMF and CH3OH experiment is 0.205 eV and 0.19 eV, respectively. Both DMF and CH3OH functionalization cause a reduction of this value, but the magnitude of the decrease of DMF functionalization (from 0.205 to 0.114 eV) is larger than that of CH3OH functionalization (from 0.19 to 0.176 eV). This suggests that the doping effect originating from DMF is stronger than that from CH3OH. When graphene is functionalized with DMF/PBASE and CH3OH/PBASE, the │∆EF│ of graphene are 0.171 eV and 0.214 eV, respectively. These observations imply that both solvent molecule and PBASE impose doping effect on graphene.
To better understand the doping effects resulting from the above three chemicals, we also investigate the Fermi level shift (∆EF) of graphene after functionalization based on the electrical measurements. In the experiments using DMF as the solvent, the ∆EF of untreated graphene is 0.201 eV, and the ∆EF of DMF-functionalized graphene is -0.116 eV, which indicates that graphene is n-doped by DMF (Detailed calculation of ∆EF can be found in Figure S4). As expected, the doping effect of CH3OH on graphene is smaller than that of DMF because the ∆EF of CH3OH-functionalized graphene is -0.184 eV. When graphene is functionalized with CH3OH/PBASE, this value increases to -0.213 eV, which is attributed to the p-doping effect of PBASE. As shown in Table 1, these results are consistent with the calculations based on Raman spectroscopy. And the changes in ∆EF estimated based on electrical measurements prove that DMF and CH3OH impose n-doping effect on graphene while PBASE imposes p-doping effect. Thus, the doping effect of the surface functionalization on graphene can be monitored by analyzing the Fermi level shift of graphene using Raman spectroscopy and electrical measurements.
17
4. Conclusion
It is known that different aromatic backbones such as naphthaline, anthracene and pyrene could impose doping effects on graphene,39 however the doping effects of organic solvents have not been discussed before. Here, we investigated the doping effects on G-FETs during the functionalization process of aromatic molecule and two different organic solvents. Raman analysis confirms that both solvents and aromatic molecule impose doping effects on graphene through the shifts of G and 2D peaks and the changes in the I2D/IG ratio. It was found that solvents, which act as an electron donor, cause left shift in both G and 2D peaks, and increase the value of I2D/IG. However, the aromatic molecule PBASE acting as an electron acceptor causes the right shifts in G and 2D peaks, and the reduction of I2D/IG. The results show that both DMF and CH3OH impose n-doping effect on graphene, while the aromatic molecule PBASE affords pdoped graphene based on the electrical measurements and Fermi level shift analysis. Moreover, DMF causes a significant reduction of carrier mobility of G-FETs (from 1630.2 cm2/V·s to 5.0 cm2/V·s), whereas the value is much smaller with CH3OH as the solvent (from 1053.5 cm2/V·s to 758.7 cm2/V·s). Hence, CH3OH can be regarded as a better solvent for PBASE functionalization due to its smaller influence on carrier mobility of graphene devices. This work provides a simple approach to investigate the doping effects of the surface functionalization on graphene.
Acknowledgment This project is supported by the GRF grant from The Research Grant Council of the Hong Kong Special Administrative Region Government (CityU 11205514). Appendix A. Supplementary data
18
Supplementary data: G-FETs functionalization process; capacitive equivalent model; G band and 2D peak shift; carrier mobility and Fermi level shift calculations of solution gated G-FETs. References 1 E. Stolyarova, K. T. Rim, S. M. Ryu, J. Maultzsch, P. Kim, L. E. Brus, T. F. Heinz, M. S. Hybertsen, G. W. Flynn, High-resolution scanning tunneling microscopy imaging of mesoscopic graphene sheets on an insulating surface, PNAS 104 (2007) 9209-9212. 2 A. K. Geim and K. S. Novoselov, The rise of graphene, Nat. Mater. 6 (2007) 183-191. 3 X. Tang, H. K. Zhang, X. B. Tang, K. W. C. Lai, Photoresponse enhancement in graphene/silicon infrared detector by controlling photocarrier collection, Mater. Res. Express 3 (2016) 076203. 4 C. H Liu, Y. C. Chang, T. B. Norris, Z. H. Zhong, Graphene photodetectors with ultrabroadband and high responsivity at room temperature, Nat. Nanotechnol. 9 (2014) 273-278. 5 P.T. Yin, S. Shah, M. Chhowalla, K. -B. Lee, Design, synthesis, and characterization of graphene–nanoparticle hybrid materials for bioapplications, Chem. Rev. 115 (2015) 24832531. 6 D. S. Yu and L. M. Dai, Self-assembled graphene/carbon nanotube hybrid films for supercapacitors, J. Phys. Chem. Lett. 1 (2010) 467-470. 7 S. Mao, K. H. Yu, G. H. Lu, J. H. Chen, Highly sensitive protein sensor based on thermallyreduced graphene oxide field-effect transistor, Nano Res. 4 (2011) 921-930.
19
8 J. L. Ping, R. Vishnubhotla, A. Vrudhula, A. T. C. Johnson, Scalable production of highsensitivity, label-free DNA biosensors based on back-gated graphene field effect transistors, ACS Nano 10 (2016) 8700-8704. 9 V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, N. Kim, K. C. Kemp, P. Hobza, R. Zboril, K. S. Kim, Functionalization of graphene: covalent and non-covalent approaches, derivatives and applications, Chem. Rev. 112 (2012) 6156. 10 N. J. Jeon, J. Lee, J. H. Noh, M. K. Nazeeruddin , M. Grätzel, S. Seok, Efficient inorganicorganic hybrid perovskite solar cells based on pyrene arylamine derivatives as holetransporting materials, J. Am. Chem. Soc. 135 (2013) 19087-19090. 11 R. S. Pantelic, W. Y. Fu, C. Schoenenberger, H. Stahlberg, Rendering graphene supports hydrophilic with non-covalent aromatic functionalization for transmission electron microscopy, Appl. Phys. Lett. 104 (2014) 134103. 12 L. N. Cella, W. Chen, N. V. Myung, A. Mulchandani, Single-walled carbon nanotube-based chemiresistive affinity biosensors for small molecules: ultrasensitive glucose detection, J. Am. Chem. Soc. 132 (2010) 5024-5026. 13 Q. Y. He, S. X. Wu, Z. Y. Yin, H. Zhuang, Graphene-based electronic sensors, Chem. Sci. 3 (2012) 1764-1772. 14 Y. P. Liu, L. Yuan, M. Yang, Y. Zheng, L. J. Li, L. B. Gao, N. Nerngchamnong, C. T. Nai, C. S. S. Sangeeth, Y. P. Feng, C. A. Nijhuis, K. P. Loh, Giant enhancement in vertical conductivity of stacked CVD graphene sheets by self-assembled molecular layers, Nat. Commun. 5 (2014) 5461.
