Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Surface Energy in Nanocrystalline Carbon Thin Films: Effect of Size-dependence and Atmospheric Exposure Manish Kumar, Amjed Javid, and Jeon Geon Han Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b04463 • Publication Date (Web): 16 Feb 2017 Downloaded from http://pubs.acs.org on February 18, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Surface Energy in Nanocrystalline Carbon Thin Films: Effect of Size-dependence and Atmospheric Exposure Manish Kumar,,1* Amjed Javid,1,2and Jeon Geon Han1* 1
Center for Advanced Plasma Surface Technology (CAPST), NU-SKKU Joint Institute for
Plasma Nano-Materials (IPNM), Advanced Materials Science and Engineering, Sungkyunkwan University, Suwon, 440-746, Republic of Korea. 2
Department of Textile Processing, National Textile University, Faisalabad-37610, Pakistan.
*Corresponding Author. Email:
[email protected],
[email protected] Keywords: surface energy; stability; carbon; neutral dominant process; size dependence.
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Abstract: Surface energy (SE) is the most sensitive and fundamental parameter for governing the interfacial interactions in nanoscale carbon materials. However, on account of the involved complexities of hybridizations states and surface-bonds, achieved SE values are often lesser in comparison to their theoretical counterparts and strongly influenced with stability aspects. Here, an advanced facing-target pulsed DC unbalanced magnetron sputtering process is presented for the synthesis of undoped and H/N doped- nanocrystalline carbon thin films. The time-dependent surface properties of undoped and H/N doped nanocrystalline carbon thin films are systematically studied. The advanced plasma process induced the dominant deposition of high energy neutral carbon species, consequently controlling the inter-columnar spacing of nanodomain morphology and surface-anisotropy of electron density. As a result, significantly higher SE values (maximum=79.24 mJ/m2) are achieved with a possible window of 79.24-66.5 mJ/m2 by controlling the experimental conditions. The intrinsic (size-effects and functionality) and extrinsic factors (atmospheric exposure) are resolved and explained on the basis of sizedependent cohesive energy model and long range van der Waal interactions between hydrocarbons molecules and the carbon surface. Obtained findings anticipate enhanced functionality of nanocrystalline carbon thin films in terms of selectivity, sensitivity and stability.
2 ACS Paragon Plus Environment
Page 2 of 41
Page 3 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
1. Introduction As the size reduces, surface properties become increasingly important. Despite the fact that bulk properties can be excellent for a specific application, the surfaces properties are needed to be physically/chemically engineered for the desired direction. Among other surface properties, surface energy (SE) is the most important one to decide material's compatibility and function with the environment. SE governs various kinds of interfacial phenomena of technological and fundamental importance related to the physical/chemical adsorption of molecules/bio-species1-8 and flow-dynamics of liquids9-11. Conventionally, wetting (desired for adhesives, paints etc) as well as nonwetting (sought for ship hulls, contact lens, integrated sensors, anti-sticking layers for lithography, self-cleaning etc) surfaces; both require specific SE engineering. New approaches have been proposed to fabricate multi-functional smart surfaces by controlling the SE.1,
5-7
Graphene layers are found to loose their protein adhesion function when SE is affected due to the physisorption of hydrocarbons.2 SE has critical consequences on the stability of nanoscale materials, as SE anisotropy is found to be a governing factor for Rayleigh-like solid state dewetting and nanowire stability.3 Further, bio-interfaces interactions have been explored in details for various biomedical applications i.e. tissue engineering, cell-cultivation and drug deliveries.2,5-8 Studies of condensation and flow of liquids in nanochannels and nanotubes has been the core agenda of nanofluidics field.9-11 Hydrophobic nanopores (~1 nm) in amorphous carbon membranes exhibited ultrafast viscous permeation of organic solvents.12 The light-driven motion of droplet has been realized in the photo-responsive surfaces through the gradient in SE.13 When it comes to the biocompatibility, the most researched material remains 'carbon' owing to its natural and prominent existence in all organic as well as in bio-species. The control
3 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 41
