CARBON
7 5 ( 2 0 1 4 ) 4 5 2 –4 6 4
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/carbon
Multifunctional three-dimensional nanodiamond-nanoporous alumina nanoarchitectures Morteza Aramesh a, Kate Fox a, Desmond W.M. Lau a, Jinghua Fang b, Kostya (Ken) Ostrikov b,c, Steven Prawer a, Jiri Cervenka a,* a b c
School of Physics, The University of Melbourne, Melbourne, Victoria 3010, Australia Plasma Nanoscience Centre Australia, CSIRO Materials Science and Engineering, PO Box 218, Lindfield, NSW 2070, Australia Plasma Nanoscience@Complex Systems, School of Physics, The University of Sydney, Sydney, NSW 2006, Australia
A R T I C L E I N F O
A B S T R A C T
Article history:
Hybrid composite nanomaterials provide an attractive and versatile material platform for
Received 22 November 2013
numerous emerging nano- and biomedical applications by offering the possibility to
Accepted 7 April 2014
combine diverse properties which are impossible to obtain within a single material. In this
Available online 13 April 2014
work, we present the fabrication of novel hybrid diamond and amorphous diamond-like carbon (DLC) coated nanoporous alumina materials that exhibit multiple functionalities, such as high surface area, quasi-ordered nanopore structure, tunable surface chemistry and electrical conductivity, excellent biological, chemical and corrosion resistance. These multifunctional nanohybrid materials are fabricated using the plasma-induced carbonization method that effectively modifies the surface and the inside of the nanopores of anodic alumina, producing a homogenous ultrathin DLC protecting layer over the whole external and internal surfaces of the membranes. We demonstrate that the interplay between internal and external carbon supply is a critical factor for the formation of the ultrathin sp3-bonded carbon layer in the nanopores. This study brings new insights in the DLC growth mechanisms in confined nanospaces and opens new avenues to fabricate hybrid, chemically resistant and biocompatible carbon-coated nanoarchitectures on other inorganic supports. 2014 Elsevier Ltd. All rights reserved.
1.
Introduction
Hybrid nanocomposites are a rapidly growing field of science in pursuit of novel materials with tailored functionality and improved properties [1,2]. Over the last decades, nanostructured single component materials, particularly those constructed with uniform nanopores, have become key components in many important industrial applications, such * Corresponding author. E-mail address:
[email protected] (J. Cervenka). http://dx.doi.org/10.1016/j.carbon.2014.04.025 0008-6223/ 2014 Elsevier Ltd. All rights reserved.
as nanoelectronics, catalysis, sensing, membrane separation and storage [2–4]. Development of novel hybrid multicomponent nanostructured materials promise to provide new possibilities for advanced applications, which are not possible with single materials alone. Functional materials currently used in nano- and biomedical applications are required to have versatile physical and chemical properties [2–6]. They should be easily nanostructured, possess tunable electrical, optical
CARBON
7 5 (2 0 1 4) 4 5 2–46 4
and other properties, variable chemistry and additionally they need to be chemically inert and biocompatible to exhibit long-term functional stability under a wide range of chemical and biological conditions. Fabrication of such multifunctional nanostructured materials is often difficult to achieve with just a single component material due to several considerable technical challenges or limitations in their chemical and physical properties. Nanoporous anodic alumina and nanoporous silica are examples of such single component inorganic materials with well-ordered pores, which have gained considerable attention in the last decade because of their ease of fabrication, rigidity, adjustable pore sizes and fairly narrow pore size distribution [7–27]. However, both porous anodic alumina and silica have limited stability in biological media, resulting in their dissolution, leaching and unintentional uptake by the mononuclear phagocyte system [16–20]. Their limited biostability is a direct consequence of low chemical resistance to alkaline and acidic solutions [16,17]. To overcome these limitations, researchers have attempted functionalize and modify the surface of anodic alumina using several different approaches, such as evaporating metals [21], coating with oxides [22,23], polymers [24,25], and carbon-based materials [26,27]. However, most of these approaches coated solely the top surface of the porous structures leaving the internal surface of pores unmodified and unprotected. Diamond and diamond-like carbon due to their exceptional properties have been long recognized as ideal materials for making devices that need to withstand harsh-environments. Diamond is biocompatible [28–30]; it has extreme hardness, wear and chemical resistivity [31–33]. Additionally, its surface chemistry [34,35], electrical and optical properties [32,33] can be tuned by atomic doping to suit the desired device application. Diamond-like carbon (DLC), an amorphous form of carbon containing a significant fraction of sp3 bonded carbon, shares many of these exceptional properties with diamond [36–39]. Despite the extensive research devoted to fabrication and processing of synthetic diamond and DLC [32–47], nanopatterning of these materials remains still challenging. Owing to the extreme hardness and chemical stability of sp3 carbon the top-down fabrication techniques for sculpting nanostructures in diamond are very limited [40–47]. Previous attempts of nanostructuring diamond have used the conventional optical, electron or mask template lithographic processes in combination with oxygen plasma etching, resulting in diamond nanostructures with pores size of the order of hundreds of nanometers and aspect ratios of about 10 (ratio of pore depth divided by pore size) [40–43]. As an alternative researchers have used the bottom-up approach to fabricate porous diamond and DLC nanostructures [44–53]. For example spatially selective diamond nucleation and growth have been applied to produce patterns in diamond films with lm resolution [45–47]. Subnanometersized pores have recently been produced in 10–40 nm thick free-standing DLC membranes by plasma polymerization of organic compounds [48]. These highly cross-linked networks of sp3 carbons demonstrated ultrafast permeation of organic solvents through the membranes. However, a facile fabrication method for producing nanoporous sp3 bonded carbon
453
nanostructures with variable pore size has not been reported yet. In this work we demonstrate a novel fabrication method of hybrid diamond and DLC coated nanoporous alumina materials that exhibit multiple functionalities. We take the advantage of the properties of both diamond and nanostructured anodic alumina templates to fabricate hybrid carbon-alumina nanomaterials with multiple functionalities, including high surface area, quasi-ordered nanopore structure (10–150 nm), tunable surface termination and electrical conductivity, excellent biological, chemical and corrosion resistance. We use an optimized plasma-enhanced chemical vapor deposition (CVD) process to coat the surface and inside of pores of anodic aluminium oxide (AAO) templates by an ultrathin DLC layer. We investigate the properties of these materials and the formation of DLC in the nanopores by Raman spectroscopy, scanning electron microscopy (SEM), high-resolution transmission microscopy (HRTEM), scanning transmission electron microscopy (STEM), electron energy-loss spectroscopy (EELS), X-ray photoemission spectroscopy (XPS) and near edge X-ray absorption fine structure (NEXAFS) spectroscopy.
2.
Experimental
2.1.
