Characterize the intrinsic water wettability of graphite with contact angle. measurement: effect of defects on the static and dynamic contact angles

Transcription

1 Characterize the intrinsic water wettability of graphite with contact angle measurement: effect of defects on the static and dynamic contact angles Andrew Kozbial a, Charlie Trouba a, Haitao Liu b, and Lei Li a,c,* a Department of Chemical & Petroleum Engineering, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, PA, USA b Department of Chemistry, University of Pittsburgh, Pittsburgh, PA 15260, USA c Department of Mechanical Engineering & Materials Science, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, PA, USA Supplementary Information *Corresponding author lel55@pitt.edu S1

2 SI 1: Wenzel analysis Influence of surface roughness was analyzed for graphite samples based on AFM topography measurement according to the Wenzel equation S1 and the detailed procedure was described elsewhere. S2 Young s WCA (θ Y ) was calculated to be 65.6 and 50.9 for ZYA and PG, respectively, as shown in Table S1. Roughness has no effect on wettability for any of the HOPG samples and decreases the WCA by 3.5 for PG. The surface would need to be extremely rough for roughness affect wettability according to the Wenzel equation. Advancing WCA for ZYA and PG is 66.3 and 74.9, respectively. Applying these values to the Wenzel equation, θ Y for ZYA is 66.3 and θ Y for PG is This indicates that surface roughness has no influence for ZYA and decreases WCA by 1.4% for PG. Considering experimental uncertainty for water contact angle measurements is ±1, the effect of surface roughness is negligible. Therefore, Wenzel analysis concludes that surface roughness has a negligible effect on wettability of fresh graphite. Table S1. Wenzel analysis of graphite samples. Data presented as average (± standard deviation). R a is the surface roughness from AFM scans. ZYA SPI-1 SPI-2 PG # of AFM scans R a (nm) (± 0.103) (± 0.401) (± 0.249) 41.4 (± 8.0) Projected Surface Area (µm) 25.0 (± 0.0) 25.0 (± 0.0) 25.0 (± 0.0) 26.8 (± 0.8) r (± 0.000) (± 0.000) (± 0.000) (± 0.032) Experimental Static WCA (θ s ) Young s WCA (θ Y ) 65.6 (± 0.0 ) 65.3 (± 0.0 ) 65.2 (± 0.0 ) 50.9 (± 1.3 ) Experimental Advancing WCA (θ a ) Young s WCA (θ Y ) 66.3 (± 0.0 ) (± 0.4 ) S2

3 SI 2: Raman spectroscopy Raman spectra Point defects on a graphitic surface include vacancies, substitutional impurity atoms, selfinterstitial atoms, and interstitial impurity atoms. S3, S4 Raman spectroscopy provides the capability to quantitatively determine the density of point defects on graphitic surfaces because intensity of the G peak (I G ) is independent of disorder while intensity of the D peak (I D ) is directly proportional to disorder up to amporphization where the sp 2 aromatic ring is destroyed. S5, S6 In the current study, Raman spectra of ZYA and PG were collected using a custom-built spectrometer with a 532 nm solid-state laser and 40x (NA: 0.60) optical objective. The graphite sample was cleaved either with tape (ZYA) or razor (PG) right before data acquisition. Figure S1 shows the Raman spectra of ZYA and PG along with the corresponding peak attributions. A single Lorentzian was used for the curve fitting of all peaks except for the 2D peak of ZYA that required deconvolution into three Lorentzian peaks. S6, S7, S8 The G peak is most salient in the spectra of both materials, as expected for graphite. The D peak is easily seen for PG but is absent for ZYA, which indicates that a significantly greater level of Raman detectable defects are present on PG. The 2D peak around 2700 cm -1 is observed for both samples, but with different line shapes. The characteristic 2D peak for HOPG is observed in the spectra of ZYA with a sharp peak and a shoulder towards lower wavenumber. Contrarily, the 2D peak for PG is a single sharp peak that is redshifted 13.5 cm -1 compared to ZYA. Lespade et al. showed that a single 2D peak occurs for turbostratic graphite (i.e., graphite without AB interlayer stacking). S9 This result was confirmed through work by Ferrari et al. which demonstrated that the 2D peak lineshape is a function of the number of graphene layers approaching the S3

