Chemical Reduction of Individual Graphene Oxide Sheets as ...
S1
Chemical Reduction of Individual Graphene Oxide Sheets as Revealed by Electrostatic Force Microscopy Dhaval D. Kulkarni†, Songkil Kim‡, Marius Chyasnavichyus†, Kesong Hu†, Andrei G. Fedorov‡, and Vladimir V. Tsukruk† †
School of Materials Science and Engineering and ‡George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332.
Supporting Information Experimental Chemicals and materials: Graphene oxide was synthesized from natural graphite flakes (325 mesh, 99.8% metal basis) purchased from Alfa Aesar as per the modified Hummer’s method.[ 1 ]
Stable dispersion of graphene oxide in a solution mixture of
methanol: water (5:1) was subjected to ultrasonication for 15 mins followed by centrifugation at 3000 rpm. The supernatant was decanted and used for LangmuirBlodgett (LB) deposition. P-doped silicon wafer with a 300 nm thermal oxide layer were cut and cleaned in piranha solution (3:1 mixture of H2SO4/H2O2) for 1 hr and then thoroughly rinsed with water.
Nanopure water (18.2 MΩ cm) was used in all the
processes. Reduction of graphene oxide flakes was performed by exposing the entire substrate to hydrazine vapors for different time periods.[2],[10] Assembly and reduction of graphene oxide: Graphene oxide was deposited on a silicon substrate with a 300 nm oxide layer on the top using LB setup at room temperature using a KSV 2000 LB minitrough.[3] Specifically, 2.0 ml of graphene oxide solution (0.1g/l) in methanol:water mixture (5:1) was uniformly distributed over the water surface. The monolayer was deposited at a surface pressure of ~0.5 mN/m in order to
S2
obtain a uniform coverage with minimum overlap. The reduction process was carried out by exposing the substrate to hydrazine vapors every 30 sec, while performing EFM measurement (~30 min per scan) between the consecutive exposures.
Instrumentation:
Coverage of the graphene oxide flakes on SiO2 substrate was
analyzed using a Hitachi-3400 SEM microscope with operating voltage of 10-15keV. Contact angle measurements were performed using a KSV CAM 101 contact angle system by dropping 10 µl Nanopure water at three different locations on the sample. Raman measurements were performed using a Witec (Alpha 300R) confocal Raman microscope with an Ar+ ion laser (λ = 514 nm) operated at 1 mW. Surface composition was analyzed using a Thermo K-Alpha XPS system with an Al Kα source. Survey spectra were obtained over the range of 0-1000eV at 1 eV steps with a spot size of 30 μm averaged over 2 scans. High resolution scans were performed in the range of relevance for specific elements at 0.1 eV steps and averaged over five scans. Topography and EFM-phase images were obtained using a Bruker Icon AFM equipped with a home-built humidity controller using n-doped silicon probes with tip radius of 10 nm, spring constant of 3 N/m, quality factor of 240, and resonant frequency of 60-80 kHz.[4] Image analysis was performed using the Bruker Nanoscope Analysis software, ImageJ, and Matlab.
Surface characterization
Contact angle: Fig.SI1 shows the optical, AFM, and SEM image of the graphene oxide flakes assembled on a 300 nm SiO 2/p-Si substrate. According to Cassie’s law5, the effective contact angle
for a liquid droplet on a composite surface (graphene oxide
plus silicon dioxide in our case) is given by: ( )
(
)
(
)
(1)
S3
where oxide,
is the coverage of graphene oxide, is the coverage of SiO2, and
is the contact angle of bulk graphene is the contact angle of bulk graphene
oxide. Contact angle for water on a freshly cleaned piece of SiO 2 surface was found to be around 5o and the substrate with 80+6% coverage of graphene oxide showed a contact angle of 52+4o.
Fig. S1. Global structural characterization of graphene oxide flakes. (a) Representative optical image, (b) AFM topography, and (c) Scanning electron microscopy (SEM) image of the graphene oxide flakes assembled in SiO 2 surface resulting in an areal density of 80 + 6% and (d) representative contact angle profile of the surface.
Considering the possibility of contaminants from the atmosphere being adsorbed on the surface during storage, we measured the contact angle on the substrate at different locations. The contact angle depends on the surface chemistry and can be used to estimate the concentration of the functional groups on the surface of graphene oxide.[6],[7] It was found that the average contact angle for the graphene oxide deposited
S4
on the SiO2 surface was ~52 + 4o (Fig. S1b). This value is lower than the contact angle of bulk graphene oxide reported in literature, 62.8 o and can be related to limited surface coverage of graphene oxide.[ 8 ]
The 20% difference between the contact angle
measurements reported in this study and the literature values supports the fact that 20% of the surface is not covered by the graphene oxide as seen from the SEM images (Fig. S1c). Moreover, the values of the water contact angle on bulk graphene oxide reported in literature varies by about 2o based on the C:O content which depends on the method used for synthesis (Staudinger’s/ Hummer’s/ modified Hummer’s). Also, the surface roughness and porosity of bulk materials might influence the contact angle.