Fabrication of porous anodic alumina templates
The free-standing porous AAO membranes used in this study were fabricated from a pure Al sheet (99.999%). First the Al sheet was sonicated in ethanol for 5 min and washed with DI water. The sheet was then ‘‘electropolished’’ in a solution of ethanol/perchloric acid (4:1 v/v) for 3 min at 18 V at room temperature (RT). This electropolishing step made the Al surface smooth and shiny. In order to anodize the samples, we used a three step anodization process. In the first step, we placed electropolished Al in 0.3 mol/L oxalic acid (RT) for 12 h at 40 V. This resulted in a pore size of approximately 45 nm. Different pore sizes have been achieved by varying different electrochemical conditions, such as potential over the electrodes, temperature and electrolyte concentration [10,15]. Then the sample was soaked in a mixture of 0.2 mol/L H2CrO4 and 0.4 mol/L H3PO4 (12 h and RT) to remove the formed oxide from pores. In the second step of the anodization process, we repeated the anodization parameters from the first step for 20 h to achieve 100 lm deep pores. Finally, a saturated solution of CuCl2 was used to selectively remove aluminium from the backside, leaving a free standing AAO membrane. The free-standing AAO membrane was placed in 0.3 mol/L H3PO4 at 30 C for 90 min to open the closed pores. The result is a 100 lm thick free-standing AAO membrane with tunable pore size (20–150 nm).
2.2.
Chemical vapor deposition
Diamond and DLC films were grown on porous anodic alumina substrates by microwave-assisted CVD in a Cyrannus system from Iplas GmbH. NCD diamond growth was done in hydrogen/methane plasma at 920 C for 10 min using a microwave power input of 2 kW, chamber pressure of 80 Torr and total gas flow of 760 sccm (1.3% CH4 in H2). N-UNCD films
454
CARBON
7 5 ( 2 0 1 4 ) 4 5 2 –4 6 4
were grown in argon/methane plasma with 1250 W, pressure 50 Torr, passive stage heating set at 850 C and gas flow of 79 sccm Ar, 20 sccm N2 and 1 sccm CH4. Ultra-thin DLC films were grown on AAO for 10–60 min using the same hydrogen/ methane plasma conditions as for NDC films. N-DLC films were grown using the same plasma conditions as N-UNCD films. Different substrate temperature (750–1100 C) in the DLC growth was achieved by varying the power input in the range of 1250–2500 W. The substrate temperature was monitored by a pyrometer during the growth. Prior to the diamond CVD growth, AAO substrates were seeded with detonation nanodiamond/water colloidal mixture (0.1% concentration) in a ultrasonic bath (5 min) to initiate diamond nucleation and growth [54,55]. Prior to seeding, the surface of nanodiamond particles (G01 powder from PlasmaChem) was hydrogen terminated by annealing the nanodiamond powder in 99.999% H2/Ar (4.14% H2) gas at 800 C for 4 h. This treatment resulted in an improved water solubility of the nanodiamonds and zeta potential of the order of 50 mV. Nanodiamond/water colloidal mixtures were centrifuged at 20,000 rcf for 6 h to get mono-dispersed colloids with the mean particle size 4 ± 2 nm as determined by dynamic light scattering (DLS) on a Malvern Zetasizer Nano ZS fitted with a 633 nm laser.
2.3.
3.
Results and discussion
3.1.
Synthesis of diamond and DLC coated AAO templates
Chemical resistivity test
Acid boil cleaning was performed in a mixture of sulfuric acid (1 ml) and sodium nitride (0.25 mg) at 200 C for 1 h [56–58]. This procedure was used to remove any strongly bonded contaminants and sp2-bonded carbon. Other resistivity tests of the samples were done in saturated NaOH and KOH solutions, 40% HF in H2O and 25% HClO4 in ethanol, in which the samples were left for 24 h. Bioresistance evaluation was done using the accelerated aging protocol [59]. Samples were soaked in medical grade sterile saline (Aerowash Sterile Sodium Chloride Eyewash Solution) using capped glass vials. Using an environmental test chamber (MicroClimate Benchtop Test Chamber Cincinnati Sub-Zero) set at 80 C, samples were kept at temperature for a time period of 18 days. Using the Arrhenius equation, this equates to a lifespan of 6 months in vivo.
2.4.
(E = 1253.6 eV) at a power of 300 W using a 400 lm spot size. Samples were grounded and an electron flood gun was used to compensate for charging during the measurements. Semiquantitative in-depth composition of elements has been obtained by XPS depth profiling using ion sputtering with 3 keV Ar+ ions. Synchrotron XPS and NEXAFS experiments were conducted at the soft X-ray beamline at the Australian Synchrotron at a base pressure of 10 10 mbar and the total energy resolution better than 0.1 eV. The samples were degassed at 400 C in ultrahigh vacuum conditions for 2 h prior to the photoemission experiments. Photon energies were chosen to ensure maximum surface sensitivity for high resolution core level scans of C 1s (330 eV), O 1s (550 eV) and Al 2p (150 eV). The binding energy of all spectra was calibrated using the Au 4f 7/2 core level at 84.0 eV. The XPS data analysis was done using a Shirley background subtraction and peak fitting using Voigt functions. NEXAFS spectra were obtained in the partial electron yield (PEY) mode. The photon flux was monitored using the drain current from a half transparent gold mesh, allowing intensity and energy normalization across different scans.
Characterization
The overall morphology of the samples was characterized by SEM in a FEI Nova dual beam apparatus operated at 20 kV. HRTEM/STEM analysis was carried out in a FEI Tecnai TF20 HRTEM and Jeol 2100F TEM/STEM 200 kV systems equipped with an electron spectrometer for EELS. Crosssectional TEM specimens were prepared using a focus ion beam in a dual beam SEM apparatus. EELS quantification of elemental composition was done from Al, C, O K-edge core loss profiles by integrating the number of counts under the excitation edge using cross section calculated by Hartree-Slater model and power law background removal. Raman spectroscopy was conducted using a Renishaw Micro Raman spectrometer with an excitation wavelength of 514 nm in backscattering geometry. Laboratory-based XPS analysis was performed in a Thermo-Fisher K-Alpha apparatus (10 9 mbar) using a Mg Ka radiation source
The free-standing AAO structures were fabricated by a conventional two-step anodizing process of an aluminium film as previously reported [10,15]. This anodization process leads to close packed honeycomb-like pores with short distance ordering which form parallel channels of the membrane (Fig. 1a). Fabrication of long-range ordered pores can be done by pre-texturing the alumina surface or using a hard anodization process as previously demonstrated by others [11–13]. In this study, we have used nanoporous AAO templates with pore size in the range of 20–150 nm and 100 lm thickness. Different pore sizes have been obtained by varying the electrochemical conditions in the anodizing process, such as potential over the electrodes, temperature and electrolyte type and concentration [15]. As prepared free-standing porous AAO templates have been used as the growth substrates for the growth of diamond and DLC by microwave-assisted CVD. Fig. 1 summarizes three different approaches to fabricate hybrid diamond and amorphous carbon coated alumina membranes from nanoporous AAO by CVD. In the first method, AAO templates have been treated by hydrogen/ methane or argon/methane plasmas in CVD to produce ultrathin DLC coating over the whole surface of nanoporous AAO. In the other two methods, the surface of the AAO templates has been first pretreated by nanodiamond to promote diamond nucleation and growth in CVD. Diamond would not grow without this nanodiamond seeding step. For this purpose, the whole AAO sample was immersed in a colloidal dispersion of nanodiamond in water [54,55]. Prior to this, the surface of nanodiamond particles was hydrogen terminated in a furnace at 800 C with a H2/Ar gas mixture (5% H2) to improve adhesion of nanodiamonds to the alumina surface
CARBON
7 5 (2 0 1 4) 4 5 2–46 4
455
Fig. 1 – Schematic and SEM images of three different plasma conversion processes of anodic alumina templates into nanoporous sp3-bonded carbon (diamond and DLC) membranes. (a) Shows a pristine AAO membrane, (b) Method 1: AAO coated by DLC, (c) Method 2: AAO seeded by nanodiamond and exposed to ultrananocrystalline CVD diamond growth conditions, (d) Method 3: AAO seeded by nanodiamond and exposed to nanocrystalline CVD diamond growth conditions. The inset images show pore size distribution of the pores in the SEM images. (A color version of this figure can be viewed online.)