20
15 C. A. Hunter, J. K. M. Sanders, The nature of π-π interactions, J. Am. Chem. Soc. 112 (1990) 5525-5534. 16 Y. L. Zhao, J. F. Stoddart, Noncovalent functionalization of single-walled carbon nanotubes, Acc. Chem. Res. 42 (2009) 1161-1171. 17 Y. H. Kwak, D. S. Choi, Y. N. Kim, H. Kim, D. H. Yoon, S. S. Ahn, J. W. Yang, W. S. Yang, S. Seo, Flexible glucose sensor using CVD-grown graphene-based field effect transistor, Biosens. Bioelectron. 37 (2012) 82-87. 18 A. Das, S. Piasana, B. Chakraborty, S. Piscanec, S. K. Saha, U. V. Waghmare, K. S. Novoselov, H. R. Krishnamurthy, A. K. Geim, A. C. Ferrari, A. K. Sood, Monitoring dopants by Raman scattering in an electrochemically top-gated graphene transistor, Nat. Nanotechnol. 3 (2008) 210-215. 19 A. Srivastava, C. Galande, L. J. Ci, L. Song, C. Rai, D. Jariwala, K. F. Kelly, P. M. Ajayan, Novel liquid precursor-based facile synthesis of large-area continuous, single, and few-layer graphene films, Chem. Mater. 22 (2010) 3457-3461. 20 A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Rot, A. K. Geim, Raman spectrum of graphene and graphene layers, Phys. Rev. Lett. 97 (2006) 187401. 21 J. Yan, Y.B. Zhang, P. Kim, A. Pinczuk, Electric field effect tuning of electron-phonon coupling in graphene, Phys. Rev. Lett. 98 (2007) 166802. 22 D. A. Neamen, Semiconductor Physics and Devices, 2012, 4th edition, Mc Graw Hill India.
21
23 F. Chen, Q. Qing, J. L. Xia, J. H. Li, N. J. Tao, Electrochemical gate-controlled charge transport in graphene in ionic liquid and aqueous solution, J. Am. Chem. Soc. 131 (2009) 9908-9909. 24 J. L Xia, F. Chen, J. H. Li, N. J. Tao, Measurement of the quantum capacitance of graphene, Nat. Nanotechnol. 4 (2009) 505-509. 25 K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Electric field effect in atomically thin carbon films, Science, 306 (2004) 666-669. 26 G. Froehlicher, S. Berciaud, Raman spectroscopy of electrochemically gated graphene transistors: Geometrical capacitance, electron-phonon, electron-electron, and electron-defect scattering, Phys. Rev. B 91 (2015) 205413. 27 Y. Ohno, K. Maehashi, Y. Yamashiro, K. Matsumoto, Electrolyte-gated graphene field-effect transistors for detecting pH and protein adsorption, Nano Lett. 9 (2009) 3318-3322. 28 B. Gelmont, M.S. Shur, R. J. Mattauch, Disk and stripe capacitances, Solid State Electron. 38 (1995) 731-734. 29 M. Ornthai, A. Siripinyanond, B. K. Gale, Effect of ionic and nonionic carriers in electrical field-flow fractionation, Anal. Chem. 88 (2016) 1794-1803. 30 R. Stine, S. P. Mulvaney, J. T. Robinson, C. R. Tamanaba, P. E. Sheehan, Fabrication, optimization, and use of graphene field effect sensors, Anal. Chem. 85 (2013) 509-521.
22
31 I. Gierz, C. Riedl, U. Starke, C. R. Ast, K. Kern, Atomic Hole Doping of Graphene, Nano Lett. 8 (2008), 4603-4607. 32 H. G. Yan, F. N. Xia, W. J. Zhu, M. Freitag, C. Dimitrakopoulos, A. A. Bol, G. Tulevski, P. Avouris, Infrared spectroscopy of wafer-scale graphene, ACS Nano 5 (2011) 9854-9860. 33 Y. B. An, A. Behnam, E. Pop, A. Ural, Metal-semiconductor-metal photodetectors based on graphene/p-type silicon Schottky junctions, Appl. Phys. Lett. 102 (2013) 013110. 34 J. K. Lee, H. Sung, M. S. Jang, H. Yoon, M. Choi, Reliable doping and carrier concentration control in graphene by aerosol-derived metal nanoparticles, J. Mater. Chem. C 3 (2015) 82948299. 35 K Berke, S Tongay, M. A. McCarthy, A. G. Rinzler, B. R. Appleton, A. F. Hebard, Current transport across the pentacene/CVD-grown graphene interface for diode applications, J. Phys. Condens. Matter 24 (2012) 255802. 36 A. Reina, X. T. Jia, J. Ho, D. Nezich, H. B. Son, V. Bulovic, M. S. Dresselhaus, J. Kong, Large area, few-layer graphene films on arbitrary substrates by chemical vapor deposition, Nano Lett. 9 (2009) 30-35. 37 Y. F. Yuan, X. T. Yu, Q. Zhang, M. F. Chang, L. Li, T. Q. Yang, Y. T. Chen, H. F. Pan, S. J. Zhang, L. Li, J. H. Xu, Sensitive detection of polycyclic aromatic molecules: surface enhanced Raman scattering via π-π stacking, Anal. Chem. 88 (2016) 4328-4335. 38 C. A. Hunter, J. K. M. Sanders, The nature of π-π interactions, J. Am. Chem. Soc. 112 (1990) 5525-5534.
23
39 B. Das, R. Voggu, C. S. Rout, C. N. R. Rao, Changes in the electronic structure and properties of graphene induced by molecular charge-transfer, Chem. Commun. 41 (2008) 5155-5157. 40 J. H. Bong, O. Sul, A. Yoon, S. Y. Choi, B. J. Cho, Facile graphene n-doping by wet chemical treatment for electronic applications, Nanoscale 6 (2014) 8503-8508. 41 F. Chen, J. L. Xia, N. J. Tao, Ionic screening of charged-impurity scattering in graphene, Nano Lett. 9 (2009) 1621-1625. 42 M. Sayadian, M. Khaleghian, M. Yari, Solvatochromism and preferential solvation in mixtures of methanol with ethanol, 1-propanol and 1-butanol, Orient. J. Chem. 30 (2014) 1897-1903. 43 G. S. Kulkarni, K. Reddy, W. Z. Zang, K. H. Lee, X. D. Fan, Z. H. Zhong, Electrical probing and tuning of molecular physisorption on graphene, Nano Lett. 16 (2016) 695-700. 44 B. Q. Peng, L. Chen, C. J. Que, K. Yang, F. Deng, X. Y. Deng, G. S. Shi, G. Xu, M. H. Wu, Adsorption of antibiotics on graphene and biochar in aqueous solutions induced by π-π interactions, Sci. Rep. 6 (2016) 31920. 45 B. Gupta, N. Kumar, K. Panda, A. A. Melvin, S. Joshi, S. Dash, A. K.Tyagi, Effective noncovalent functionalization of poly(ethylene glycol) to reduced graphene oxide nanosheets through γ-radiolysis for enhanced lubrication, J. Phys. Chem. C 120 (2016) 2139-2148. 46 J. Lee, E. Hwang, E. Lee, S. Seo, H. Lee, Tuning of n- and p-type reduced graphene oxide transistors with the same molecular backbone, Chem. Eur. J. 18 (2012) 5155-5159.
24
Figure 1 Schematic illustration of the G-FET fabrication process. Inset: optical image of the G-FET. The scale bar is 500 μm.
Figure 2 Illustration of PBASE-functionalized G-FET in DMF or CH3OH. The black triangle represents adsorbed DMF or CH3OH molecule on the graphene surface.
25
Figure 3 Energy band diagrams of the (a) p-doped graphene, (b) pristine graphene and (c) ndoped graphene.
26
Figure 4 Optical image of graphene on SiO2/Si substrate. For each graphene area, width = 1 mm and length = 1 mm (scale bar, 800 μm). The red crosses represent Raman sampling spots.