Thus, based on these parameters, the effective contact angle of graphene oxide (
) was calculated to be 58.5+2o and was close to the values reported in the literature.
This agreement negates the possibility of significant adsorbed contaminants which increase the contact angle.
Next, based on the contact angle measurements, the
concentration of oxygenated groups on the graphene oxide can also be estimated by assuming that the oxygenated domains show low contact angle of about 15 o and the graphitic domains behave as a graphite surface with a contact angle of 98o,[ 9 ] the oxygenated domains would constitute about 63% of the total surface area which is close to typical estimations.[10]
Compositional characterization:
The chemical composition of the graphene oxide
was confirmed using spectroscopic techniques. Fig. S2 shows the Raman map of the surface obtained by recording the intensity of the G-band (1550-1650 cm-1) and the Raman spectra of graphene oxide flakes. The spectrum showed all the characteristic peaks corresponding to the silicon substrate (1 st order at 520 cm-1 and 2nd order between 950-1000 cm-1) and the graphene oxide (D-band at 1359 cm-1, G-band at 1613 cm-1, and a broad 2D band between 2500-3000 cm-1).[ 11 ],[ 12 ]
A uniform intensity
distribution of the G-band and the D-band was observed over the surface of flakes. This further supports the presence of a majority of monolayer flakes. XPS data for heavily reduced graphene oxide shows the presence of all C=C, C-N, and C=O bonds after reduction.
S5
(a)
(b) C=C 284.6
Counts, (a.u.)
C–N 286.4
C=O 288.5
280
285
290
295
Binding Energy, (eV) -1
Fig S2. Raman mapping of 1608 cm peak intensity of graphene oxide (a); (b) XPS data for heavily reduced graphene oxide (600 secs),
Table SI1: Raman spectra parameters
Flake
GO rGO
D peak Position, Width, -1 cm cm-1 1349 121 1353 115
Surface potential:
G peak Position, Width, -1 cm cm-1 1608 65 1609 60
D/G Peak Ratio 1.06 0.96
To quantify the surface potential distribution, we conducted
electrostatic force spectroscopy (EFS) by measuring the phase shift at a given point in different areas as a function of continuous variation of tip bias. As expected, the phase shift was negative over all the surface areas probed and showed a parabolic dependence on the tip bias. Judging from these dependencies, it can be concluded that the bright and dark areas in EFM-phase images showed a very different response in comparison with fully reduced graphene oxide surface. Comparing the bright and dark areas within the graphene oxide itself, the spectroscopic curves showed small but quantified difference well beyond standard deviations (Table SI2).
S6
Fig. S3. (a) EFM phase shift plotted as a function of tip bias; scattered points represent experimental data (only 1 data point from 10 was selected for this plot), R2 =0.997; lines represent data fit with the second order polynomial. EFM phase images of the (b) graphene oxide flake, and (c) reduced graphene oxide obtained with a tip bias of +5V at a lift height of 50 nm, showing the regions where phase shift curves in (a) was collected.
The phase shift vs. tip bias data obtained over the different surface areas can be analyzed with the equation in the form [ ( where
is the constant parameter
represents the surface potential. Also, since the minimum phase shift occurs at further used to calculate
(
)) ]
(2)
represents an offset to 0o, and
, (
(
)
) can be calculated from the plot itself (
)
.
From these fits,
can be
gradient from the known quality factor and spring constant
of the EFM probe. Large tip-bias reveals clear differences in overall shape and peak positions of surface areas on and off the flakes
S7
Table SI2: Parabolic fitting parameters with standard deviations.
Bright Areas Dark Areas RGO SiO2
A -0.00510±0.00007 -0.00499±0.00007 -0.00743±0.00010 -0.00503±0.00007
B -0.1370±0.0028 -0.3050±0.0029 -0.1396±0.0027 -0.1490±0.0027
This way, different surface areas were analyzed to obtain quantitative estimates of the electrostatic interactions and it was observed that the 3), showed a little variation with
values being constant (eq.
of 1.29 + 0.044 C/cm2 and 1.22 + 0.042 C/cm2 in the
bright areas and dark areas respectively.