4 lineshape of graphite at 5 layers. S8 Further details of the 2D peak behavior for HOPG and turbostratic graphite were studied by Jorio et al. S10 Single layer graphene has no interlayer coupling with adjacent graphene layers resulting in a single 2D peak; whereas, interlayer coupling in multilayer graphene resolves two 2D peaks (D 1 and D 2 ). S7 Thus, the 2D peak lineshape of ZYA qualifies this sample as high quality graphite with ABAB graphene stacking. The single 2D peak of PG, along with its peak position nearer D 2, indicates that PG is turbostratic graphite with little (or no) interlayer coupling. Fitting results shown in Table S2 and Table S3 illustrate several differences between ZYA and PG. FWHM of the G peak is 15 cm -1 for ZYA, corresponding to the literature value for HOPG, and 22 cm -1 for PG. S6, S10, S11, S12 Since the G peak width is related to charge doping (e.g., intercalates, defects, etc.), S13, S14, S15 PG is qualified as having more defects than ZYA based on (a) presence of the D peak, (b) wider G peak, and (c) redshifted 2D peak resolved as a single peak. S12, S16, S17 The intrinsic disorder in ZYA and PG is well within the first stage of disorder classification proposed by Ferrari et al. considering (a) the expected graphitic nature of the samples, (b) FWHM of the G peak, (c) position of the G peak, and (d) sharp well defined D, G, and 2D peaks. S6 This validates the use of Equation S2, Equation S3, and Equation S4 to calculate defect density as described in the following section since these equations were derived from samples within the first stage of disorder. S5 S4

5 (a) ZYA PG G Normalized Intensity D 2D 1 2D 2 0 D (b) 100 G ZYA Lorentzian fit (c) 100 G PG Lorentzian fit Normalized Intensity D Normalized Intensity D D D Wavenumber (cm -1 ) Wavenumber (cm -1 ) Figure S1. Raman spectra of graphite samples. (a) Normalized spectra of ZYA of PG with peak attributions. (b) ZYA spectrum with the D and G peaks each fit with a single Lorentzian and the 2D peak fit with three Lorentzian curves. (c) PG spectrum with the D, G, and 2D peaks each fit with a single Lorentzian. Table S2. Raman spectra fitting results of ZYA and PG. Data fit using Lorentzian functions. Intensity (I) is determined from the peak height which is an extrinsic parameter of the Lorentzian function, thus there is no associated error. I 2D and FWHM 2D for ZYA are taken from the cumulative fit of the three sub-peaks. The fitting error of FWHM D for ZYA is unrealistically large because the peak was forced to fit yet no peak was conspicuously resolved. I D I G I 2D I D /I G I G /I 2D FWHM D FWHM G FWHM 2D ZYA PG (± ) (± 1.64) (± 0.79) (± 0.69) (± 5.58) (± 5.75) S5

6 Table S3. Raman spectra peak centers of ZYA and PG. x c of the D peak for ZYA was held constant to allow fit to converge. D G 2D 1 2D 2 2D 3 ZYA cm (± 0.3) cm (± 1.3) cm (± 0.3) cm (± 6.7) cm -1 PG (± 0.4) cm (± 0.2) cm (± 2.0) cm Quantify the surface defects Raman spectroscopy is not a surface sensitive technique and probes into the bulk of the sample, thus defects detected by Raman originate throughout the sample and not just at its surface. WCA, on the other hand, is a very surface sensitive technique and probes only the uppermost layers of the sample. S18 Therefore, in order to quantify only surface defects through Raman spectroscopy and correlate to wettability, it is necessary to determine the contribution of surface defects compared to the total amount of Raman detectable defects. The Beer-Lambert equation can be used to calculate penetration depth of the Raman laser into graphite: =e = Equation S1 where α=4πk/λ, k is the extinction coefficient of graphite (k=1.3), and λ is the laser wavelength (λ=532 nm). The penetration depth (z) is calculated to be 32.6 nm that is equivalent to ca. 97 graphene layers. This means that Raman detectable defects originate from the uppermost 97 graphene layers. To estimate the I D /I G ratio of the uppermost layer of graphene, the following assumptions are made: S19 (1) Only the top layer of S6