27
Figure 5 Raman spectra of graphene functionalized in (a) DMF and (b) CH3OH. Green rectangle represents the disorder of D peak. Raman G peak spectra for graphene functionalized in (c) DMF and (d) CH3OH. Blue arrows represented the location of new peaks.
28
Figure 6 The changes in the I2D/IG ratio of graphene during PBASE functionalization. Blue and red columns represent I2D/IG ratio of graphene during PBASE functionalization by using DMF and CH3OH, respectively. Error bars indicate the standard deviations of the experimental Raman peak positions (12 measurements from graphene samples).
29
Figure 7 G-FET transfer curves (Vds = 0.1 V) when incubated in (a) DMF and (b) CH3OH. The time interval is 0, 30 min, 60 min, 120 min and 240 min. Inset: the changes in VDP of the G-FETs with different incubation times. ∆EF represents the Fermi level of graphene. The changes in carrier mobility (hole and electron) of the G-FETs incubated in (c) DMF and (d) CH3OH with different incubation times.
30
Figure 8 Transfer curves (Vds = 0.1 V) of G-FETs incubated with (a) 5 mM PBASE in DMF and (b) 5 mM in CH3OH. The time interval is 0, 30 min, 60 min, 120 min and 240 min. Inset: the changes in VDP of the G-FETs with different incubation times. ∆EF represents the Fermi level shift of graphene. The changes in carrier mobility (hole and electron) of the G-FETs incubated in (c) DMF/PBASE and (d) CH3OH/PBASE with different incubation times.
31
Figure 9 (a) Transfer curves (Vds = 0.1 V) of G-FETs incubated with 10 mM PBASE in DMF. The time interval is 0, 30 min, 60 min, 120 min and 240 min. Inset: the changes in VDP of the GFETs with different incubation times. ∆EF represents the Fermi level shift of graphene. (b) The changes in carrier mobility (hole and electron) of the G-FETs incubated in DMF with different incubation times.
32
Table 1 Fermi level shift of graphene based on Raman spectroscopy and electrical measurements
Item
Raman spectroscopy
Electrical measurements
DMF
ωG (cm-1)
VDP (V)
Graphene
1588.6
0.205
0.24
-0.201
Graphene/DMF
1584.8
0.114
0.08
-0.116
Graphene/DMF/PBASE
1587.2
0.171
0.18
-0.174
Graphene
1588
0.190
0.22
-0.193
Graphene/CH3OH
1587.4
0.176
0.2
-0.184
Graphene/CH3OH/PBASE
1589
0.214
0.27
-0.213
│∆EF│ (eV)
∆EF (eV)
CH3OH
33
S0169-4332(17)32037-8 http://dx.doi.org/doi:10.1016/j.apsusc.2017.07.048 APSUSC 36582
To appear in:
APSUSC
Received date: Revised date: Accepted date:
8-3-2017 26-6-2017 5-7-2017
Please cite this article as: Guangfu Wu, Xin Tang, M.Meyyappan, King Wai Chiu Lai, Doping Effects of Surface Functionalization on Graphene with Aromatic Molecule and Organic Solvents, Applied Surface Sciencehttp://dx.doi.org/10.1016/j.apsusc.2017.07.048 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Doping Effects of Surface Functionalization on Graphene with Aromatic Molecule and Organic Solvents Guangfu Wua, Xin Tanga, M. Meyyappanb, and King Wai Chiu Laia,* a. Department of Mechanical and Biomedical Engineering, City University of Hong Kong, 83 Tat Chee Ave., H.K. SAR, China b. NASA Ames Research Center, Moffett Field, CA, USA
* Corresponding author. Tel: +852-3442-9099; Fax: +852-3442-0172 E-mail address: [email protected]
1
Graphic Abstract:
We have investigated the doping effects of surface functionalization and its influence on the carrier mobility of graphene.
Highlights
For the first time, the doping effects of the surface functionalization on graphene with aromatic molecule and organic solvents were investigated by Raman and electronic properties. Raman spectra showed that both PBASE and these two solvents imposed doping effects on graphene. Electrical measurements further revealed that PBASE imposed a p-doping effect while DMF and CH3OH imposed n-doping effect. CH3OH causes a much smaller reduction in the carrier mobility of G-FETs than that of DMF.
Abstract
Aromatic molecule functionalization plays a key role in the development of graphene field-effect transistors (G-FETs) for bio-detection. We have investigated the doping effects of surface functionalization and its influence on the carrier mobility of graphene. The aromatic
2
molecule (1-pyrenebutanoic acid succinimidyl ester, PBASE), which is widely used as a linker to anchor bio-probes, was employed here to functionalize graphene. Dimethyl formamide (DMF) and methanol (CH3OH) were used as two solvents to dissolve PBASE. Raman spectra showed that both PBASE and these two solvents imposed doping effects on graphene. The PBASE was stably immobilized on the graphene surface, which was confirmed by the new peak at around 1623.5 cm-1 and the disordered D peak at 1350 cm-1. Electrical measurements and Fermi level shift analysis further revealed that PBASE imposes a p-doping effect while DMF and CH3OH impose an n-doping effect. More importantly, CH3OH causes a smaller reduction in the carrier mobility of G-FETs (from 1095.6 cm2/V·s to 802.4 cm2/V·s) than DMF (from 1640.4 cm2/V·s to 5.0 cm2/V·s). Therefore, CH3OH can be regarded as a better solvent for the PBASE functionalization. This careful study on the influence of organic solvents on graphene during PBASE functionalization process provides an effective approach to monitor the surface functionalization of graphene.