However, the surface potential (
(
))
showed a difference of -34 + 11 mV for the brightest area and -327 + 30 mV for the darkest surface areas as calculated from EFM data fit. A lower surface potential in the darker areas further confirms that these areas correspond to the oxygenated functionalities and vice-versa. On the other hand, the bright areas showed a slightly higher phase shift at a negative tip bias and the darker areas showed a higher phase shift at a positive tip bias. This behavior again confirms that the darker areas in the EFM-phase image obtained at a positive tip bias should correspond to the oxygenated functionalities with high electronegativity and the bright areas are occupied by positively charged graphene-like domains. Considering that around 60% of total surface area of the graphene oxide flakes is composed of the negatively charged oxygen functionalities, a net negative charge should be observed as in fact confirmed by the z-potential measurements. 13,14
The surface potential then corresponds to the potential across the capacitor system, and the surface charge density on the plates can be calculated using: [
(
)
(3)
S8
where,
is the surface charge density,
between the plates (water,
),
is the relative permittivity of the medium (
) is the surface potential, and
separation between the plates (thickness of water layer,
is the
).
Fig. S4. C:O ratio compositional distribution of the graphene oxide surface, with flake topography overlay. From this relationship, the charge densities in the brightest and the darkest areas in the EFM phase image of graphene oxide surface were calculated to be 0.06 C/m2 and 1.0 C/m2, respectively. Approximating that the oxygenated groups contains one oxygen atoms and the benzene ring in graphitic lattice contains effectively 2 carbon atoms (each carbon is shared by 3 neighboring rings), and using the charge values estimated above, the C:O ratio within the heavily oxygenated domains (darkest EFM phase areas) was estimated to be 2.4:1. The values of the C:O were found to be close to the values obtained using high-resolution XPS. On the other hand, the least oxygenated domains (brightest EFM phase areas) should contain only one oxygen atom per 40 carbon atoms (or 5% of benzene rings with defects).
By expanding this analysis, the C:O ratio
variation of individual graphene oxide flakes can be plotted (Fig.S4). This mapping represents the local variation of chemical composition from fully oxidized to virtually completely graphitic within randomized surface regions of 50-100 nm across obtained with truly nanoscale resolution.
S9
Local structural and compositional characterization
Effect of tip polarity on EFM-imaging of graphene oxide
Fig. S5.
AFM topography and EFM-phase image of (a) graphene oxide and (b)
reduced graphene oxide flake at different tip biases.
The different chemical functionalities (hydroxyl, epoxy, and carboxyl) having an oxygen with a lone pair of electrons should have a lower surface potential ( ( graphitic domains.15,16 Thus, at a positive tip bias,
(
)) than the
) will be higher in the
negatively charged oxidized areas and will result in a greater phase lag (darker areas) in the corresponding EFM-phase image. Morever, the situation will be reversed on reversal of the tip polarity since absolute value.,17
(
) will be negative and smaller than
in
S10
This phenomenon is clearly evident in the EFM-phase image of the graphene oxide flake obtained at
and 50nm lift height. The oxidized domains which cover
most of the graphene oxide surface correspond to the dark areas which constitute over 60+15% of the EFM-phase image. Interestingly, presence of bound surface charges is responsible for the reversal in the EFM-phase contrast upon reversal of tip polarity. However, the electrostatic interaction for materials with no surface bound charges (For eg., reduced graphene oxide) results from the tip-sample capacitance which primarily depends on the material properties (dielectric constant of the medium between the tip and surface, geometry, and tip-sample separation). The EFM-phase image of these materials is expected to show no reversal of contrast upon reversal of tip polarity.
To verify this effect, EFM-phase images of graphene oxide flakes were obtained at two opposite biases before and after reduction under low humidity conditions (Fig. S5). As expected, no apparent difference was observed in the flake topography. However, the EFM phase images of the graphene oxide flakes clearly showed a contrast reversal dominant over a major area of the surface upon changing the polarity of the tip bias (highlighted by squares). On the other hand, the EFM-phase images of the reduced graphene oxide flakes show the same contrast irrespective of the tip bias (Fig. S5b). These consistent changes suggest that the observed EFM-phase contrast over the different areas on the graphene oxide flake is indeed due to the areas consisting of charge species resulting in different surface potentials.
The absence of contrast reversal with a relative change in the EFM phase shift observed over the reduced graphene oxide surface implies that the electrostatic signal over the surface is related to the increase in permittivity. Reduction of graphene oxide is known to increase the size of the graphitic domains thereby improving the electrical conductivity. The oxygenated functionalities on the graphene oxide surface disappear after reduction leading to an increase in the sp 2 character and transitioning from an insulator to an electrically conductive material. The mobile charges in the electrically conductive material are free (unlike the bound charges on graphene oxide) and can orient themselves in response to the external electrical field induced by the charged
S11
AFM tip. Thus, reversal of the tip polarity shows no effect other than a relative decrease in the phase shift over the reduced graphene oxide surface.