in water. Nanodiamond seeds have homogenously covered the entire surface of the anodic alumina as observed in SEM after the CVD growth in Fig. 1c and d. Diamond films have grown only on the top surface of the AAO substrates, while the internal surface of nanopores was coated by DLC as will be discussed later. The only difference between Method 2 and 3 was in the CVD diamond growth recipe used. In Method 3, we have applied the polycrystalline CVD diamond growth using the hydrogen/methane plasma, which resulted in a relatively rough holey nanocrystalline diamond (NCD) film with a 50 nm thickness and broad pore size distribution (Fig. 1b). From the distribution of pore size it can be seen that the pores have been significantly reduced in the NCD film compared to the pristine AAO membrane. By using the ultrananocrystalline (UNCD) diamond growth plasma in Method 2 (Fig. 1c), the roughness and the dispersion of the pore size distribution of the holey diamond film has been improved. The nonconductive diamond films have been made conductive by adding nitrogen in the UNCD growth [28], resulting in conductive holey nitrogen-incorporated ultrananocrystalline diamond (N-UNCD) films. Nevertheless both of these two methods have lead to a significant reduction and broadening of the
pore size distribution of the diamond coated AAO membranes. This can be explained by an ‘‘overlayer effect’’ caused by a finite roughness of the deposited diamond layer, which gives rise to a Gaussian broadening of the pore size distribution. In Method 1 in Fig. 1b, we have overcome the problem with the broadening of the pore size distribution by coating AAO templates by an ultrathin DLC layer in CVD. The pore size of the resulting DLC–AAO membrane was found the same as the original AAO template (46 ± 3 nm), suggesting that the DLC layer thickness is below the resolution limit of our SEM. The advantage of this plasma-induced surface modification method is that DLC coats both the surface and inside of the pores of an AAO template without compromising pore size and it does not require any nanodiamond preseeding. Additionally, this method can be combined with the two previous diamond growth methods when using optimized CVD growth conditions that produce both diamond and DLC to fabricate free-standing nanoporous diamond/DLC–AAO membranes with pore size smaller than of the original AAO templates as shown in Fig. 1c and d. Overall, we could tune the pore size and porosity of diamond and DLC coated AAO membranes by varying the pore size of the AAO substrate (see Fig. S1,
456
CARBON
7 5 ( 2 0 1 4 ) 4 5 2 –4 6 4
Supporting Information) and thickness of the grown diamond layer in the range of 10–100 nm. The electrical properties of the insulating diamond layers have been modified by incorporating nitrogen in the UNCD films to make a conducting top layer of the nanoporous diamond/DLC–AAO nanostructures.
3.2.
Chemical and biological resistance
To check the quality and permeability of the grown ultrathin DLC layer on the surface and inside of porous AAO templates we have tested its chemical stability in various strong acidic and alkali environments in which anodic alumina usually dissolves. Table 1 summarizes the chemical and biological resistivity of the DLC–AAO samples in comparison to other associated materials: pristine AAO, sapphire and diamond. The plasma-treated DLC–AAO survived in all strong alkali and acidic environments in which the uncoated AAO templates dissolved. This indicates that DLC coated-AAO provides a highly stable material for use in harsh chemical and biological environments. Furthermore, the quality and the stability of the formed diamond-like carbon layer have been verified by an acid boil treatment, a technique routinely used in the diamond community to clean off any residual impurities and sp2-bonded carbon from diamond surfaces [56,57]. Overall, the DLC coated alumina templates have exhibited excellent chemical and corrosion resistance as good as the extreme corrosion resistance of diamond [58]. The DLC coating on AAO has also significantly improved the biostability of AAO materials in an accelerated aging test, in which an uncoated AAO template have been severely degraded. Accelerated aging of the samples was done in sterile saline solution at 80 C for 18 days, which corresponds to a life span of 6 month in vivo at 37 C [59]. Diamond and DLC are acknowledged as biocompatible materials and are used in a variety of medical applications, such as bionic devices, orthopedic implants and heart valves [28–30,36–38]. Accordingly, the biocompatibility of our DLC coatings is anticipated.
3.3. AAO
Raman spectroscopy characterization of plasma-treated
To understand the growth mechanism of DLC layer on anodic alumina in CVD, we have characterized differently plasma modified AAO templates by Raman spectroscopy in Fig. 2. Fig. 2a shows the effect of different plasma gasses on the carbon layer formation on AAO templates. Pristine and hydrogen
Fig. 2 – Raman spectroscopy of AAO templates after exposure to different CVD plasma conditions. (a) The effects of different plasma gases at Tsub 900 C: H2, H2/CH4 mixture for DLC and seeded NCD, and N2/Ar/H2/CH4 mixture for N-DLC. (b) Raman spectroscopy of DLC–AAO as a function of the substrate temperature in CVD. The spectra have been obtained with a 514 nm excitation laser in the backscattering mode using a 100· objective. (A color version of this figure can be viewed online.)
plasma treated AAO materials exhibited no obvious carbon Raman bands in the Raman spectra, while all other samples treated by carbon containing plasmas showed clear carbon
Table 1 – Comparison of chemical and biological resistance of AAO, sapphire, DLC–AAO and diamond. Chemical
pH
T (C)
Saturated KOH or NaOH Saturated KOH or NaOH HF 40% HClO4 25% Acid boil (H2SO4 + NaNO3) 1 h Bioresistivity test for 6 months in vivo using accelerated aging
14 14 3.5 1 1 5.5
20 60 20 20 200 80
+ Resistant, not resistant (dissolution of the samples). Partly etched. b 100 lm thick membranes with 50 nm pores. a
AAOb
Sapphire
DLC–AAOb
Diamond
+
+ + + + + +
+ + + + +
+ a a
CARBON
7 5 (2 0 1 4) 4 5 2–46 4
D and G Raman modes at 1350 and 1600 cm 1, respectively. These Raman spectral features are signatures of sp2-bonded carbon in nanoscopic diamond, diamond-like carbon or glassy carbon films [60]. The sp2 sites dominate the Raman spectra of amorphous carbon films because the cross-section of sp2 phase is much higher (50–250 times) than that of the sp3 phase in visible excitation [60]. The first order diamond Raman band at 1332 cm 1, corresponding to the crystalline sp3 carbon phase, was observed only in nanocrystalline diamond films grown on detonation nanodiamond seeded AAO templates. In DLC–AAO materials, the width and ratio between D and G carbon Raman peaks (I(D)/I(G) ratio) could be modified by changing the substrate synthesis temperature in the CVD process in Fig. 2b, suggesting a temperature driven growth process. The highest I(D)/I(G) ratio and the broadest G peak have been measured in samples fabricated at temperature of 850–950 C. These samples had also the highest sp3 carbon content as determined by other techniques. The
457
carbon coatings became considerably more graphitic above 1100 C.