Keywords: Raman analysis; Surface functionalization; Graphene; Aromatic; Doping effect
1. Introduction
Graphene exhibits remarkable properties1-2 such as high carrier mobility, large specific surface area and excellent electrical conductivity, which make it a candidate material to develop photodetectors,3-4 capacitors5 and sensors.6 Recently, the field-effect transistor (FET) structure
3
employing graphene as the sensing material has been shown to have potential for sensitive biodetection.7-8 Generally, the graphene surface needs to be functionalized with bio-probes to realize specific detection. In general, the bio-probes are anchored on graphene by linker molecules. Typically covalent and non-covalent methods are employed to functionalize graphene with linker molecules.9 Covalent methods can stably and specifically introduce functional groups on the surface of graphene, but they will inevitably alter the physical properties and native electronic structure of graphene by converting the sp2 carbons to sp3 ones. Therefore, non-covalent approaches are preferable in order to preserve the inherent properties of pristine grapheme; in this case, aromatic molecules such as pyrene and its derivatives have a strong affinity for graphene because of the stable π-π stacking across the basal plane.10-11 1-pyrenebutanoic acid succinimidyl ester (PBASE) is an aromatic molecule which has both an aromatic ring and a reactive tail and has been widely used as a linker molecule in biosensors.12 For example, glucose oxidase is introduced to the surface of graphene and carbon nanotubes by using the linker molecule PBASE.12-13 In these examples, the role of PBASE is to serve as an anchor for introducing recognition elements such as an antibody onto the nanomaterial surface. The attachment of PBASE onto the surface of graphene is based on the stable π-π stacking between graphene and the pyrene group of PBASE, which has been confirmed in previous studies.14 Since graphene can be regarded as an infinite aromatic molecule and pyrene has four π-π binding sites, the interaction between pyrene and graphene adopts parallel-stacking orientation to maximize the PBASE-graphene interaction.15 It was also revealed that the distance of PBASE-graphene stacking was about 2.3 Å.14 This value is smaller than that of C6F6-C6H6 π-π stacking (3.6-3.8 Å), which is ascribed to that the electron withdrawing group in PBASE reduced the repulsive electrostatic interaction between pyrene and graphene, enhancing the π-π interaction.15
4
Aromatic molecules consisting of electron-withdrawing or electron-donating groups could dope carbon nanotubes,16 but similar phenomenon for graphene have been rarely studied. Besides, organic solvents such as dimethyl formamide (DMF) and methanol (CH3OH) are necessary to dissolve the aromatic molecule for graphene functionalization. The effects of these organic solvents during aromatic molecule functionalization process on the graphene properties have not yet been investigated. The graphene field-effect transistor (G-FET) is utilized here to explore the doping effect of aromatic molecule PBASE and organic solvents on graphene. The graphene layer serves as the conducting channel bridging the source and drain electrodes. A solution-gate electrode is usually employed in the G-FET to tune graphene’s Fermi level, and to obtain the transfer characteristics that reveal the doping condition (or Fermi level shift) and the carrier mobility of graphene.17 Besides, Raman spectroscopy is an effective method to investigate the inherent physical properties of carbon nanomaterials.18-19 The characteristic D peak (~1350 cm-1), G peak (~1580 cm-1) and 2D peak (~2685 cm-1) in Raman spectra are able to reveal the number of graphene layers, the defects in graphene and change in carrier concentration (electron or hole).20-21
In this paper, the doping effects of the aromatic molecule PBASE and two different organic solvents (DMF and CH3OH) on graphene are investigated carefully by monitoring the changes in Raman spectrum and the shift of Dirac point voltage of the G-FETs during the chemical functionalization process. The Fermi level shift of graphene is also estimated by these two methods. The results indicate that the PBASE imposes a p-doping effect on graphene because of its electron-withdrawing property while DMF and CH3OH impose an n-doping effect. Besides, CH3OH causes a smaller reduction in carrier mobility of G-FETs than DMF. This work
5
provides a two-in-one method to explore the effect of both the solvent and aromatic molecule on graphene.
2. Materials and methods
2.1 Fabrication and characterizations
2.1.1 G-FETs fabrication process
The fabrication of G-FETs is described in detail in Figure 1. CVD-grown graphene on copper foil was first spin-coated with a thin layer of PMMA. Then, the PMMA/graphene/Cu film was immersed in FeCl3·6H2O solution (1 M) until the Cu was completely etched away. The released PMMA/graphene film was washed by deionized water and transferred onto a SiO2/Si substrate. After removing the PMMA layer with acetone, patterned structures were made on the surface of graphene by photolithography, and the patterned graphene was transferred onto another Si/SiO2 substrate. The source and drain metal electrodes were created at the two ends of graphene film (1 × 1 mm2) by using silver conductive paint. Finally, silicone rubber was attached on top of the substrate to insulate the electrodes (source and drain) and to confine the liquid solution during experiment. The inset represents the optical image of the G-FET.
2.1.2 Characterizations
The as-fabricated G-FETs were functionalized with PBASE in two different organic solvents (DMF and CH3OH). The detailed functionalization approaches can be found in the supporting information. A schematic of the functionalized G-FET structure is shown in Figure 2. The graphene functionalization was characterized by Raman spectroscopy and electrical
6
measurements. All Raman measurements were conducted under room temperature (Renishaw Raman Spectroscopy 1000B) with 514 nm laser excitation. All electrical characterization was conducted using a semiconductor device analyzer coupled with a probe station (Semiconductor Analyzer, B1500, Keysight Co. Ltd., CA, USA) under room temperature. The aromatic molecule functionalized G-FETs were biased at 100 mV for all measurements, and the gate voltage was applied to the G-FETs via an Ag/AgCl electrode that was immersed in PBS buffer solution (0.1 mM, pH 7.2).
2.2. Carrier Mobility
The carrier mobility, which indicates the ability of the carrier (hole and electron) to drift in a material, is the essential feature of electronic devices.22 Capacitive equivalent model, which has been used to describe the capacitive property of the solution-gate G-FET,23 is employed here to estimate the carrier mobility of G-FETs integrated with a solution gate during the functionalization process. The details can be found under Supporting Information. It is assumed that the total gate capacitance (CTG) is a series of electrostatic double layer capacitance (CG_EDL) and quantum capacitance of graphene layer (CQ). This model is expressed by the following equation:17
CTG
CG EDL CQ CG EDL CQ
(1)
where the quantum capacitance of graphene layer CQ was estimated as ~2 µF·cm-2 based on previous work.24 CG_EDL is the capacitance of electrostatic double layer (EDL) that forms between graphene and the solution interface. For a typical back-gate G-FET, in which graphene
7
layer was put on dielectric layer (300 nm SiO2), the back gate capacitance (CG_BG) is simply described by the equation below:25
CG _ BG
SiO 0 2
(2)
d SiO2
where εSiO2 is the relative permittivity of SiO2; ε0 is the vacuum permittivity (ε0 = 8.854 × 10-12 F·m-1); and dSiO2 is the thickness of SiO2. For the solution-gate G-FET, the EDL, which works as a top-gate dielectric layer when a voltage is applied via the gate electrode, can be modeled as a parallel plate capacitor. This assumption has been used to estimate the electronic properties of graphene devices,26 and the thickness of the EDL can be calculated according to the DebyeHückel equation.27-28 In this work, the capacitance of EDL can be expressed as below:26
CG _ EDL
EDL 0 d EDL
(3)
where εEDL and dEDL are the relative permittivity (~78)29 of 0.1 mM PBS solution and the thickness of EDL (~7.7 nm for 0.1 mM PBS).30 The CTG, which is obtained from the capacitive equivalent model, can be used to calculate the carrier mobility of G-FET. The carrier mobility (hole or electron), in the linear region of transfer curve, can be calculated based on the following equation:17
I ds l 1 Vg w CTG Vds
(4)
8
where the length to width ration (l/w) of graphene channel was set to 1 in this work; Ids is source-drain current; Vg is gate voltage; and Vds is source-drain voltage.
2.3 Fermi level shift of graphene
The Fermi level for pristine graphene is located at the Dirac point; for doped (p-doped or n-doped) graphene, the Fermi level will shift away from the Dirac point.31 As shown in Figure 3, the Fermi level shift (∆EF) of the doped graphene is defined as the energy difference between the Fermi level (EF) and the Dirac point.
In this study, the Fermi level shift of graphene was estimated by two methods. One involved the G peak energy (ωG) of graphene in the Raman spectrum. For pristine graphene, the G peak is located at ~1580 cm-1, and both p-doped and n-doped graphene have right-shifted G peaks.21 Therefore, the absolute value of the Fermi level shift (│∆EF│) with respect to the Dirac point, is expressed as below:32
EF
G 1580cm1 42cm1 eV
(5)
The other one involves electrical measurement of the G-FET. The Fermi level shift of graphene in this case is expressed by the following equation:33
EF sgn(n) vF n
(6)
9
where νF is Fermi velocity of graphene; ħ is reduced Plank constant (1.0546 ×10-34 J·s); carrier density n is defined to be positive for holes, and negative for electrons. For solution gated (top gate) G-FETs, the carrier density of graphene is estimated by the following equation:34
n
CTG Vg e
(7)
where CTG is the total gate capacitance as shown in equation (1); e is elementary charge; and Vg is the gate voltage at Dirac point (VDP).