Changes in the EFM-phase image of graphene oxide at different stages of reduction monitored by changing the tip polartiy
Fig. S6.
Effect of chemical reduction time (exposure to hydrazine vapor) on the
graphene oxide composition for the same individual flake edge (fixed phase window and position).
The reduction process of graphene oxide flakes using hydrazine vapor and monitored the surface charge distribution of oxidized areas. EFM images were obtained for the same flake upon subjecting the substrate to reducing environment for different periods of time.
An example showing the changes in the EFM-phase of graphene oxide
subjected to reduction has been included in the main text. The shifting of the phase scale to increase the image contrast inadvertently resulted in a change in the SiO 2
S12
contrast as seen in the Fig. 2. In reality, the EFM-phase of the SiO2 surface does not change upon reduction of graphene oxide as is evident from the Fig. 3b presented in the main text. Here, we convince the readers that the EFM-phase contrast change on the SiO2 surface seen in the Fig. 2 is indeed due the phase scale used to enhance the image contrast. Fig S6 shows the topography of the same flake and the corresponding EFM-phase images taken at different reduction times. Clearly, the SiO2 phase contrast does not change upon reduction of graphene oxide.
Thickness of graphene oxide 0.68 nm 1.47 nm
1.2
1.0 2
2 0.5
std = (0.19 +0.18 ) = 0.27 (nm)
nm
-1
0.8
Flake height: 0.78±0.27 nm
0.6
0.4
0.2
500 nm
0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
Height, nm 2.2
From ten profiles: H = 0.84±0.09 nm
2.0 1.8
Height, nm
1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0
500
1000
1500
2000
2500
Position, nm
Fig. S7. (a) Topography image of graphene oxide flake with and two methods of flake height calculation from topography data: (b) bearing analysis for height histogram (GO sheet and silicon) and (c) line by line step function fitting. Results presented as mean ± standard deviation.
S13 Bearing Peak1 Peak2 PeakSum
4.0 3.5 3.0 2.5
, °
-1
2.0 1.5 1.0 0.5 0.0 -0.5 0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Phase, °
Fig. S8. Bearing analysis of phase shift histogram for graphene oxide sheet and silicon substrate after reducing for 30 sec.
References: [1]
W. S. Hummers, R. E. Offeman, J. Amer. Chem. Soc. 1958, 80, 1339. S. Stankovich, D. A. Dikin, R. D. Piner, K .A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S. T. Nguyen, R. S. Ruoff, Carbon 2007, 45, 1558. [3] D. D. Kulkarni, I. Choi, S. Singamaneni, ACS Nano 2010, 4, 4667. [4] M. E. McConney, S. Singamaneni, V. V. Tsukruk, Poly. Rev. 2010, 50, 235. [5] Cassie, A. B. D.; Baxter, S. Wettability of porous surfaces. Trans. Faraday Soc. 1944, 40, 546. [6] R. N. Wenzel, J. Phys. Chem. 1949, 53, 1466. [7] L. Gao, T. J. McCarthy, Langmuir 2006, 22, 6234. [8] I. K. Moon, J. Lee, R. S. Ruoff, H. Lee, Nat. Comm. 2010, 1, 1. [9] S. Wang, Y. Zhang, Y.N. Abidi, L. Cabrales, Langmuir 2008, 25, 11078. [10] D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev. 2010, 39, 228. [ 11 ] A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Roth, A. K. Geim, Phys. Rev. Lett. 2006, 97, 187401. [ 12 ] D. Yang, A. Velamakanni, G. Bozoklu, S. Park, M. Stoller, R. D. Piner, S. Stankovich, I. Jung, D. A. Field, C. A. Ventrice Jr., R. S. Ruoff, Carbon 2009, 47, 145. [ 13 ] D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. B. Dommett, G. Evmenenko,; S. T. Nguyen, R. S. Ruoff, Nature 2007, 448, 457. [14] B. Konkena, S. Vasudevan, J. Phys. Chem. Lett. 2012, 3, 867. [15] Sugimura, H.; Ishida, Y.; Hayashi, K.; Takai, O. Appl. Phys. Lett. 2002, 80, 14591461 [16] Vezenov, D. V.; Noy, A.; Rozsnyai, L. F.; Lieber, C. M. J. Am. Chem. Soc. 1997, 119, 2006-2015. [17] Dianoux, R.; Martins, F.; Marchi, F.; Alandi, C.; Comin, F.; Chevrier. J. Phys. Rev. B 2003, 68, 045403. [2]
Chemical Reduction of Individual Graphene Oxide Sheets as Revealed by Electrostatic Force Microscopy Dhaval D. Kulkarni†, Songkil Kim‡, Marius Chyasnavichyus†, Kesong Hu†, Andrei G. Fedorov‡, and Vladimir V. Tsukruk† †
School of Materials Science and Engineering and ‡George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332.