over microstructures and surface properties have led the advances of carbon nanostructures in nanophotonics,14 damping,15 field emission,16 electrochemical energy storage17,18 and sodium battery applications19. However, the size effects on SE of carbon based nanostructures have not much addressed. Theoretical calculation on the basis of density functional theory with local density approximations predicted SE values in order of ~70-80 mJ/m2 for hexagonal graphite.20 In general, the experimentally observed values of SE in various carbon materials are far lesser than this. For engineering the SE, main approaches move around the control over surface roughness and surface functionalization by physical/chemical methods.1-2,
21-27
Among others, plasma
surface deposition/treatment is an interesting choice since it may imply functionalization as well as the tailoring of roughness.21,23-26 It has been found that despite no correlation of SE with thickness for pristine graphene, O2 plasma treated graphene demonstrates a linear correlation with SE owing to the plasma induced defects.21 The values for SE of graphene and graphene oxide are found as 46.7 and 62.1 mJ/m2, respectively, whereas natural graphite flake has SE value as 54.8 mJ/m2.4 The incorporation of dopants in carbon strongly influence the SE values, however the generalization of trend is difficult to establish. For example, SE of the plasma treated Si incorporated diamond like carbon (DLC) films showed as 39.7± 8.9 mJ/m2, subjecting to reduction of SE on H2 and CF4 treatment whereas enhancement in SE on N2 and O2 treatments.23 An opposite trend was found when O2 plasma treatment decreased the SE of DLC from 40.8 to 15.3 mJ/m2.25 The treatment by allylcyanide plasma polymers increased the SE value of carbon fibers from 37.2 mJ/m2 to 44.3 mJ/m2.26 Mild heat treatments (in the range of 37 o
C-95 oC) decreased the SE in DLC, ta-C films as well as in stainless steel and titanium metal
surfaces.27 The doping of phosphorus in DLC increased the SE from 42.9 mJ/m2 to 72.4 mJ/m2.24 4 ACS Paragon Plus Environment
Page 5 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
The doping of Fe metals (10% in target) reduced SE of a-C from 42.8 to 25 mJ/m2.28 All these reports indicate that the experimentally observed SE values remain smaller relative to the theoretical values. Further, the contradictory reports23,25 with similar doping lead to ambiguity in understanding of SE variation and its correlation to experimental conditions. Numbers of methods are used for the evaluation of SE in nanoscale materials. For example, SE values were estimated using the calculations of contact angles data based on Fowkes, Owens-Wendth (extended Fowkes), and Neumann models.22 Approximation of solidliquid interfacial energy with the equation of state theory was applied to determine the graphene SE.4 Roenback et al. utilized in-situ scanning electron microscope peeling to quantify SE between multiwall carbon nanotubes and graphene.29 Surface tensiometric analyses of highly branched hydrocarbons surfactant indicated that a subtle structural modification in the tails and headgroup results in significant effects on limiting surface tensions at the critical micelle concentration where a higher level of branching and an increased counter-ion size promotes an effective reduction of surface tension to lower values (~24 mN m-1).30 The SE evaluation from heat of immersion measurements in various liquids was found to result mainly from Lifshitz-van der Waals intermolecular interactions.31 Recently, we found that low temperature plasma sputtering process provided high adhesion and proliferation for fibroblast as well as osteoblast cells in carbon thin films.32,33 When doped with Cu, these films also exhibited excellent antibacterial properties.34 In general, low temperature synthesis processes yield amorphous-carbon and crystallization initiates on increasing the substrate temperature or applying the substrate biasing.35,36 In specific plasma conditions, sp2 (graphitic phase character) and sp3 bonding (diamond phase character) can be
5 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
formed with relatively large-range order than amorphous structures (still short-range order with conventional crystals), and nanoclusters/ crystallites can be grown. This class of carbon coating is interesting in the sense that these can be deposited on flexible substrate and thus extending the possibilities of tailored performances beyond the conventional applications. For obtaining a new regime of SE values, we selected a different approach of synthesis, based on effective neutrals deposition in place of conventional ions governed deposition of carbon thin films. The objective of the present work is to study the dependence of experimental conditions on resulting SE values of undoped and N/H doped nanocrystalline carbon thin films synthesized by neutral governed sputtering technique. The synthesis conditions are studied through the plasma diagnostics and correlated to the films surface properties. The size-effect and stability of SE values are presented in details and explained by theoretical justifications.