3. Results and discussion
3.1 Raman spectra
Raman spectroscopy was used to characterize five types of graphene: (1) untreated graphene; (2) DMF functionalized graphene; (3) CH3OH functionalized graphene; (4) DMF/PBASE functionalized graphene; (5) CH3OH/PBASE functionalized graphene. The changes in the characteristic bands were carefully examined during the functionalization process with different chemical treatments. In order to minimize the errors due to the variations in graphene samples, the data from 12 sampling points of the graphene sample before and after functionalization were averaged. The graphene sample size was 1 × 1 mm2 as shown in Figure 4.
DMF and CH3OH were chosen as candidate solvents for the functionalization experiments, and the Raman spectra of graphene before and after PBASE functionalization in these two solvents are shown in Figure 5. The black curve represents the Raman spectrum of untreated graphene, blue and red curves represent PBASE functionalized graphene in DMF and
10
CH3OH, respectively. The shift of G and 2D peaks and their intensity ratio (I2D/IG) are the indicators of doping effect on graphene.18
In Figure 5a and 5b, the positions of the G peak and 2D peak of untreated graphene are consistent with reported values.20 The small variations of these two peaks are attributed to the PMMA residues resulting from the fabrication process.35 These are few layer (one or two layers) graphene samples because the initial value of I2D/IG (the intensity ratio of 2D and G peak) is larger than 2.036 as shown in Figure 5. For the DMF and CH3OH-treated graphene, both G and 2D peaks exhibit an obvious left-shift as shown in Figure S3. The left shifts are attributed to the doping effect arising from the solvent molecule due to the existence of N in DMF and O atoms (electron donor) in CH3OH. The decrease in I2D/IG of the DMF and CH3OH-treated graphene further confirms that both DMF and CH3OH impose doping effect on graphene.
For PBASE functionalization in DMF (or in CH3OH), as shown in Figure 5c and 5d, a new peak located at 1623.5 cm-1 (1621.7 cm-1) appears after PBASE functionalization. This peak is attributed to pyrene group resonance,14 which directly proves that PBASE is attached onto the graphene surface. In a previous study, the number of the π-π stacking binding sites in polycyclic aromatic molecules was revealed to affect the Raman intensity significantly when they were adsorbed on the aromatic polysodium styrenesulfonate (PSSS) templates via π-π interactions.37 We believe that the published study will help us to explain the attachment of the aromatic molecule PBASE onto the graphene surface. Since graphene has many π-π binding sites, we think that the aromatic molecule PBASE can be immobilized on the graphene surface stably via π-π stacking. Meanwhile, slight right-shifts in both G and 2D peaks are observed compared with DMF and CH3OH-treated graphene, which are ascribed to the doping effect of PBASE on
11
graphene. The G peak (~1580 cm-1) is corresponding to a doubly degenerate phonon mode at the Γ point. The 2D peak (~2685 cm-1) results from simultaneous excitation of two phonons with wavevector close to the K point in the graphene Brillouin zone. Doping of graphene decreased the probability of excited charge carrier recombination.38 The decreased probability afforded non-adiabatic photon perturbations, which removed the Kohn anomaly and increased the energy of the phonons at the Γ point and K point. The increased phonon energy resulted in the Raman G and 2D peaks with higher frequency and larger wavelength. Therefore, the right-shifts (larger wavelength) in the G peak and 2D peak can be observed in Raman spectrum of PBASE treated graphene. This phenomenon has been discussed in the previous studies.18, 21 The I2D/IG ratio shows an increase after PBASE functionalization in DMF and CH3OH as shown in Figure 6. The right shift in both G and 2D peaks and the decrease in I2D/IG indicate that PBASE imposes an opposite doping effect on graphene which is attributed to the existence of the carbonyl group (electron acceptor) in PBASE. These observations of G and 2D peak shifts are in agreement with previous experimental results.39 Besides, the D peak for untreated graphene at 1350 cm-1 is not obvious, which indicates that few defects exist in graphene samples. The PBASE functionalization mildly increases the disorder in D peak, which is attributed to the orbital hybridization of the PBASE molecule with graphene plane.
3.2 Electronic properties of functionalized G-FETs
Although Raman spectrum can be used to confirm the doping effect on graphene, it is difficult to distinguish whether it is p- or n-doping effect. Moreover, it is difficult to obtain the detailed information on electronic properties such as carrier mobility by Raman spectrum. Therefore, we conducted transfer curve measurements on four groups of functionalized G-FETs:
12
(1) DMF functionalized G-FETs; (2) CH3OH functionalized G-FETs; (3) DMF/PBASE functionalized G-FET; (4) CH3OH/PBASE functionalized G-FET. The functionalization conditions and the experimental measurement conditions were kept the same for the timedependent measurement. The temperature was kept at 25 °C and the humidity was kept at 50%.
First, the transfer curves on DMF (and CH3OH) functionalized G-FETs are discussed. The transfer characteristics of DMF functionalized G-FET after different incubation times are shown in Figure 7. There are two noteworthy features: (1) the left shift of the Dirac point voltage (VDP); (2) the reduction of conductance of DMF functionalized G-FETs. As shown in the inset of Figure 7a, the VDP shifts from 0.24 V to 0.08 V. This provides evidence that DMF imposes ndoping effect on G-FETs.40 The conductance decreases dramatically during the DMF functionalization process, and we extracted the carrier mobility (hole and electron) from the transfer curves using equations (1)-(3). The results are summarized in Figure 7c. The DMF causes a significant reduction in hole mobility from 1640.4 cm2/V·s to 5.0 cm2/V·s (Detailed calculation of hole mobility can be found in Figure S4 and Table S1). The electron mobility shows a similar change with the increase in incubation time. The reduction of carrier mobility could be attributed to the charged impurity scattering caused by adsorption of the DMF molecules.41
Compared with DMF-functionalized G-FET, CH3OH causes a smaller left-shift of VDP from 0.22 V to 0.20 V as shown in Figure 7b. Moreover, in Figure 7d, the reduction of carrier mobility is from 1095.6 cm2/V·s to 802.4 cm2/V·s, which is much smaller than that caused by DMF. The reason could be that DMF has a larger molecular dipole moment (3.82 Debye) and polarizability (7.8 × 10-24 cm3)42 than CH3OH (1.7 Debye; 0.6 × 10-24 cm3).43 For DMF, the
13
larger dipole moment and polarizability contribute to a larger noncovalent DMF-graphene binding energy,43 making DMF on graphene surface more difficult to be removed during the functionalization process. Therefore, the DMF molecule is adsorbed more stably on graphene surface than CH3OH, causing a heavier scattering effect on the graphene device.14 Thus, the reduction in carrier mobility of DMF-functionalized G-FET is much larger than that of the CH3OH case.
We now discuss the doping effect of PBASE functionalization on G-FETs in the above two solvents with the transfer characteristics shown in Figure 8a and 8b. Unlike the continuous left-shift of VDP for DMF-functionalized G-FETs, PBASE (5 mM in DMF) functionalization causes a right-shift of the VDP from 0.22 V to 0.24 V in the first 30 min, and stays stable from 30 min to 60 min. From 60 to 120 min, the VDP shows a left shift (from 0.24 V to 0.18 V). Results discussed earlier showed that DMF imposed n-doping effect on graphene. Therefore, the rightshift in the first 30 min can be attributed to the p-doping effect of PBASE. The p-doping effect of PBASE is also evidenced from the transfer curves of the CH3OH/PBASE-functionalized G-FET.