Supporting Information Experimental Chemicals and materials: Graphene oxide was synthesized from natural graphite flakes (325 mesh, 99.8% metal basis) purchased from Alfa Aesar as per the modified Hummer’s method.[ 1 ]
Stable dispersion of graphene oxide in a solution mixture of
methanol: water (5:1) was subjected to ultrasonication for 15 mins followed by centrifugation at 3000 rpm. The supernatant was decanted and used for LangmuirBlodgett (LB) deposition. P-doped silicon wafer with a 300 nm thermal oxide layer were cut and cleaned in piranha solution (3:1 mixture of H2SO4/H2O2) for 1 hr and then thoroughly rinsed with water.
Nanopure water (18.2 MΩ cm) was used in all the
processes. Reduction of graphene oxide flakes was performed by exposing the entire substrate to hydrazine vapors for different time periods.[2],[10] Assembly and reduction of graphene oxide: Graphene oxide was deposited on a silicon substrate with a 300 nm oxide layer on the top using LB setup at room temperature using a KSV 2000 LB minitrough.[3] Specifically, 2.0 ml of graphene oxide solution (0.1g/l) in methanol:water mixture (5:1) was uniformly distributed over the water surface. The monolayer was deposited at a surface pressure of ~0.5 mN/m in order to
S2
obtain a uniform coverage with minimum overlap. The reduction process was carried out by exposing the substrate to hydrazine vapors every 30 sec, while performing EFM measurement (~30 min per scan) between the consecutive exposures.
Instrumentation:
Coverage of the graphene oxide flakes on SiO2 substrate was
analyzed using a Hitachi-3400 SEM microscope with operating voltage of 10-15keV. Contact angle measurements were performed using a KSV CAM 101 contact angle system by dropping 10 µl Nanopure water at three different locations on the sample. Raman measurements were performed using a Witec (Alpha 300R) confocal Raman microscope with an Ar+ ion laser (λ = 514 nm) operated at 1 mW. Surface composition was analyzed using a Thermo K-Alpha XPS system with an Al Kα source. Survey spectra were obtained over the range of 0-1000eV at 1 eV steps with a spot size of 30 μm averaged over 2 scans. High resolution scans were performed in the range of relevance for specific elements at 0.1 eV steps and averaged over five scans. Topography and EFM-phase images were obtained using a Bruker Icon AFM equipped with a home-built humidity controller using n-doped silicon probes with tip radius of 10 nm, spring constant of 3 N/m, quality factor of 240, and resonant frequency of 60-80 kHz.[4] Image analysis was performed using the Bruker Nanoscope Analysis software, ImageJ, and Matlab.
Surface characterization
Contact angle: Fig.SI1 shows the optical, AFM, and SEM image of the graphene oxide flakes assembled on a 300 nm SiO 2/p-Si substrate. According to Cassie’s law5, the effective contact angle
for a liquid droplet on a composite surface (graphene oxide
plus silicon dioxide in our case) is given by: ( )
(
)
(
)
(1)
S3
where oxide,
is the coverage of graphene oxide, is the coverage of SiO2, and
is the contact angle of bulk graphene is the contact angle of bulk graphene
oxide. Contact angle for water on a freshly cleaned piece of SiO 2 surface was found to be around 5o and the substrate with 80+6% coverage of graphene oxide showed a contact angle of 52+4o.
Fig. S1. Global structural characterization of graphene oxide flakes. (a) Representative optical image, (b) AFM topography, and (c) Scanning electron microscopy (SEM) image of the graphene oxide flakes assembled in SiO 2 surface resulting in an areal density of 80 + 6% and (d) representative contact angle profile of the surface.
Considering the possibility of contaminants from the atmosphere being adsorbed on the surface during storage, we measured the contact angle on the substrate at different locations. The contact angle depends on the surface chemistry and can be used to estimate the concentration of the functional groups on the surface of graphene oxide.[6],[7] It was found that the average contact angle for the graphene oxide deposited
S4
on the SiO2 surface was ~52 + 4o (Fig. S1b). This value is lower than the contact angle of bulk graphene oxide reported in literature, 62.8 o and can be related to limited surface coverage of graphene oxide.[ 8 ]
The 20% difference between the contact angle
measurements reported in this study and the literature values supports the fact that 20% of the surface is not covered by the graphene oxide as seen from the SEM images (Fig. S1c). Moreover, the values of the water contact angle on bulk graphene oxide reported in literature varies by about 2o based on the C:O content which depends on the method used for synthesis (Staudinger’s/ Hummer’s/ modified Hummer’s). Also, the surface roughness and porosity of bulk materials might influence the contact angle.