2. Experimental Details 2.1 Process chamber and sample preparation Carbon thin films were prepared using facing target sputtering process having unbalanced close-field magnetic confinement. As shown in Fig. 1(a), the process chamber consists of two 4 inch graphite (99.99%) targets facing to each other, and substrates are placed off-axis to the centre of targets (1 cm below the bottom of targets). The magnetron consisted permanent magnets of high field strength. The process chamber was evacuated to 1 × 10−5 Torr base pressure by a turbo-molecular pump in conjunction with rotary pump. Ar gas was used to initiate the sputtering process. For the doping cases, H2 or N2 gases were used with 3 sccm flow rate, and reducing Ar flow appropriately to keep fixed working pressure. The doping level of both H2 and
6 ACS Paragon Plus Environment
Page 6 of 41
Page 7 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
N2 was 1.3% calculated from flow rate (H2/N2 :3 sccm, Ar: 230 sccm). A pulse DC (150 kHz frequency, 2.9 µsec pulse reversal time) power with varying the power density (4 to 20 W cm−2) was supplied to the facing electrodes at different working pressures (2.5 to 8.5 mTorr). Thin films of identical thickness (=200 nm) were deposited on silicon wafer substrates (1 cm × 1 cm). Prior to deposition, the silicon wafers were cleaned ultrasonically with acetone and ethanol each for 15 min. and dried with air-gun. 2.2 Plasma diagnostics Plasma diagnostics was performed by optical emission spectroscopy (OES) through Acton spectra 500i spectrometer. The spectrometer had resolution ≈ 0.05 nm, 10 µm wide entrance slit and grating of 1200 grooves mm− 1. The emitted photons were detected through CCD camera (PRIMAX Princeton Instrument), and spectra were acquired through WinSpec32™ software. Ion current density is measured using an oscilloscope (Tektronix TDS 3052B) assisted with a current probe amplifier (Tektronix TM502A). Total flux was calculated using the growth rate and measured density of thin films. 2.3 Characterization of prepared samples The film thickness was estimated by averaging the values measured through surface profiler (KLA-Tencor, Alpha-Step IQ) at five different positions. The surface morphological study was carried out through Field emission scanning electron microscopy (FE-SEM, JEOL JSM-6500F). X-ray photoelectron spectroscopy (XPS) was studied out using ESCA 2000 Instrument (VG Microtech, UK) having monochromatic Al Kα X-ray source. Core level peaks were deconvoluted using the XPSPEAK4.1 software. The electrical resistivity of films was measured by 4-point probe van der Pauw method (Chang Min Co. Ltd) and 2 point probe
7 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
(TrekModel 152–1 Resistance Meter). Raman spectroscopy was performed using 532 nm source wavelength (Alpha 300 M, WI Tec). Contact angle measurements were carried out using sessile drop method through contact angle goniometry (DSA 100, Krüss). A drop of distilled water (with 4 µL volume) was carefully added to the surface followed by capturing of drop image using adjacent camera. Measurements were observed at different locations and contact angle was calculated as average of five measurements. The same procedure was repeated with di-iodo-methane to measure the contact angle with nonpolar component.
3. Results and discussion The plasma chamber used for the films deposition, schematically shown in Fig. 1(a), consisted closed field type magnetron and two sputtering targets facing to each other. This configuration provided sufficient interaction volume for the sputtered ionic species to be neutralized through the recombination of electrons-ions. Therefore, unlike the conventional sputtering, present set-up provides the neutrals-dominated deposition. This was confirmed by actual monitoring of ion flux and total flux during the deposition; as shown in Fig. 1(b) and 1(c), respectively. It is observed that ion flux and total flux both increase with plasma power density as well as with the working pressure. Such behavior is justified due to the higher sputtered species at enhanced power and increase of Ar atoms in processing chamber. The comparison of both fluxes (at fixed power density and working pressure conditions) shows that total flux is almost 2 to 3 orders bigger in magnitude than ion flux., which confirms the dominance of neutrals species during the deposition of film. For the further plasma diagnostics, OES was performed in 350 to 550 nm wavelength range to study the effect of variation of the power 8 ACS Paragon Plus Environment
Page 8 of 41
Page 9 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