Figure 8b shows continuous right-shift (from 0.21 V to 0.27 V) of the VDP of CH3OH/PBASE-functionalized G-FETs, which is quite different from that of DMF/PBASEfunctionalized G-FETs. Experimental results discussed earlier confirmed that both DMF and CH3OH imposed n-doping effect on graphene. Therefore, the n-doping effect from these two solvents and the p-doping effect from PBASE are competitive. The n-doping effect of CH3OH is slightly based on the previously measured VDP shift (from 0.22 to 0.20 V). Therefore, the pdoping effect of PBASE dominates the CH3OH/PBASE functionalization process. Besides, when
14
the incubation time increases, the stronger n-doping effect of DMF neutralizes the p-doping effect of PBASE, and dominates the DMF/PBASE functionalization process.
The carrier mobility (hole and electron) of the DMF/PBASE and CH3OH/PBASE functionalized G-FETs is summarized in Figure 8c and 8d. For CH3OH/PBASE functionalized G-FET, the hole mobility decreases from 1377.5 cm2/V·s to 1144.3 cm2/V·s, which is much smaller than that of DMF/PBASE functionalized cases (977.0 cm2/V·s to 76.3 cm2/V·s). These results are consistent with that of DMF and CH3OH functionalized G-FETs, which further suggests that CH3OH has a smaller influence on the carrier mobility of G-FET during PBASE functionalization process.
In order to investigate the influence of PBASE concentration, 10 mM PBASE was used to functionalize graphene devices in DMF. The transfer characteristics and carrier mobility of PBASE (10 mM in DMF) functionalized graphene devices after different incubation times are shown in Figure 9. This functionalization causes an obvious change of VDP, which shows a larger right shift (from 0.22 V to 0.26 V) in the first 30 min. The observed right-shift of VDP further confirms that PBASE imposes p-doping effect on G-FETs which is also observed for PBASE (5 mM in DMF) functionalized G-FETs. From 60 to 120 min, the VDP changes from 0.26 to 0.21 V which is smaller than that of 5 mM PBASE. The observed right-shift of VDP further confirms that PBASE imposes p-doping effect on G-FETs which is also observed for PBASE (5 mM in DMF) functionalized G-FETs. From 60 to 120 min, the VDP changes from 0.26 to 0.21 V which is smaller than that of 5 mM PBASE. Furthermore, the carrier mobility is still reducing after 240 min for 10 mM PBASE, which is different from that of 5 mM PBASE. The performances of Dirac point voltage shift and carrier mobility change indicate that the number of the aromatic
15
molecule PBASE has an influence on graphene during the functionalization process. The volume of PBASE solution was kept same for all functionalization experiments. Larger concentration afforded larger amount of PBASE molecule that took a longer time to make the PBASEgraphene interaction reach to the saturation. In Figure 7 and Figure 8, the carrier mobility became stable after 120 min for 5 mM PBASE because most of the PBASE molecules were attached onto the surface of graphene. Similar phenomenon has been reported in a previous study.44 The adsorption of antibiotics on graphene reached balance or complete absorption due to the low concentration of antibiotics. In Figure 9, 10 mM PBASE was used to functionalize graphene, the larger number of PBASE molecule caused the interaction to reach the saturation in a longer time. Therefore, the carrier mobility is still reducing after 240 min for 10 mM PBASE. This phenomenon is similar with that in other two studies.37, 45 Larger number of FAD caused the shift in the absorbance of PSSS-templated AgNPs in longer time, and larger number of rGO molecule costed a longer time to make the absorbance of PEG200 decrease to the stable state. The attachment of the PBASE to graphene by π-π stacking appears to have no effect on the conductance of the devices.46 Therefore, the reduction of conductance in the experiments could be mainly attributed to the adsorption of DMF.
3.3 Fermi level shift analysis
The Fermi level shift (∆EF) of graphene is regarded as an indicator for chemical doping.31 To further understand the doping effects caused by different dopants, we then discuss the ∆EF of graphene after DMF, CH3OH and PBASE functionalization based on Raman spectroscopy and electrical measurements. In Raman spectrum, functionalization with the above three chemicals causes the changes in G peak, which has been used to estimate the │∆EF│ of graphene.32 Based
16
on equation (5), the │∆EF│ of graphene after DMF, CH3OH and PBASE functionalization is summarized in Table 1. The │∆EF│ of untreated graphene sample used in DMF and CH3OH experiment is 0.205 eV and 0.19 eV, respectively. Both DMF and CH3OH functionalization cause a reduction of this value, but the magnitude of the decrease of DMF functionalization (from 0.205 to 0.114 eV) is larger than that of CH3OH functionalization (from 0.19 to 0.176 eV). This suggests that the doping effect originating from DMF is stronger than that from CH3OH. When graphene is functionalized with DMF/PBASE and CH3OH/PBASE, the │∆EF│ of graphene are 0.171 eV and 0.214 eV, respectively. These observations imply that both solvent molecule and PBASE impose doping effect on graphene.
To better understand the doping effects resulting from the above three chemicals, we also investigate the Fermi level shift (∆EF) of graphene after functionalization based on the electrical measurements. In the experiments using DMF as the solvent, the ∆EF of untreated graphene is 0.201 eV, and the ∆EF of DMF-functionalized graphene is -0.116 eV, which indicates that graphene is n-doped by DMF (Detailed calculation of ∆EF can be found in Figure S4). As expected, the doping effect of CH3OH on graphene is smaller than that of DMF because the ∆EF of CH3OH-functionalized graphene is -0.184 eV. When graphene is functionalized with CH3OH/PBASE, this value increases to -0.213 eV, which is attributed to the p-doping effect of PBASE. As shown in Table 1, these results are consistent with the calculations based on Raman spectroscopy. And the changes in ∆EF estimated based on electrical measurements prove that DMF and CH3OH impose n-doping effect on graphene while PBASE imposes p-doping effect. Thus, the doping effect of the surface functionalization on graphene can be monitored by analyzing the Fermi level shift of graphene using Raman spectroscopy and electrical measurements.
17
4. Conclusion
It is known that different aromatic backbones such as naphthaline, anthracene and pyrene could impose doping effects on graphene,39 however the doping effects of organic solvents have not been discussed before. Here, we investigated the doping effects on G-FETs during the functionalization process of aromatic molecule and two different organic solvents. Raman analysis confirms that both solvents and aromatic molecule impose doping effects on graphene through the shifts of G and 2D peaks and the changes in the I2D/IG ratio. It was found that solvents, which act as an electron donor, cause left shift in both G and 2D peaks, and increase the value of I2D/IG. However, the aromatic molecule PBASE acting as an electron acceptor causes the right shifts in G and 2D peaks, and the reduction of I2D/IG. The results show that both DMF and CH3OH impose n-doping effect on graphene, while the aromatic molecule PBASE affords pdoped graphene based on the electrical measurements and Fermi level shift analysis. Moreover, DMF causes a significant reduction of carrier mobility of G-FETs (from 1630.2 cm2/V·s to 5.0 cm2/V·s), whereas the value is much smaller with CH3OH as the solvent (from 1053.5 cm2/V·s to 758.7 cm2/V·s). Hence, CH3OH can be regarded as a better solvent for PBASE functionalization due to its smaller influence on carrier mobility of graphene devices. This work provides a simple approach to investigate the doping effects of the surface functionalization on graphene.