Thus, based on these parameters, the effective contact angle of graphene oxide (
) was calculated to be 58.5+2o and was close to the values reported in the literature.
This agreement negates the possibility of significant adsorbed contaminants which increase the contact angle.
Next, based on the contact angle measurements, the
concentration of oxygenated groups on the graphene oxide can also be estimated by assuming that the oxygenated domains show low contact angle of about 15 o and the graphitic domains behave as a graphite surface with a contact angle of 98o,[ 9 ] the oxygenated domains would constitute about 63% of the total surface area which is close to typical estimations.[10]
Compositional characterization:
The chemical composition of the graphene oxide
was confirmed using spectroscopic techniques. Fig. S2 shows the Raman map of the surface obtained by recording the intensity of the G-band (1550-1650 cm-1) and the Raman spectra of graphene oxide flakes. The spectrum showed all the characteristic peaks corresponding to the silicon substrate (1 st order at 520 cm-1 and 2nd order between 950-1000 cm-1) and the graphene oxide (D-band at 1359 cm-1, G-band at 1613 cm-1, and a broad 2D band between 2500-3000 cm-1).[ 11 ],[ 12 ]
A uniform intensity
distribution of the G-band and the D-band was observed over the surface of flakes. This further supports the presence of a majority of monolayer flakes. XPS data for heavily reduced graphene oxide shows the presence of all C=C, C-N, and C=O bonds after reduction.
S5
(a)
(b) C=C 284.6
Counts, (a.u.)
C–N 286.4
C=O 288.5
280
285
290
295
Binding Energy, (eV) -1
Fig S2. Raman mapping of 1608 cm peak intensity of graphene oxide (a); (b) XPS data for heavily reduced graphene oxide (600 secs),
Table SI1: Raman spectra parameters
Flake
GO rGO
D peak Position, Width, -1 cm cm-1 1349 121 1353 115
Surface potential:
G peak Position, Width, -1 cm cm-1 1608 65 1609 60
D/G Peak Ratio 1.06 0.96
To quantify the surface potential distribution, we conducted
electrostatic force spectroscopy (EFS) by measuring the phase shift at a given point in different areas as a function of continuous variation of tip bias. As expected, the phase shift was negative over all the surface areas probed and showed a parabolic dependence on the tip bias. Judging from these dependencies, it can be concluded that the bright and dark areas in EFM-phase images showed a very different response in comparison with fully reduced graphene oxide surface. Comparing the bright and dark areas within the graphene oxide itself, the spectroscopic curves showed small but quantified difference well beyond standard deviations (Table SI2).
S6
Fig. S3. (a) EFM phase shift plotted as a function of tip bias; scattered points represent experimental data (only 1 data point from 10 was selected for this plot), R2 =0.997; lines represent data fit with the second order polynomial. EFM phase images of the (b) graphene oxide flake, and (c) reduced graphene oxide obtained with a tip bias of +5V at a lift height of 50 nm, showing the regions where phase shift curves in (a) was collected.
The phase shift vs. tip bias data obtained over the different surface areas can be analyzed with the equation in the form [ ( where
is the constant parameter
represents the surface potential. Also, since the minimum phase shift occurs at further used to calculate
(
)) ]
(2)
represents an offset to 0o, and
, (
(
)
) can be calculated from the plot itself (
)
.
From these fits,
can be
gradient from the known quality factor and spring constant
of the EFM probe. Large tip-bias reveals clear differences in overall shape and peak positions of surface areas on and off the flakes
S7
Table SI2: Parabolic fitting parameters with standard deviations.
Bright Areas Dark Areas RGO SiO2
A -0.00510±0.00007 -0.00499±0.00007 -0.00743±0.00010 -0.00503±0.00007
B -0.1370±0.0028 -0.3050±0.0029 -0.1396±0.0027 -0.1490±0.0027
This way, different surface areas were analyzed to obtain quantitative estimates of the electrostatic interactions and it was observed that the 3), showed a little variation with
values being constant (eq.
of 1.29 + 0.044 C/cm2 and 1.22 + 0.042 C/cm2 in the
bright areas and dark areas respectively.