density and doping conditions, as shown in (d), (e) and (f) correspond to Ar, Ar+N and Ar+H plasma, respectively. The important spectral lines are indexed at 419.83 nm, 426.7 nm and 488.1 nm corresponding to Ar I, C II and Ar II transitions, respectively. The intensities of all these species enhance with increase in power density. An additional spectral line for N- addition, a spectral line at 357.7 nm is observed in corresponding to the excitation of N2 molecules. Similarly, spectral line at 656.28 nm is observed for H- addition corresponding to Hα. Furthermore, the intensity ratio of Ar II (488.1 nm) to Ar I (419.83 nm) emission lines are calculated. It is found that the intensity ratio of Ar II (488.1 nm) to Ar I (419.83 nm) emission lines, which is associated to the electron temperature, is increased with the increase in power density. At higher power density, the electrons get excited to the higher energy levels and gain much energy and exhibit the higher electron temperature. Whereas, this ratio is decreased when the working pressure is increased suggesting the decrease in electron temperature of plasma. The increase in pressure enhances the electron-neutral collision frequency because of shortening the mean free path that causes to decrease the electron energy. It has been explained in our earlier work.33 The microstructural information and size variation of grown nanocrystalline carbon films was investigated through Raman spectroscopy. After resolving each spectrum into two Gaussian peaks associated with D and G peaks of carbon, the ratio of area (ID/IG) under D and G peaks was calculated. The D peak contributed to the disordered graphitic carbon (bond stretching of sp2 atoms configured in both chain and ring structures) and G peak contributed to the graphitic carbon (the breathing modes of ring structures). The variation in ID/IG ratio of the Raman spectra of films prepared at varying power density and working pressure is shown in Fig. 2(a). On increasing the power density ID/IG ratio, initially, increases fast and then stabilizes. On the other 9 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 41
hand, on increasing the working pressure, ID/IG ratio decreases. Using the ID/IG ratio and the wavelength of laser, the crystallite size (La) of graphite was calculated using the expression;37
() = (2.4 × 10 )λ ( )
(1)
where λL is the wavelength of the laser source used in Raman spectrophotometer. ID and IG are integrated intensity (area under peaks) of D and G peaks, respectively. This formula is a generalized form of Tuinstra and Koening’s method38 for the estimation of crystallite size of nanocrystalline graphite or graphitic nanocrystals in carbon films. The generalized form does not depend explicitly on the laser wavelength and had been validated for different laser sources in visible region.37 The variation in average size of graphitic nanocrystals for the pristine films as a function of power density and working pressure is shown in Fig. 2(b). Similarly, ID/IG ratio corresponding to the films with N- and H- doping are shown in Fig. 2(c), whereas respective average crystallite sizes are shown in Fig. 2(d). It is found that average crystallite size slightly increases with increasing the working pressure (from 1.21 nm to 1.27 nm). Whereas, irrespective to doping/undoping it significantly decreases on increasing the power density. Relatively, crystallite size were bigger (1.33 nm- 2.24 nm) in N- doped carbon films and biggest (2.04 nm3.65 nm) in H- doped carbon films when compared to pristine films (1.24 nm -1.50 nm). Surface morphology of the films deposited at varying the plasma density are displayed as micrographs (a)-(d) in Fig. 3. From the micrographs, it is clear that films grow in columnar structures and exhibit nanodomain surface morphology. In cross-sectional images the columnar growth through the entire thickness is visible along a very thin over-layer of amorphous carbon. The reason for such amorphous like carbon as top layer is the inherent range of energy relaxation of fast coming ions/atoms on the surface. The densification always occurs beneath the top layer. 10 ACS Paragon Plus Environment
Page 11 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
The increase in power density produces the films with compact structures restricting the intercolumnar spacing. The dependence of working pressure on surface properties was reported in our earlier work,33 where it was found that increase in working pressure enhances the inter-columnar spacing. For the chemical bonding information of the surface, high resolution XPS was employed on selected samples (with extreme power densities) to obtain C-1s core level spectra. Fig. 3(e) and 3(f) represent the deconvolution of C-1s core level spectra into four components centered at 284.4, 285.4, 286.6 and 287.6 eV. The first two components correspond to C=C (sp2 hybridization) and C-C (sp3 hybridization) where as the last two components correspond to C bonded with O and O/OH, respectively. The relative hybridization of sp2, sp3 are calculated using the area under deconvoluted components of C-1s, and it is found that deposited films are majorly consisting sp2 hybridization (~60%) with significant contribution of sp3 hybridization (~26%). The local variation in hybridization can be understood with the consideration of observed total flux variation and energy consideration of deposited flux. As the power density increases, both of the total flux and kinetic energy of ionic species increase. However, the profile of energy distribution may not remain exactly similar when power density is increased; thus the processes which require higher energies are restricted in comparison to low energy processes. There was small increase in the intensity of the C=C with increase of power density. This observed variation is very small in comparison to the variation observed in Raman spectroscopy. The reason behind this is that despite the definite increase of C=C, the surface adsorbed carbon diminishes the quantitative values of C=C contribution. For the insight on the stability of films with atmospheric exposure, the surface electrical properties of the films were measured in fresh and after exposing to atmosphere for 7 days, as shown in Fig. 4. The increase in power density reduces the electrical resistivity almost to 3 order 11 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 41