7 graphene contributes to the D peak. (2) Each graphene layer adsorbs 2.3% of incoming laser light as well as the Raman signal from the underlying graphene layers. (3) The Raman emission from each graphene layer is proportional to the laser intensity it experiences. (4) Exfoliation does not cause additional Raman detectable defects. With these assumptions, contribution of each graphene layer to the observed G peak intensity will decrease by ( ) 2 for each increase of layer depth; contribution from the top layer of graphene accounts for 4.5% of the observed G peak intensity. S19 This is reflected in Table S2 where I G is 4.5% of the actual experimental peak height to correct for surface sensitivity. We assume that only the top layer of graphene contributes to the D peak based on previous reports S6, S19, S20 showing that defects are substantially restricted to the uppermost graphene layer. For PG with more defect, the graphene layers below might also contribute to the D peak. However, this will not change our conclusion qualitatively. Cançado et al. showed that I D (L L /L d ) 2 where L L is the Raman laser spot size and L d is the average distance between defects. S5 Additionally, I G is independent of defect density: I G L 2 L. S5 Combining these equations yields I D /I G 1/L 2 d. S5, S6, S21 The distance between defects (L d ) can be calculated from I D /I G by Equation S and defect density can be calculated from Equation S where E L is the laser energy (λ=532 nm; E L =2.33 ev): S5 L nm = ± Equation S2 n defects cm = Equation S3 S7

8 Therefore, combining Equation S2 and Equation S3 allows for the defect density to be calculated: n defects cm = 7.4e9 ±2.5e9 E Equation S4 These equations were empirically determined for single layer graphene on SiO 2, thus they are relevant only for the uppermost graphite surface, which is also the area of interest in WCA analysis. S5, S21 These equations have also been used to calculate defect density on HOPG. S22 An important consideration for graphite samples is the size of crystalline domains (i.e., crystallite size). S12 Knight and White showed that the crystallite size (L a ) is inversely related to the ratio I D /I G (L a 1/(I D /I G )) for 2.5 < L a < 300 nm. S6, S16, S17 The average crystallite size for ZYA and PG are unknown but L a is proportional to the distance between defects (L d ) so one can roughly estimate that the actual crystalline domain is roughly equivalent to the distance between defects; therefore, L a L d 1/(I D /I G ). Calculation results using Raman spectra are shown in Table S4. Distance between defects is substantially larger for ZYA than PG, roughly 116 nm and 13 nm, respectively. Defect density of ZYA and PG is ca. 24 and 1778 defects/µm 2. Assuming that the domains are square, the domain area for ZYA and PG is calculated to be 13,380 nm 2 and 179 nm 2, respectively. The domains are ca. 75x larger on ZYA. Likewise, ZYA has substantially fewer defects per carbon atom than PG. In fact, ZYA has about an order of magnitude less defects than CVD synthesized graphene on copper (see Table S1 of Reference 23): S23 the defect density of G/Cu and ZYA is ca. 4.5 x defects/cm 2 and S8

9 0.24 x defects/cm 2, respectively. S23 For comparison, SPI-2 has 5.1 x defects/cm 2 and less than 13 x 10-6 defects/carbon atom, which is a greater defect density than ZYA. S19 Table S4. Defect calculations for graphite samples. L d is the distance between defects and n d is the defect density. Domain size is calculated assuming the defect free area is a square. The number density of carbon atoms is 3.85 x /cm 2. L d (nm) Defect Density n d (defects/µm 2 ) Domain area: L a (nm 2 ) Defects per carbon atom (x10-6 ) ZYA 116 (± 16) 24 (± 16) 13,380 (± 3778) 0.6 (± 0.4) PG 13 (± 2) 1778 (± 1187) 179 (± 51) 46.2 (± 30.8) References S1. Wenzel, R. N. Surface Roughness and Contact Angle. J. Phys. Colloid Chem. 1949, 53, S2. Kozbial, A.; Li, Z.; Conaway, C.; McGinley, R.; Dhingra, S.; Vahdat, V.; Zhou, F.; D Urso, B.; Liu, H.; Li, L. Study on the Surface Energy of Graphene by Contact Angle Measurements. Langmuir 2014, 30, S3. Eckmann, A.; Felten, A.; Mishchenko, A.; Britnell, L.; Krupke, R.; Novoselov, K. S.; Casiraghi, C. Probing the Nature of Defects in Graphene by Raman Spectroscopy. Nano Lett. 2012, 12, S4. Russell, A. M.; Lee, K. L. Defects and Their Effects on Materials Properties; John Wiley & Sons: Hoboken, NJ, S5. Cançado, L. G.; Jorio, A.; Ferreira, E. H. M.; Stavale, F.; Achete, C. A.; Capaz, R. B.; Moutinho, M. V. O.; Lombardo, A.; Kulmala, T. S.; Ferrari, A. C. Quantifying Defects in Graphene via Raman Spectroscopy at Different Excitation Energies. Nano Lett. 2011, 11, S6. Ferrari, A. C.; Robertson, J. Interpretation of Raman Spectra of Disordered and Amorphous Carbon. Phys. Rev. B 2000, 61, S7. Ferrari, A. C. Raman Spectroscopy of Graphene and Graphite: Disorder, Electron Phonon Coupling, Doping and Nonadiabatic Effects. Solid State Commun. 2007, 143, S8. Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K. Raman Spectrum of Graphene and Graphene Layers. Phys. Rev. Lett. 2006, 97. S9