However, the surface potential (
(
))
showed a difference of -34 + 11 mV for the brightest area and -327 + 30 mV for the darkest surface areas as calculated from EFM data fit. A lower surface potential in the darker areas further confirms that these areas correspond to the oxygenated functionalities and vice-versa. On the other hand, the bright areas showed a slightly higher phase shift at a negative tip bias and the darker areas showed a higher phase shift at a positive tip bias. This behavior again confirms that the darker areas in the EFM-phase image obtained at a positive tip bias should correspond to the oxygenated functionalities with high electronegativity and the bright areas are occupied by positively charged graphene-like domains. Considering that around 60% of total surface area of the graphene oxide flakes is composed of the negatively charged oxygen functionalities, a net negative charge should be observed as in fact confirmed by the z-potential measurements. 13,14
The surface potential then corresponds to the potential across the capacitor system, and the surface charge density on the plates can be calculated using: [
(
)
(3)
S8
where,
is the surface charge density,
between the plates (water,
),
is the relative permittivity of the medium (
) is the surface potential, and
separation between the plates (thickness of water layer,
is the
).
Fig. S4. C:O ratio compositional distribution of the graphene oxide surface, with flake topography overlay. From this relationship, the charge densities in the brightest and the darkest areas in the EFM phase image of graphene oxide surface were calculated to be 0.06 C/m2 and 1.0 C/m2, respectively. Approximating that the oxygenated groups contains one oxygen atoms and the benzene ring in graphitic lattice contains effectively 2 carbon atoms (each carbon is shared by 3 neighboring rings), and using the charge values estimated above, the C:O ratio within the heavily oxygenated domains (darkest EFM phase areas) was estimated to be 2.4:1. The values of the C:O were found to be close to the values obtained using high-resolution XPS. On the other hand, the least oxygenated domains (brightest EFM phase areas) should contain only one oxygen atom per 40 carbon atoms (or 5% of benzene rings with defects).
By expanding this analysis, the C:O ratio
variation of individual graphene oxide flakes can be plotted (Fig.S4). This mapping represents the local variation of chemical composition from fully oxidized to virtually completely graphitic within randomized surface regions of 50-100 nm across obtained with truly nanoscale resolution.
S9
Local structural and compositional characterization
Effect of tip polarity on EFM-imaging of graphene oxide
Fig. S5.
AFM topography and EFM-phase image of (a) graphene oxide and (b)
reduced graphene oxide flake at different tip biases.
The different chemical functionalities (hydroxyl, epoxy, and carboxyl) having an oxygen with a lone pair of electrons should have a lower surface potential ( ( graphitic domains.15,16 Thus, at a positive tip bias,
(
)) than the
) will be higher in the
negatively charged oxidized areas and will result in a greater phase lag (darker areas) in the corresponding EFM-phase image. Morever, the situation will be reversed on reversal of the tip polarity since absolute value.,17
(
) will be negative and smaller than
in
S10
This phenomenon is clearly evident in the EFM-phase image of the graphene oxide flake obtained at
and 50nm lift height. The oxidized domains which cover
most of the graphene oxide surface correspond to the dark areas which constitute over 60+15% of the EFM-phase image. Interestingly, presence of bound surface charges is responsible for the reversal in the EFM-phase contrast upon reversal of tip polarity. However, the electrostatic interaction for materials with no surface bound charges (For eg., reduced graphene oxide) results from the tip-sample capacitance which primarily depends on the material properties (dielectric constant of the medium between the tip and surface, geometry, and tip-sample separation). The EFM-phase image of these materials is expected to show no reversal of contrast upon reversal of tip polarity.
To verify this effect, EFM-phase images of graphene oxide flakes were obtained at two opposite biases before and after reduction under low humidity conditions (Fig. S5). As expected, no apparent difference was observed in the flake topography. However, the EFM phase images of the graphene oxide flakes clearly showed a contrast reversal dominant over a major area of the surface upon changing the polarity of the tip bias (highlighted by squares). On the other hand, the EFM-phase images of the reduced graphene oxide flakes show the same contrast irrespective of the tip bias (Fig. S5b). These consistent changes suggest that the observed EFM-phase contrast over the different areas on the graphene oxide flake is indeed due to the areas consisting of charge species resulting in different surface potentials.
The absence of contrast reversal with a relative change in the EFM phase shift observed over the reduced graphene oxide surface implies that the electrostatic signal over the surface is related to the increase in permittivity. Reduction of graphene oxide is known to increase the size of the graphitic domains thereby improving the electrical conductivity. The oxygenated functionalities on the graphene oxide surface disappear after reduction leading to an increase in the sp 2 character and transitioning from an insulator to an electrically conductive material. The mobile charges in the electrically conductive material are free (unlike the bound charges on graphene oxide) and can orient themselves in response to the external electrical field induced by the charged
S11
AFM tip. Thus, reversal of the tip polarity shows no effect other than a relative decrease in the phase shift over the reduced graphene oxide surface.