magnitude in pristine (a), as well as in N-doped (c) or H-doped (d) carbon films. On the other hand, the increase in working pressure increases the resistivity (b). The observed variation of resistivity values with the power density and working pressure can be justified on the basis of structural compactness, which increases with power density and decreases with working pressure. This can be further supported by the consideration of almost linear dependence of electron temperature with power density, that means increase in electron energy. This will result in high ionization via inelastic collision and the acceleration of Ar ion to induce high deposition rate of carbon film with high sp2 content and therefore high conductivity is observed. The increase in pressure decreases the structure compactness that increases the number of grain boundaries acting as the barriers for transportation of charge. This phenomenon leads to increase the film resistivity. Further, the decrease in electron energy with the increase in pressure develops the film structure with reduced sp2 hybridization leading to decrease ID/IG. The presence of N further reduces the resistivity of the film that can be justified on the basis of increase in tunneling current density in the films followed by the decrease in energy gap due to higher graphitization.39 It is found that resistivity values do not remain same after the 7 days exposure. In general, resistivity values were higher after 7 days, which clearly shows that atmospheric exposure is a strong influencing parameter to govern the surface electrical properties of the films. On a relative note, the variation in resistivity after atmospheric exposure was highest for the case of working pressure, and almost stabilized surface electrical properties for the hydrogenated carbon films prepared power densities >15 W/cm2. The reason why films prepared at higher power density are more stable in properties can be understood by consideration of structural compactness. At higher power density the structure is more compact decreasing the available surface area for adsorption of atmospheric hydrocarbons and moisture. Large variation in N-doped films can be 12 ACS Paragon Plus Environment
Page 13 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
justified on the basis of atmospheric hydrocarbons/moisture molecules' interaction with surface through the van der Waal forces. The van der Waal interactions are stronger for higher electron density. Since the N-doped carbon films were more conductive, these films have stronger affinity for the adsorption of atmospheric molecular species. The contact angles of water and di-iodo methane were measured on carbon films prepared under various conditions, The SE values were calculated using the polar and nonpolar components consideration and repeated after 1, 3 and 7 days atmospheric exposure. The variation of water contact angles is shown in Table 1. As can be seen from table, in all conditions, fresh films exhibit very low values of water contact angles which increase with time on exposing to atmosphere. The atmospheric exposure dependence of SE in 4 cases (plasma power density, working pressure, and doping of H- and N- in career gas) are given as histograms of Fig. 5(a-d) depicting the role of experimental parameters and atmospheric exposure. The SE values of fresh films were very high 79.08 mJ/m2 for undoped film, 78.63 mJ/m2 for N- doped, 79.24 mJ/m2 for H- doped films. It is found that irrespective of doping/undoping, increasing the power density or decreasing the working pressure induce reduction in SE values in the fresh films which can be correlated to the size effects, explained later in the discussion. Size effects and doping effects can be considered as intrinsic factors whereas atmospheric exposure can be considered as extrinsic factor for governing the SE values. It can be further noted that in fresh prepared films the effect of doping is not huge on SE. Whereas, the variation in SE is large in all cases when these films are subjected to atmospheric exposure. By tailoring of intrinsic and extrinsic effects a SE window of 79.24-66.5 mJ/m2 is achieved. From stability point of view, H-doped C films exhibit the least variation in SE values when subjected to the atmospheric exposure for 7 days, which is in agreement to the observed electrical properties variation in these films, as presented in Fig. 4. 13 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 41