10 S9. Lespade, P.; Marchand, A.; Couzi, M.; Cruege, F. Caracterisation De Materiaux Carbones Par Microspectrometrie Raman. Carbon 1984, 22, S10. Jorio, A.; Saito, R.; Dresselhaus, G.; Dresselhaus, M. S. Dispersive G -Band and Higher-Order Processes: The Double Resonance Process. In Raman Spectroscopy in Graphene Related Systems; Wiley-VCH Verlag GmbH & Co. KGaA, 2011, DOI: / ch12, pp S11. Lazzeri, M.; Piscanec, S.; Mauri, F.; Ferrari, A. C.; Robertson, J. Phonon Linewidths and Electron-Phonon Coupling in Graphite and Nanotubes. Phys. Rev. B 2006, 73, S12. Dresselhaus, M. S.; Kalish, R. Ion Implantation in Diamond, Graphite and Related Materials; Springer-Verlag Berlin Heidelberg1992. p 202. S13. Mélinon, P.; Hannour, A.; Bardotti, L.; Prével, B.; Gierak, J.; Bourhis, E.; Faini, G.; Canut, B. Ion Beam Nanopatterning in Graphite: Characterization of Single Extended Defects. Nanotechnology 2008, 19, S14. Meng, X.; Tongay, S.; Kang, J.; Chen, Z.; Wu, F.; Li, S.-S.; Xia, J.-B.; Li, J.; Wu, J. Stable P- and N-Type Doping of Few-Layer Graphene/Graphite. Carbon 2013, 57, S15. Ferrari, A. C.; Basko, D. M. Raman Spectroscopy as a Versatile Tool for Studying the Properties of Graphene. Nat. Nanotechnol. 2013, 8, S16. Knight, D. S.; White, W. B. Characterization of Diamond Films by Raman Spectroscopy. J. Mater. Res. 1989, 4, S17. Tuinstra, F.; Koenig, J. L. Raman Spectrum of Graphite. J. Chem. Phys. 1970, 53, S18. Hoffman, A. S.; Ratner, B. Lecture on Contact Angles. University of Washington, 2005, pdf. S19. Kozbial, A.; Li, Z.; Sun, J.; Gong, X.; Zhou, F.; Wang, Y.; Xu, H.; Liu, H.; Li, L. Understanding the Intrinsic Water Wettability of Graphite. Carbon 2014, 74, S20. Isaac Childres; Jauregui, L. A.; Park, W.; Cao, H.; Chen, Y. P. Raman Spectroscopy of Graphene and Related Materials. In New Developments in Photon and Materials Research, Jang, J. I., Ed.; Nova Science Publishers, S21. Lucchese, M. M.; Stavale, F.; Ferreira, E. H. M.; Vilani, C.; Moutinho, M. V. O.; Capaz, R. B.; Achete, C. A.; Jorio, A. Quantifying Ion-Induced Defects and Raman Relaxation Length in Graphene. Carbon 2010, 48, S22. Zeng, J.; Yao, H. J.; Zhang, S. X.; Zhai, P. F.; Duan, J. L.; Sun, Y. M.; Li, G. P.; Liu, J. Swift Heavy Ions Induced Irradiation Effects in Monolayer Graphene and Highly Oriented Pyrolytic Graphite. Nucl. Instrum. Meth. B 2014, 330, S23. Li, Z.; Wang, Y.; Kozbial, A.; Shenoy, G.; Zhou, F.; McGinley, R.; Ireland, P.; Morganstein, B.; Kunkel, A.; Surwade, S. P.; Li, L.; Liu, H. Effect of Airborne Contaminants on the Wettability of Supported Graphene and Graphite. Nature Mater. 2013, 12, S10