Changes in the EFM-phase image of graphene oxide at different stages of reduction monitored by changing the tip polartiy
Fig. S6.
Effect of chemical reduction time (exposure to hydrazine vapor) on the
graphene oxide composition for the same individual flake edge (fixed phase window and position).
The reduction process of graphene oxide flakes using hydrazine vapor and monitored the surface charge distribution of oxidized areas. EFM images were obtained for the same flake upon subjecting the substrate to reducing environment for different periods of time.
An example showing the changes in the EFM-phase of graphene oxide
subjected to reduction has been included in the main text. The shifting of the phase scale to increase the image contrast inadvertently resulted in a change in the SiO 2
S12
contrast as seen in the Fig. 2. In reality, the EFM-phase of the SiO2 surface does not change upon reduction of graphene oxide as is evident from the Fig. 3b presented in the main text. Here, we convince the readers that the EFM-phase contrast change on the SiO2 surface seen in the Fig. 2 is indeed due the phase scale used to enhance the image contrast. Fig S6 shows the topography of the same flake and the corresponding EFM-phase images taken at different reduction times. Clearly, the SiO2 phase contrast does not change upon reduction of graphene oxide.
Thickness of graphene oxide 0.68 nm 1.47 nm
1.2
1.0 2
2 0.5
std = (0.19 +0.18 ) = 0.27 (nm)
nm
-1
0.8
Flake height: 0.78±0.27 nm
0.6
0.4
0.2
500 nm
0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
Height, nm 2.2
From ten profiles: H = 0.84±0.09 nm
2.0 1.8
Height, nm
1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0
500
1000
1500
2000
2500
Position, nm
Fig. S7. (a) Topography image of graphene oxide flake with and two methods of flake height calculation from topography data: (b) bearing analysis for height histogram (GO sheet and silicon) and (c) line by line step function fitting. Results presented as mean ± standard deviation.
S13 Bearing Peak1 Peak2 PeakSum
4.0 3.5 3.0 2.5
, °
-1
2.0 1.5 1.0 0.5 0.0 -0.5 0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
Phase, °
Fig. S8. Bearing analysis of phase shift histogram for graphene oxide sheet and silicon substrate after reducing for 30 sec.
References: [1]
W. S. Hummers, R. E. Offeman, J. Amer. Chem. Soc. 1958, 80, 1339. S. Stankovich, D. A. Dikin, R. D. Piner, K .A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S. T. Nguyen, R. S. Ruoff, Carbon 2007, 45, 1558. [3] D. D. Kulkarni, I. Choi, S. Singamaneni, ACS Nano 2010, 4, 4667. [4] M. E. McConney, S. Singamaneni, V. V. Tsukruk, Poly. Rev. 2010, 50, 235. [5] Cassie, A. B. D.; Baxter, S. Wettability of porous surfaces. Trans. Faraday Soc. 1944, 40, 546. [6] R. N. Wenzel, J. Phys. Chem. 1949, 53, 1466. [7] L. Gao, T. J. McCarthy, Langmuir 2006, 22, 6234. [8] I. K. Moon, J. Lee, R. S. Ruoff, H. Lee, Nat. Comm. 2010, 1, 1. [9] S. Wang, Y. Zhang, Y.N. Abidi, L. Cabrales, Langmuir 2008, 25, 11078. [10] D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev. 2010, 39, 228. [ 11 ] A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Roth, A. K. Geim, Phys. Rev. Lett. 2006, 97, 187401. [ 12 ] D. Yang, A. Velamakanni, G. Bozoklu, S. Park, M. Stoller, R. D. Piner, S. Stankovich, I. Jung, D. A. Field, C. A. Ventrice Jr., R. S. Ruoff, Carbon 2009, 47, 145. [ 13 ] D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. B. Dommett, G. Evmenenko,; S. T. Nguyen, R. S. Ruoff, Nature 2007, 448, 457. [14] B. Konkena, S. Vasudevan, J. Phys. Chem. Lett. 2012, 3, 867. [15] Sugimura, H.; Ishida, Y.; Hayashi, K.; Takai, O. Appl. Phys. Lett. 2002, 80, 14591461 [16] Vezenov, D. V.; Noy, A.; Rozsnyai, L. F.; Lieber, C. M. J. Am. Chem. Soc. 1997, 119, 2006-2015. [17] Dianoux, R.; Martins, F.; Marchi, F.; Alandi, C.; Comin, F.; Chevrier. J. Phys. Rev. B 2003, 68, 045403. [2]