Acknowledgment This project is supported by the GRF grant from The Research Grant Council of the Hong Kong Special Administrative Region Government (CityU 11205514). Appendix A. Supplementary data
18
Supplementary data: G-FETs functionalization process; capacitive equivalent model; G band and 2D peak shift; carrier mobility and Fermi level shift calculations of solution gated G-FETs. References 1 E. Stolyarova, K. T. Rim, S. M. Ryu, J. Maultzsch, P. Kim, L. E. Brus, T. F. Heinz, M. S. Hybertsen, G. W. Flynn, High-resolution scanning tunneling microscopy imaging of mesoscopic graphene sheets on an insulating surface, PNAS 104 (2007) 9209-9212. 2 A. K. Geim and K. S. Novoselov, The rise of graphene, Nat. Mater. 6 (2007) 183-191. 3 X. Tang, H. K. Zhang, X. B. Tang, K. W. C. Lai, Photoresponse enhancement in graphene/silicon infrared detector by controlling photocarrier collection, Mater. Res. Express 3 (2016) 076203. 4 C. H Liu, Y. C. Chang, T. B. Norris, Z. H. Zhong, Graphene photodetectors with ultrabroadband and high responsivity at room temperature, Nat. Nanotechnol. 9 (2014) 273-278. 5 P.T. Yin, S. Shah, M. Chhowalla, K. -B. Lee, Design, synthesis, and characterization of graphene–nanoparticle hybrid materials for bioapplications, Chem. Rev. 115 (2015) 24832531. 6 D. S. Yu and L. M. Dai, Self-assembled graphene/carbon nanotube hybrid films for supercapacitors, J. Phys. Chem. Lett. 1 (2010) 467-470. 7 S. Mao, K. H. Yu, G. H. Lu, J. H. Chen, Highly sensitive protein sensor based on thermallyreduced graphene oxide field-effect transistor, Nano Res. 4 (2011) 921-930.
19
8 J. L. Ping, R. Vishnubhotla, A. Vrudhula, A. T. C. Johnson, Scalable production of highsensitivity, label-free DNA biosensors based on back-gated graphene field effect transistors, ACS Nano 10 (2016) 8700-8704. 9 V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, N. Kim, K. C. Kemp, P. Hobza, R. Zboril, K. S. Kim, Functionalization of graphene: covalent and non-covalent approaches, derivatives and applications, Chem. Rev. 112 (2012) 6156. 10 N. J. Jeon, J. Lee, J. H. Noh, M. K. Nazeeruddin , M. Grätzel, S. Seok, Efficient inorganicorganic hybrid perovskite solar cells based on pyrene arylamine derivatives as holetransporting materials, J. Am. Chem. Soc. 135 (2013) 19087-19090. 11 R. S. Pantelic, W. Y. Fu, C. Schoenenberger, H. Stahlberg, Rendering graphene supports hydrophilic with non-covalent aromatic functionalization for transmission electron microscopy, Appl. Phys. Lett. 104 (2014) 134103. 12 L. N. Cella, W. Chen, N. V. Myung, A. Mulchandani, Single-walled carbon nanotube-based chemiresistive affinity biosensors for small molecules: ultrasensitive glucose detection, J. Am. Chem. Soc. 132 (2010) 5024-5026. 13 Q. Y. He, S. X. Wu, Z. Y. Yin, H. Zhuang, Graphene-based electronic sensors, Chem. Sci. 3 (2012) 1764-1772. 14 Y. P. Liu, L. Yuan, M. Yang, Y. Zheng, L. J. Li, L. B. Gao, N. Nerngchamnong, C. T. Nai, C. S. S. Sangeeth, Y. P. Feng, C. A. Nijhuis, K. P. Loh, Giant enhancement in vertical conductivity of stacked CVD graphene sheets by self-assembled molecular layers, Nat. Commun. 5 (2014) 5461.
20
15 C. A. Hunter, J. K. M. Sanders, The nature of π-π interactions, J. Am. Chem. Soc. 112 (1990) 5525-5534. 16 Y. L. Zhao, J. F. Stoddart, Noncovalent functionalization of single-walled carbon nanotubes, Acc. Chem. Res. 42 (2009) 1161-1171. 17 Y. H. Kwak, D. S. Choi, Y. N. Kim, H. Kim, D. H. Yoon, S. S. Ahn, J. W. Yang, W. S. Yang, S. Seo, Flexible glucose sensor using CVD-grown graphene-based field effect transistor, Biosens. Bioelectron. 37 (2012) 82-87. 18 A. Das, S. Piasana, B. Chakraborty, S. Piscanec, S. K. Saha, U. V. Waghmare, K. S. Novoselov, H. R. Krishnamurthy, A. K. Geim, A. C. Ferrari, A. K. Sood, Monitoring dopants by Raman scattering in an electrochemically top-gated graphene transistor, Nat. Nanotechnol. 3 (2008) 210-215. 19 A. Srivastava, C. Galande, L. J. Ci, L. Song, C. Rai, D. Jariwala, K. F. Kelly, P. M. Ajayan, Novel liquid precursor-based facile synthesis of large-area continuous, single, and few-layer graphene films, Chem. Mater. 22 (2010) 3457-3461. 20 A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Rot, A. K. Geim, Raman spectrum of graphene and graphene layers, Phys. Rev. Lett. 97 (2006) 187401. 21 J. Yan, Y.B. Zhang, P. Kim, A. Pinczuk, Electric field effect tuning of electron-phonon coupling in graphene, Phys. Rev. Lett. 98 (2007) 166802. 22 D. A. Neamen, Semiconductor Physics and Devices, 2012, 4th edition, Mc Graw Hill India.
21
23 F. Chen, Q. Qing, J. L. Xia, J. H. Li, N. J. Tao, Electrochemical gate-controlled charge transport in graphene in ionic liquid and aqueous solution, J. Am. Chem. Soc. 131 (2009) 9908-9909. 24 J. L Xia, F. Chen, J. H. Li, N. J. Tao, Measurement of the quantum capacitance of graphene, Nat. Nanotechnol. 4 (2009) 505-509. 25 K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Electric field effect in atomically thin carbon films, Science, 306 (2004) 666-669. 26 G. Froehlicher, S. Berciaud, Raman spectroscopy of electrochemically gated graphene transistors: Geometrical capacitance, electron-phonon, electron-electron, and electron-defect scattering, Phys. Rev. B 91 (2015) 205413. 27 Y. Ohno, K. Maehashi, Y. Yamashiro, K. Matsumoto, Electrolyte-gated graphene field-effect transistors for detecting pH and protein adsorption, Nano Lett. 9 (2009) 3318-3322. 28 B. Gelmont, M.S. Shur, R. J. Mattauch, Disk and stripe capacitances, Solid State Electron. 38 (1995) 731-734. 29 M. Ornthai, A. Siripinyanond, B. K. Gale, Effect of ionic and nonionic carriers in electrical field-flow fractionation, Anal. Chem. 88 (2016) 1794-1803. 30 R. Stine, S. P. Mulvaney, J. T. Robinson, C. R. Tamanaba, P. E. Sheehan, Fabrication, optimization, and use of graphene field effect sensors, Anal. Chem. 85 (2013) 509-521.