3.4. Plasma modification of chemical composition of AAO templates X-ray photoelectron spectroscopy investigation of differently CVD plasma modified AAO templates have been performed to gain more information about the influence of the plasma on the chemical composition and local bonding environment in the hybrid materials. Fig. 3a shows wide XPS survey spectra of uncoated and DLC coated AAO membrane materials. Both spectra show that the samples contain only three elements: carbon, oxygen and aluminium. XPS depth profiling of the pristine AAO template by argon ion sputtering revealed a high amount of carbon (5–10%) in the anodic alumina materials. This has been caused by incorporation of different organic chemicals (oxalic acid) and OH ions used
Fig. 3 – XPS spectra (E = 1253.6 eV) of AAO templates after treatment in different plasma gases: H2, H2/CH4 mixture for DLC and seeded NCD, N2/Ar/H2/CH4 mixture for N-DLC, and O2 plasma on DLC–AAO. (a) Survey XPS spectra of pristine AAO and DLC– AAO materials. (b) XPS depth profile showing the relative chemical composition of the samples obtained by Ar+ ion sputtering. High resolution XPS spectra of the bulk porous AAO materials at 1 lm sputter depth: (c) carbon C 1s, (d) oxygen O 1s, (e) Al 2p, and (f) N 1s spectra. (A color version of this figure can be viewed online.)
458
CARBON
7 5 ( 2 0 1 4 ) 4 5 2 –4 6 4
during the anodization process as previously reported [61– 63]. The observation of AlOOH and OH groups in the high resolution XPS scans further confirms this finding. It is important to note here, however, that the XPS spectra of sputtered samples reveal information not only about the bulk material at the sputtered depth but also about the surface of the pores inside the membranes owing to their porous structure. For this reason we show all high resolution XPS spectra in Fig. 3 after Ar+ ion sputtering to a 1 lm depth of the samples to account for both bulk and internal nanopore surface modification of the AAO material. The high resolution XPS spectra of the external membrane surface followed similar trend and can be found in Fig. S2 in Supporting Information. Fig. 3b shows that the elemental composition ratio has changed after the CVD modification from uncoated AAO to DLC-coated AAO. Surprisingly, however, the carbon content has decreased after the H2/CH4 plasma treatment. This result agrees with the observation of no detectable carbon in the H2 plasma treated AAO in high resolution C 1s spectrum in Fig. 3c, where all carbon has been removed from both the surface and bulk of anodic alumina. The same result has been found for Ar plasma treated AAO. The carbon depletion in the DLC–AAO material has followed an exponential decay profile from the sample surface (Fig. S3, Supporting Information), suggesting that the carbon removal is induced by plasma. Interestingly, DLC has not been formed in the DLC growth plasma on the already hydrogen-treated AAO templates with diminished carbon content. By including nitrogen in the CVD plasma gas mixture we could clarify the prominent influence of the plasma on the DLC formation. Nitrogen-incorporated DLC has been formed on the surface and inside of nanoporous AAO. This result confirms that the growth of DLC inside of nanopores is not due to high temperature induced transformation but it is caused by reactive species from the plasma. Nitrogen signatures are seen in the XPS spectra in Fig. 3 as N–C, N–Al and N–O at binding energy of 289.9 (C 1s), 397.9 (N 1s), 533.4 (O 1s) and 73.6 eV (Al 2p). The positions of these XPS peaks are in accordance to previous photoelectron studies of nitrogen implanted DLC and aluminium nitride [64,65]. The occurrence of Al–C bonds at binding energy of 281.5 eV in the C 1s spectrum is attributed to the Ar+ ion sputtering and atom intermixing, since this peak has not been observed on unsputtered samples (Fig. S2, Supporting Information). The heat treatment of AAO templates at 900 C during the CVD process has not caused any significant change in the chemical bonding of the AAO material, as seen by no change of the O 1s and Al 2p XPS peaks in H2 plasma treated and pristine AAO materials. A slight upshift of the Al2O3 peaks in the O 1s and Al 2p XPS spectra of DLC–AAO compared to the pristine AAO can be explained by increased carbon content in the top surface layer. By exposing the nanoporous DLC–AAO structures to oxygen plasma (50 W) for 30 s, we have changed the as grown hydrogen-terminated surface of DLC in nanopores to oxygen-terminated DLC surface (O-DLC) as seen in C 1s lines in Fig. 3. Oxygen termination on diamond and DLC surfaces opens up a vast variety of chemical strategies for further surface functionalization by different organic functional groups [34,35].
3.5. Determination of carbon bonding in DLC–AAO by synchrotron-based XPS and NEXAFS So far, based on the laboratory-based XPS and Raman spectroscopy data we can conclude that CVD treatment of AAO gives rise to carbon layer formation inside of the nanopores, however none of the used techniques could provide us with information about the carbon–carbon bonding in these layers. To obtain this information we have performed synchrotron-based
Fig. 4 – Synchrotron XPS spectra of the top surface of DLC– AAO: (a) C 1s core-level spectrum taken with 330 eV X-ray radiation, (b) O 1s core-level spectrum taken with 550 eV Xray radiation, and (c) Al 2p core-level spectrum taken with 150 eV X-ray radiation. (A color version of this figure can be viewed online.)
CARBON
7 5 (2 0 1 4) 4 5 2–46 4
Fig. 5 – C K-edge NEXAFS spectra of DLC–AAO, highly ordered pyrolytic graphite (HOPG) and H-terminated detonation nanodiamond, and an energy-loss near-edge structure spectrum of a DLC layer in the bulk DLC–AAO obtained in cross-sectional STEM–EELS (Fig. 6). The NEXAFS spectra were obtained using the partial electron yield mode and normalized to the area under the spectra. (A color version of this figure can be viewed online.)