The properties of the surface, such as chemical structure, homogeneity, crystallinity, and the strength of cohesive interactions between atoms and molecules, as well as the physical shape; cumulatively provide the insights of various surface-interface phenomena. Condensed systems in equilibrium state minimize their SE by stabilizing their shapes with the smallest possible surface area. Thus the hybridization constituents with least SE are more probable towards the surface region. For the graphitic structure, 3 of the 4 valence electrons of carbon form normal covalent bonds (σ-bonds) with adjacent carbon atoms. Whereas 4th valence electron (π-electron) resonates between the valence band structures. The layer planes consist strong chemical bonding forces, yet the bonding energy between the planes is very small (1.3-4 kcal/gram atom) in comparison to that within the planes (150-170 kcal/gram atom). This difference of bonding energy was mostly explained on the basis of van der Waal forces between inter-layers. However, Spain found π orbitals as the source of inter-layer bonding, and overlapping of these π orbitals responsible for the high electronic mobility.40 These findings were mainly constrained for the bulk graphitic surfaces. It is to be noted that most of the theoretical calculations were carried out for very small cluster/unit cell using input parameters of bulk graphite.20,40 However theoretical studies didn't find any correlation of the classical SE with the slab thickness, which could be correlated to the size of nanocluster. The size dependence of SE in nanocrystalline materials is expressed with the consideration of size dependent cohesive energy;41 ()
= 1 −
exp(−
$%&
) '(
(2)
14 ACS Paragon Plus Environment
Page 15 of 41
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
where, )%* (+) is size dependent solid-vapor interface energy, D is the diameter of nanoparticles, Sb is the bulk coherent entropy of crystal, h denotes the atomic number, and R is the ideal gas constant. The expression provides that )%* (+) decreases with the decrease in size of the nanocrystal. Though this model was validated for Be, Mg, Na and Al thin films,41 we find that it gives fair justification to the observed variation of SE with the crystalline size, mentioned earlier as the intrinsic effect. The experimental size dependence on SE values is demonstrated in Fig. 6(a) for fresh films prepared under different conditions. It exhibits that for undoped and doped conditions, the variation of SE values shows a positive slope whose sharpness increases when the crystalline size is reduced. The possible mechanism for extrinsic effect (atmospheric exposure) on SE variation of carbon thin films is schematically presented in Fig. 6(b). Since, the growth of carbon films is in the form of columnar structures, the nano-openings between the columns may act as capillaries to accumulate the water molecules. In case of freshly prepared carbon films, the nanoopenings are empty to accommodate more water molecules, as a result films show lower water contact angle and higher values of SE. On interacting with the atmospheric environment, the increase in water contact angle and decrease in SE can be attributed to the adsorption of hydrocarbons on the surface of films. The hydrocarbons adsorption increases the hydrophobicity of the carbon films via long range van der Waal interactions between hydrocarbons molecules and the carbon surface. Using the second-order perturbation theory van der Waal coefficients, describing the dielectric response of the bulk solid to instantaneous dipole and quadrupole of adsorbed molecules, are expressed in terms of dynamic multipole polarizability of the adsorbed particles.42 Tao and Rappe calculated the value of van der Waal coefficients for several semiconductor surfaces including carbon (in diamond form).42 Since the bulk dielectric function 15 ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 41
values depend on the electron density, the nanodomain morphology of observed hybridization states suggests high values of the van der Wall's coefficients. During deposition the substrate temperature was 80 °C (due to plasma). This temperature was far below than the required for the crystallization. However, plasma process involves many excitation, dissociation, and ionization reactions of reactive species, atoms, molecules, and radicals, and hence multiple possibilities of energy deposition at the substrate location. The energy balance is controlled by heating processes generated by impinging particles/species, surface reactions and incoming radiation along with loss processes such as heat conduction and convection by emitted radiation and through the surrounding gas. It is not straight forward to calculate quantitative values accurately for each process. Note the activation energy for amorphous-crystallization transformation ~2.3 eV-2.7 eV. The maximum values of electron temperature was obtained as ~ 2.6 eV by Langmuir probe method in a closer plasma process for conductive carbon in our group.43 W. D. Westwood listed the various heat sources for substrates in a typical sputtering process as sputtered atom (5-25 eV), heat of condensation (2-5 eV), reflected Ar atom (10-200 eV), plasma radiation (