22
31 I. Gierz, C. Riedl, U. Starke, C. R. Ast, K. Kern, Atomic Hole Doping of Graphene, Nano Lett. 8 (2008), 4603-4607. 32 H. G. Yan, F. N. Xia, W. J. Zhu, M. Freitag, C. Dimitrakopoulos, A. A. Bol, G. Tulevski, P. Avouris, Infrared spectroscopy of wafer-scale graphene, ACS Nano 5 (2011) 9854-9860. 33 Y. B. An, A. Behnam, E. Pop, A. Ural, Metal-semiconductor-metal photodetectors based on graphene/p-type silicon Schottky junctions, Appl. Phys. Lett. 102 (2013) 013110. 34 J. K. Lee, H. Sung, M. S. Jang, H. Yoon, M. Choi, Reliable doping and carrier concentration control in graphene by aerosol-derived metal nanoparticles, J. Mater. Chem. C 3 (2015) 82948299. 35 K Berke, S Tongay, M. A. McCarthy, A. G. Rinzler, B. R. Appleton, A. F. Hebard, Current transport across the pentacene/CVD-grown graphene interface for diode applications, J. Phys. Condens. Matter 24 (2012) 255802. 36 A. Reina, X. T. Jia, J. Ho, D. Nezich, H. B. Son, V. Bulovic, M. S. Dresselhaus, J. Kong, Large area, few-layer graphene films on arbitrary substrates by chemical vapor deposition, Nano Lett. 9 (2009) 30-35. 37 Y. F. Yuan, X. T. Yu, Q. Zhang, M. F. Chang, L. Li, T. Q. Yang, Y. T. Chen, H. F. Pan, S. J. Zhang, L. Li, J. H. Xu, Sensitive detection of polycyclic aromatic molecules: surface enhanced Raman scattering via π-π stacking, Anal. Chem. 88 (2016) 4328-4335. 38 C. A. Hunter, J. K. M. Sanders, The nature of π-π interactions, J. Am. Chem. Soc. 112 (1990) 5525-5534.
23
39 B. Das, R. Voggu, C. S. Rout, C. N. R. Rao, Changes in the electronic structure and properties of graphene induced by molecular charge-transfer, Chem. Commun. 41 (2008) 5155-5157. 40 J. H. Bong, O. Sul, A. Yoon, S. Y. Choi, B. J. Cho, Facile graphene n-doping by wet chemical treatment for electronic applications, Nanoscale 6 (2014) 8503-8508. 41 F. Chen, J. L. Xia, N. J. Tao, Ionic screening of charged-impurity scattering in graphene, Nano Lett. 9 (2009) 1621-1625. 42 M. Sayadian, M. Khaleghian, M. Yari, Solvatochromism and preferential solvation in mixtures of methanol with ethanol, 1-propanol and 1-butanol, Orient. J. Chem. 30 (2014) 1897-1903. 43 G. S. Kulkarni, K. Reddy, W. Z. Zang, K. H. Lee, X. D. Fan, Z. H. Zhong, Electrical probing and tuning of molecular physisorption on graphene, Nano Lett. 16 (2016) 695-700. 44 B. Q. Peng, L. Chen, C. J. Que, K. Yang, F. Deng, X. Y. Deng, G. S. Shi, G. Xu, M. H. Wu, Adsorption of antibiotics on graphene and biochar in aqueous solutions induced by π-π interactions, Sci. Rep. 6 (2016) 31920. 45 B. Gupta, N. Kumar, K. Panda, A. A. Melvin, S. Joshi, S. Dash, A. K.Tyagi, Effective noncovalent functionalization of poly(ethylene glycol) to reduced graphene oxide nanosheets through γ-radiolysis for enhanced lubrication, J. Phys. Chem. C 120 (2016) 2139-2148. 46 J. Lee, E. Hwang, E. Lee, S. Seo, H. Lee, Tuning of n- and p-type reduced graphene oxide transistors with the same molecular backbone, Chem. Eur. J. 18 (2012) 5155-5159.
24
Figure 1 Schematic illustration of the G-FET fabrication process. Inset: optical image of the G-FET. The scale bar is 500 μm.
Figure 2 Illustration of PBASE-functionalized G-FET in DMF or CH3OH. The black triangle represents adsorbed DMF or CH3OH molecule on the graphene surface.
25
Figure 3 Energy band diagrams of the (a) p-doped graphene, (b) pristine graphene and (c) ndoped graphene.
26
Figure 4 Optical image of graphene on SiO2/Si substrate. For each graphene area, width = 1 mm and length = 1 mm (scale bar, 800 μm). The red crosses represent Raman sampling spots.
27
Figure 5 Raman spectra of graphene functionalized in (a) DMF and (b) CH3OH. Green rectangle represents the disorder of D peak. Raman G peak spectra for graphene functionalized in (c) DMF and (d) CH3OH. Blue arrows represented the location of new peaks.
28
Figure 6 The changes in the I2D/IG ratio of graphene during PBASE functionalization. Blue and red columns represent I2D/IG ratio of graphene during PBASE functionalization by using DMF and CH3OH, respectively. Error bars indicate the standard deviations of the experimental Raman peak positions (12 measurements from graphene samples).
29
Figure 7 G-FET transfer curves (Vds = 0.1 V) when incubated in (a) DMF and (b) CH3OH. The time interval is 0, 30 min, 60 min, 120 min and 240 min. Inset: the changes in VDP of the G-FETs with different incubation times. ∆EF represents the Fermi level of graphene. The changes in carrier mobility (hole and electron) of the G-FETs incubated in (c) DMF and (d) CH3OH with different incubation times.
30
Figure 8 Transfer curves (Vds = 0.1 V) of G-FETs incubated with (a) 5 mM PBASE in DMF and (b) 5 mM in CH3OH. The time interval is 0, 30 min, 60 min, 120 min and 240 min. Inset: the changes in VDP of the G-FETs with different incubation times. ∆EF represents the Fermi level shift of graphene. The changes in carrier mobility (hole and electron) of the G-FETs incubated in (c) DMF/PBASE and (d) CH3OH/PBASE with different incubation times.
31
Figure 9 (a) Transfer curves (Vds = 0.1 V) of G-FETs incubated with 10 mM PBASE in DMF. The time interval is 0, 30 min, 60 min, 120 min and 240 min. Inset: the changes in VDP of the GFETs with different incubation times. ∆EF represents the Fermi level shift of graphene. (b) The changes in carrier mobility (hole and electron) of the G-FETs incubated in DMF with different incubation times.
32
Table 1 Fermi level shift of graphene based on Raman spectroscopy and electrical measurements
Item
Raman spectroscopy
Electrical measurements
DMF
ωG (cm-1)
VDP (V)
Graphene
1588.6
0.205
0.24
-0.201
Graphene/DMF
1584.8
0.114
0.08
-0.116
Graphene/DMF/PBASE
1587.2
0.171
0.18
-0.174
Graphene
1588
0.190
0.22
-0.193
Graphene/CH3OH
1587.4
0.176
0.2
-0.184
Graphene/CH3OH/PBASE
1589
0.214
0.27
-0.213
│∆EF│ (eV)
∆EF (eV)
CH3OH
33