XPS and NEXAFS measurements on DLC–AAO in Figs. 4 and 5. Unlike Raman spectroscopy, NEXAFS represents a powerful technique for probing near surface structural composition of carbon materials and quantitative determination of the sp2 and sp3 carbon content [66–68]. Moreover, both XPS and NEXAFS (obtained in partial electron yield mode) measurements are extremely surface sensitive (1 nm) when using low energy Xray photons, which is essential for the investigation of the ultrathin DLC coating on AAO templates. To ensure that the sample surfaces were clean of any possible hydrocarbon contamination, DLC–AAO samples have been annealed to 400 C in ultrahigh vacuum (10 10 mbar) for 2 h prior to the photoemission experiments. Fig. 4a shows the C 1s XPS spectrum of the top surface of diamond-like carbon film on DLC–AAO obtained with a 330 eV excitation. The deconvolution of C 1s spectrum in Fig. 4a shows that the carbon peak is composed from two C–C and a broad range of C–O components (fitted by a broad peak at 285.7 eV). The sp3 and sp2 carbon is observed at binding energy of 284.4 and 284.8 eV, respectively. The ratio between the sp3/sp2 peak areas is about 70%. This result is consistent with the carbon K-edge NEXAFS measurements below. Fig. 5 shows a comparison of C K-edge NEXAFS and EELS spectra between DLC–AAO, highly oriented pyrolytic graphite (HOPG) and hydrogen terminated detonation nanodiamond (4 nm). The carbon K-edge NEXAFS of DLC–AAO shows both sp2 (p* and r*) and sp3 (r*) resonances [66]. A similar carbon K-edge spectrum has also been measured in EELS on DLC layers in the bulk DLC–AAO obtained on cross-sectional TEM samples (Fig. 6). The smaller pre-edge resonance at 285.4 eV is due to transitions of C 1s electrons to the p* orbital of sp2 bonded carbon, similarly like the dominant carbon K-edge in HOPG. The band-edge at 289 eV corresponds to the r* sp3
459
carbon and C–H bonding. The spectral features above this energy represents overlapping transitions from C 1s to r* sp2 and sp3 carbon sites. A nanodiamond sample exhibits an additional dip in the spectrum at 302.5 eV, the second band gap of diamond [67], which is not present in the DLC films. Since the untreated diamond does not have any p* sp2 components in front of the carbon K-edge, this peak (surface area) can be used as a measure of the sp2 carbon content in DLC [67,68]. The sp2 carbon content in the DLC film on AAO (determined from the p* K-edge) is about 35% and therefore the majority of the film is made of sp3 bonded carbon. This sp3/ sp2 ratio in DLC corresponds to tetrahedral hydrogenated amorphous carbon (ta-C:H) [39]. Although some of the carbon atoms have formed bonds to Al atoms, as indicated by a peak at 79.3 eV in the Al 2p XPS spectrum in Fig. 4c, we believe that this signal originates from the subsurface carbon present in the c-alumina layer as shown in Fig. 6. This is also supported by observation of no Al–C peak in the C 1s XPS spectrum, which should be present below 284 eV [69].
3.6.
Atomic structure of DLC–AAO membranes
HRTEM and STEM images of the atomic structure of the hybrid porous DLC–AAO materials are shown in Fig. 6. Since HRTEM is rather insensitive to amorphous phase compared to crystalline phase, this study reveals information predominantly about the crystalline constituents of the DLC–AAO material. Both cross-sectional HRTEM and STEM images in Fig. 6 show that the surface of the inner pores was strongly modified by carbon at the surface of pores. A comparison between uncoated and DLC-coated AAO can be found in Fig. S4 in Supporting Information. Spatial STEM–EELS mapping across the pore walls of DLC–AAO membrane with a ˚ ) clearly shows that the carbon focused electron beam (2 A layer is located within 2–5 nm from the pore surface (Fig. 6a). The elemental composition in the EELS line scan was determined from the energy-loss near-edge structures of C, O and Al K-edges. The overall structure of a DLC–AAO pore wall seen in a HRTEM cross section in Fig. 6c can be divided into four material layers. The bulk part of the material consists of amorphous anodic alumina that has been partly crystallized by the plasma-induced heat treatment at 850–950 C. As we go closer to the surface the alumina undergoes phase transition from amorphous carbon-rich anodic alumina to nanocrystalline c-alumina. Previous studies have shown that anodic alumina undergoes a rather complex smooth transition from amorphous hydrated anodic alumina to c-AlOOH/Al2O3 and c-Al2O3, via other transition alumina phases to a-Al2O3, when heated from room temperature to 1200 C [70–72]. This phase transition strongly depends on the type of the starting alumina material and the exact sequence of the transition alumina crystal phases is still heavily debated in literature [70– 72]. The nanocrystalline phase of c-alumina is observed in the subsurface layer of DLC–AAO pore walls with a nanoparticle size of the order of 1–5 nm and a lattice spacing of 2.04 ˚ . This spacing represents a distance between the (400) A planes. The buffer layer and bulk anodic alumina layers, on ˚ corresponding the other hand, show a lattice spacing of 2.42 A to the (311) crystal planes. The top surface layer of the hybrid
460
CARBON
7 5 ( 2 0 1 4 ) 4 5 2 –4 6 4
Fig. 6 – Cross sectional HRTEM/STEM images of a DLC–AAO membrane. (a) A STEM image of pore walls of the membrane with a line EELS profile of an elemental composition across the red line that was determined from Al, O and C K-edge energy-loss near-edge structures. (b) A selected area diffraction pattern of DLC–AAO. (c) A cross sectional HRTEM image of a membrane wall showing four material layers. HRTEM and FFT images of the atomic structure of (d) the uppermost DLC/c-Al2O3 surface layer, (e) the buffer layer composed of a mixture of different Al2O3 phases and (f) a bulk AAO material composed predominantly of amorphous alumina and small crystal nuclei. (A color version of this figure can be viewed online.) nanostructure wall is assigned to diamond-like hydrogenated amorphous carbon. The selected area diffraction pattern of the DLC–AAO membrane in Fig. 6b shows only the typical signatures of thermally treated anodic alumina represented by the (311), (400), (440) and (444) diffraction rings of the c-alumina phase [73,74]. The characteristic weak and broad diffuse diffraction rings of amorphous DLC could not be distinguished in the diffraction pattern because they coincide with the much stronger diffraction pattern of nanocrystalline c-alumina [39,75].
3.7. Growth mechanism of ultrathin DLC on/inside nanoporous AAO A summary of the key findings of our experiments is as follows. (a) An utrathin (2–5 nm) DLC layer with a majority of sp3 carbon bonding is grown on the surface and inside of the pores of nanoporous AAO in MW-CVD at temperatures in the range of 850–950 C. (b) DLC forms on AAO in both
the polycrystalline and ultrananocrystalline CVD growth gas mixtures. (c) The DLC layer thickness has not been found to increase with prolonged growth time (10–60 min). (d) DLC has not grown on the AAO samples that have been previously treated by hydrogen plasma, which removed most of the carbon from the AAO. Additionally, the overall carbon content in AAO has been found to be lower after the DLC growth. (e) Annealing of AAO membranes to 800–1000 C has not formed DLC layers on AAO, suggesting that the temperature-induced phase transformation is not sufficient to produce DLC without the use of the plasma. (f) DLC forms a uniform coating over the whole surface of nanoporous AAO templates, providing an impermeable protecting layer that is resistant to highly corrosive chemicals even at elevated temperatures. Based on these observations we propose that the growth and nucleation mechanism of the ultrathin diamond-like carbon film in/on the surface of AAO nanopores occurs as schematically shown in Fig. 7. This process is accompanied by a temperature-induced phase transformation of the
CARBON
7 5 (2 0 1 4) 4 5 2–46 4
Fig. 7 – Schematic illustration of the plasma-induced selflimited growth mechanism of a ta-C:H layer in anodic alumina nanopores accompanied by temperature-induced phase transformation of anodic alumina in the bulk and subsurface AAO.
amorphous hydrated carbon-rich anodic alumina to nanocystalline c-Al2O3 in the bulk and subsurface of the AAO material. The growth mechanism of DLC in the porous structure of anodic alumina can be considered formally similar to the CVD growth of DLC on flat substrates [39]. However, the plasma conditions differ significantly due to the confined space of nanopores as well as the supply of carbon atoms through micrometer long nanopores [76]. The three steps of the self-limited DLC growth mechanism are as follows. Step (1): plasma electron and ion collisions with the substrate atoms cause strong heating of the sample, giving rise to segregation of carbon, water and CO2 from the bulk and subsurface carbon- and hydroxyl-rich anodic alumina material to the surface. Anodic alumina contains high amount of chemical residues from the anodization process (H2CO4 and H2O), which are present close to the surface of nanopores [62]. The segregated carbon forms a layer on the surface of nanopores, and most likely acts as a nucleation layer for the further DLC growth. The DLC growth has not been found to occur on samples (H2-treated AAO) with low carbon content. The supply of carbon in the nucleation and growth process is therefore not only from the plasma gas but also from the bulk and pore surfaces of anodic alumina. Step (2): The segregated carbon layer undergoes transformation to a dense amorphous hydrogenated carbon phase via a plasma subplantation process by energetic carbon, hydrocarbon, and hydrogen species [39,77]. These bombard the surface and are subsequently stopped and incorporated in subsurface layers. Carbon-ions can be implanted also into alumina or sapphire as previously observed [78]. The density of this plasma transformed carbon phase increases with subplantation until it reaches saturation and all reactive carbon sites are saturated by hydrogen termination. The amorphous carbon matrix contains a significant amount of hydrogen similarly to a-C:H due to the much higher number of energetic hydrogen atoms (two orders of magnitude larger) than
461
that of energetic carbon ions [77]. Since alumina and sp3 carbon are good insulators, the AAO membrane will be strongly charged by the plasma. Owing to a very high aspect ratio of the AAO pores (20–150 nm pore diameter and 100 lm pore length), it is highly improbable that ions would pass through the narrow nanopores without being stopped by a collision with the charged walls. This argues against the proposition that ionized species are responsible for DLC growth deep in the pores. It is thus expected that neutral atomic and molecular species and radicals play the key role in the DLC formation in AAO nanopores. The overall growth mechanism of DLC is most probably controlled interplay between the subplantation process from the plasma and carbon supply from the AAO material. Further numerical modeling studies should elucidate their relative importance. Step (3): The growth is spontaneously terminated when the supply of carbon from the subsurface is finished. This has been confirmed by different time exposures (10–60 min) of the anodic alumina templates to the CVD plasma in which the plasma has not changed the film thickness and structure. This is likely due to the strong sp3 carbon bonding under which the film is not significantly etched by hydrogen plasma. This self-limited DLC growth model can be useful for understanding of the growth of diamond-like carbon in different nanostructured materials and induced by plasma polymerization of organic compounds [39,48]. If the nucleation and the growth of diamond-like carbon really does occur in this manner the understanding that our findings provide should assist in the development of methods for growing three-dimensional DLC coatings inside of various porous materials.
4.
Conclusions
We have demonstrated a new technique to fabricate hybrid diamond and DLC coated nanoporous alumina materials that exhibit multiple functionalities, such as high surface area, quasi-ordered nanopore structure, tunable surface termination and electrical conductivity, excellent biological, chemical and corrosion resistance. These multifunctional nanohybrid materials have been fabricated using the plasma-induced carbonization method that effectively modifies the surface and the inside of the nanopores of anodic alumina, producing a homogenous ultrathin (2–5 nm) DLC protecting layer over the whole external and internal surfaces of the membranes. Conductive and nonconductive nano-diamond films have been grown on the surface of AAO templates by CVD using the high nucleation density nanodiamond seeding method. Our measurements have shown that the ultrathin DLC coatings provide excellent corrosion resistance to the anodic alumina support against all tested harsh chemical (acid/alkaline) environments. Such an improved corrosion resistance of AAO by a thin sp3 bonded carbon layer with only a few nanometer thickness can be seen as a carbon parallel to protection that provide thin native oxide layers on other materials, such as aluminium oxide on aluminium. We demonstrate that the interplay between internal and external carbon supply is a critical factor for the formation of the ultrathin DLC layer in the
462
CARBON
7 5 ( 2 0 1 4 ) 4 5 2 –4 6 4
nanopores. Based on the experimental observations we have proposed a new model of the self-limited growth mechanism and nucleation of sp3 bonded carbon inside the porous structures. This study opens new avenues to protect the surface of porous nanostructured materials by corrosion resistant, chemically inert and biocompatible sp3-bonded carbon, which is useful for many technological applications.
Acknowledgements The authors wish to acknowledge the facilities and the scientific assistance of the Australian Micrsoscopy & Microanalysis Research Facility at RMIT and the Soft X-ray beamline at the Australian Synchrotron, Victoria, Australia. The authors also wish to thank Owen Burns and the Bionics Institute for access to the environmental test chamber. We acknowledge financial support from the University of Melbourne research and CSIRO top-up scholarships and an Australian Research Council Grant: ARC DECRA: DE120101100.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.carbon.2014. 04.025.
R E F E R E N C E S
[1] Kim Y, Lee J, Yeom JMS, Shin JW, Kim H, Cui Y, et al. Strengthening effect of single-atomic-layer graphene in metal-graphene nanolayered composites. Nat Commun 2013;4:2114. [2] Merhari L. Hybrid nanocomposites for nanotechnology. Boston, MA: Springer-Verlag, US; 2009. [3] Pendergast MTM, Hoek EMV. A review of water treatment membrane nanotechnologies. Energy Environ Sci 2011;4:1946–71. [4] Stroeve P, Ileri N. Biotechnical and other applications of nanoporous membranes. Trends Biotechnol 2011;29:259–66. [5] Simovic S, Losic D, Vasilev K. Controlled drug release from porous materials by plasma polymer deposition. Chem Commun 2010;46:1317–9. [6] Xie X, Xu AM, Leal-Ortiz S, Cao Y, Garner CC, Melosh A. Nanostraw-electroporation system for highly efficient intracellular delivery and transfection. ACS Nano 2013;7:4351–8. [7] Adiga SP, Jin C, Curtiss LA, Monteiro-Riviere NA, Narayan RJ. Nanoporous membranes for medical and biological applications. Wiley Interdiscip Rev Nanomed Nanobiotechnol 2009;1:568–81. [8] Jeon G, Yang SY, Kim JK. Functional nanonporous membranes for drug delivery. J Mater Chem 2012;22:14814–34. [9] Jani AMM, Losic D, Voelcker NH. Nanoporous anodic aluminium oxide: advances in surface engineering and emerging applications. Prog Mater Sci 2013;58:636–704. [10] Fang JH, Spizzirri P, Cimmino A, Rubanov S, Prawer S. Extremely high aspect ratio alumina transmission nanomasks: their fabrication and characterization using electron microscopy. Nanotechnology 2009;20:065706.
[11] Masuda H, Yamada H, Satoh M, Asoh H, Nakao M, Tamamura T. Highly ordered nanochannel-array architecture in anodic alumina. Appl Phys Lett 1997;71:2770–2. [12] Asoh H, Nishio K, Nakao M, Tamamura T, Masuda H. Conditions for fabrication of ideally ordered anodic porous alumina using pretextured Al. J Electrochem Soc 2001;148:B152–6. [13] Lee W, Ji R, Go¨sele U, Nielsch K. Fast fabrication of long-range ordered porous alumina membranes by hard anodization. Nat Mater 2006;5:741–7. [14] Kipke S, Schmid G. Nanoporous alumina membranes as diffusion controlling systems. Adv Funct Mater 2004;14:1184–8. [15] Lei Y, Cai W, Wilde G. Highly ordered nanostructures with tunable size, shape and properties: a new way to surface nano-patterning using ultra-thin alumina masks. Prog Mater Sci 2007;52:465–539. [16] Lee CW, Kang HS, Chang YH, Hahm YM. Thermotreatment and chemical resistance of porous alumina membrane prepared by anodic oxidation. Korean J Chem Eng 2000;17:266–72. [17] Karlsson M, Palsgard E, Wilshaw PR, Di Silvio L. Initial in vitro interaction of osteoblasts with nano-porous alumina. Biomaterials 2003;24:3039–46. [18] Lee CH, Cheng SH, Wang YJ, Chen YC, Chen NT, Souris J, et al. Near-infrared mesoporous silica nanoparticles for optical imaging: characterization and in vivo biodistribution. Adv Funct Mater 2009;19:215–22. [19] La Flamme KE, Popat KC, Leoni L, Markiewicz E, La Tempa TJ, Roman BB, et al. Biocompatibility of nanoporous alumina membranes for immunoisolation. Biomaterials 2007;28:2638–45. [20] Bru¨ggemann D. Nanoporous aluminium oxide membranes as cell interfaces. J Nanomater 2013;2013:1–18. [21] Velleman L, Shapter JG, Josic D. Gold nanotube membranes functionalised with fluorinated thiols for selective molecular transport. J Membr Sci 2009;328:121–6. [22] Xiong G, Elam JW, Feng H, Han CY, Wang HH, Iton LE, et al. Effect of atomic layer deposition coatings on the surface structure of anodic aluminum oxide membranes. J Phys Chem B 2005;109:14059–63. [23] Elam JW, Routkevitch D, Mardilovich PP, George SM. Conformal coating on ultrahigh-aspect-ratio nanopores of anodic alumina by atomic layer deposition. Chem Mater 2003;15:3507–17. [24] Popat KC, Mor G, Grimes CA, Desai TA. Surface modification of nanoporous alumina surfaces with poly(ethylene glycol). Langmuir 2004;20:8035–41. [25] Losic D, Cole MA, Dollmann B, Vasilev K, Griesser HJ. Surface modification of nanoporous alumina membranes by plasma polymerization. Nanotechnology 2008;19:245704. [26] Popp A, Engstler J, Schneider JJ. Porous carbon nanotubereinforced metals and ceramics via a double templating approach. Carbon 2009;47:3208–14. [27] Narayan RJ, Aggarwal R, Jin WWC, Monteiro-Riviere NA, Crombez R, Shen W. Mechanical and biological properties of nanoporous carbon membranes. Biomed Mater 2008;3:034107. [28] Garrett DJ, Ganesan K, Stacey A, Fox K, Mefin H, Prawer S. Ultra-nanocrystalline diamond electrodes: optimisation for neural stimulation. J Neural Eng 2012;9:016002. [29] Fox K, Palamara J, Judge R, Greentree AD. Diamond as a scaffold for bone growth. J Mater Sci – Mater Med 2013;24:849–61. [30] Booth L. Synthesis and characterization of multilayered diamond coatings for biomedical implants. Materials 2011;4:857–68.
CARBON
7 5 (2 0 1 4) 4 5 2–46 4
[31] Irifune T, Kurio A, Sakamoto S, Inoue T, Sumiya H. Ultrahard polycrystalline diamond from graphite. Nature 2003;421:599–600. [32] Gracio JJ, Fan QH, Madaleno JC. Diamond growth by chemical vapour deposition. J Phys D: Appl Phys 2010;43:374017. [33] Koizumi S, Nebel CE, Nesla´dek M. Physics and applications of CVD diamond. Weinheim, Germany: Wiley-VCH Verlag GmbH & Co. KGaA; 2008. [34] Szunerits S, Boukherroub R. Different strategies for functionalization of diamond surfaces. J Solid State Electrochem 2008;12:1205–18. [35] Krueger A. The structure and reactivity of nanoscale diamond. J Mater Chem 2008;18:1485–92. [36] Tiainen VM. Amorphous carbon as a bio-mechanical coating – mechanical properties and biological applications. Diamond Relat Mater 2001;10:153–60. [37] Hauert R, Thorwarth K, Thorwarth G. An overview on diamond-like carbon coatings in medical applications. Surf Coat Technol 2013;233:119–30. [38] Roy RK, Lee KR. Biomedical applications of diamond-like carbon coatings: a review. J Biomed Mater Res B 2007;83:72–84. [39] Robertson J. Diamond-like amorphous carbon. Mater Sci Eng R 2002;37:129–281. [40] Aharonovich I, Lee JC, Magyar AP, Buckley BB, Yale CG, Awschalom DD, et al. Homoepitaxial growth of single crystal diamond membranes for quantum information processing. Adv Mater 2012;24:OP54–>OP59. [41] Burek MJ, de Leon NP, Shields BJ, Hausmann BJM, Chu Y, Quan Q, et al. Free-standing mechanical and photonic nanostructures in single crystal diamond. Nano Lett 2012;12:6084–9. [42] Masuda H, Watanabe M, Yasui K, Tryk D, Rao T, Fujishima A. Fabrication of a nanostructured diamond honeycomb films. Adv Mater 2000;12:444–7. [43] Mammana VP, Salvadoria MC, Brown IG. Study of porosity in permeable diamond membranes. Thin Solid Films 1997;308– 309:258–62. [44] Wang X, Ocola LE, Divan RS, Sumant AV. Nanopatterning of ultrananocrystalline diamond nanowires. Nanotechnology 2012;23:075301. [45] Fox NA, Youh MJ, Steeds JW, Wang WN. Patterned diamond particle films. J Appl Phys 2000;87:8187–91. [46] Ando Y, Kuwabarra J, Suzuki K, Sawabe A. Patterned growth of heteroepitaxial diamond. Diamond Relat Mater 2004;13:1975–9. [47] Kurdyukov DA, Feoktistov NA, Nashchekin AV, Zadiranov YM, Aleksenskii AE, Vul’ AY, et al. Ordered porous diamond films fabricated by colloidal crystal templating. Nanotechnology 2012;23:015601. [48] Karan S, Samitsu S, Peng X, Kurashima K, Ichinose I. Ultrafast viscous permeation of organic solvents through diamondlike carbon nanosheets. Science 2012;335:444–7. [49] Zakhidov AA, Baughman RH, Iqbal Z, Cui C, Khayrullin I, Dantas SO, et al. Carbon structures with three-dimensional periodicity at optical wavelengths. Science 1998;282:897–901. [50] Ralchenko VG, Sovyk DN, Bolshakov AP, Homich AA, Vlasov II, Kurdyukov DA, et al. Diamond direct and inverse opal matrices produced by chemical vapor deposition. Phys Solid State 2011;53:1131–4. [51] Sovyk DN, Ralchenko VG, Kurdyukov DA, Grudinkin SA, Golubev VG, Khomich AA, et al. Photonic crystals of diamond spheres with the opal structure. Phys Solid State 2013;55:1120–3. [52] Masuda H, Yanagishita T, Yasui K, Nishio K, Yagi I, Rao TN, et al. Synthesis of well-aligned diamond nanocylinders. Adv Mater 2001;13:247–9. [53] Li L, Zhang K, Qiu J, Wang S, Van HH, Zhang M. Solution assembly of vertically aligned diamond nanotube arrays from diamond nanocrystals. Diamond Relat Mater 2012;29:79–83.
463
[54] Hees J, Kriele A, Williams OA. Electrostatic self-assembly of diamond nanoparticles. Chem Phys Lett 2011;509:12–5. [55] Cervenka J, Lau DWM, Dontschuk N, Shimoni O, Silvestri L, Ladouceur F, et al. Nucleation and chemical vapor deposition growth of polycrystalline diamond on aluminum nitride: role of surface termination and polarity. Cryst Growth Des 2013;13:3490–7. [56] Fairchild BA, Olivero P, Rubanov S, Greentree AD, Waldemann F, Taylor RA, et al. Fabrication of ultrathin single-crystal diamond membranes. Adv Mater 2008;20:4793–8. [57] Rezek B, Ukraintsev E, Kromka A, Ledinsky´ M, Brozˇ A, Noskova´ L, et al. Assembly of osteoblast cell micro-arrays on diamond guided by protein pre-adsorption. Diamond Relat Mater 2010;19:153–7. [58] Auciello O, Sumant AV. Status of the science and technology of ultrananocrystalline diamond (UNCD) films and applications to multifunctional devices. Diamond Relat Mater 2010;19:699–718. [59] Hemmerich KJ. General aging theory and simplified protocol for accelerated aging of medical devices. Med Plastics Biomater 1998;4:16–24. [60] Ferrari AC, Robertson J. Raman spectroscopy of amorphous, nanostructured, diamond-like carbon, and nanodiamond. Phil Trans R Soc Lond A 2004;362:2477–512. [61] Li F, Zhang L, Metzger RM. On the growth of highly ordered pores in anodized aluminum oxide. Chem Mater 1998;10:2470–80. [62] Alexander MR, Beamson G, Bailey P, Noakes TQ, Skeldon P, Thompson GE. The distribution of hydroxyl ions at the surface of anodic alumina. Surf Interface Anal 2003;35:649–57. [63] Shih TS, Chen PC, Huang YS. Effects of the hydrogen content on the development of anodic aluminum oxide film on pure aluminum. Thin Solid Films 2011;519:7817–25. [64] Zou YS, Wang QM, Du H, Song GH, Xiao JQ, Gong J, et al. Structural characterization of nitrogen doped diamond-like carbon films deposited by arc ion plating. Appl Surf Sci 2005;241:295–302. [65] Rosenberger L, Baird R, McCullen E, Auner G, Shreve G. XPS analysis of aluminum nitride films deposited by plasma source molecular beam epitaxy. Surf Interface Anal 2008;40:1254–61. [66] Abbas GA, Roy SS, Papakonstantinou P, McLaughlin JA. Structural investigation and gas barrier performance of diamond-like carbon based films on polymer substrates. Carbon 2005;43:303–9. [67] Osswald S, Yushin G, Mochalin V, Kucheyev SO, Gogotsi Y. Control of sp2/sp3 carbon ratio and surface chemistry of nanodiamond powders by selective oxidation in air. J Am Chem Soc 2006;128:11635–42. [68] Lenardi C, Piseri P, Briois V, Bottani CE, Li Bassi A, Milani P. Near-edge X-ray absorption fine structure and raman characterization of amorphous and nanostructured carbon films. J Appl Phys 1999;85:7159–67. [69] Nguyen TP, Ip J, Jolinat P, Destruel P. XPS and sputtering study of the Alq3/electrode interface in organic light emitting diodes. Appl Surf Sci 2001;172:75–83. [70] Paglia G, Rohl AL, Buckley CE, Gale JD. Determination of the structure of c-alumina from interatomic potential and firstprinciples calculations: the requirement of significant numbers of nonspinel positions to achieve an accurate structural model. Phys Rev B 2005;71:224115. [71] Liza´rraga R, Holmstro¨m E, Parker SC, Arrouvel C. Structural characterization of amorphous alumina and its polymorphs from first-principles XPS and NMR calculations. Phys Rev B 2011;83:094201. [72] Carrier X, Marceau E, Lambert JF, Che M. Transformations of c-alumina in aqueous suspensions. J Colloid Interface Sci 2007;308:429–37.
464
CARBON
7 5 ( 2 0 1 4 ) 4 5 2 –4 6 4
[73] Yuan Q, Yin AX, Luo C, Sun LD, Zhang YW, Duan WT, et al. Facile synthesis for ordered mesoporous c-aluminas with high thermal stability. J Am Chem Soc 2008;130:3465–72. [74] Boumaza A, Favaro F, Le´dion J, Sattonnay G, Brubach JB, Berthet P, et al. Transition alumina phases induced by heat treatment of boehmite: an X-ray diffraction and infrared spectroscopy study. J Solid State Chem 2009;182:1171–6. [75] Dai D, Wang A. Deposition and properties of Al-containing diamond-like carbon films by a hybrid ion sources. J Alloys Compd 2011;509:4626–31.
View publication stats
[76] Ostrikov K, Neyts EC, Meyyappan M. Plasma nanoscience: from nano-solids in plasmas to nano-plasmas in solids. Adv Phys 2013;62:113–224. [77] Lifshitz Y, Ko¨hler T, Frauenheim T, Guzmann I, Hoffman A, Zhang RQ, et al. The mechanism of diamond nucleation from energetic species. Science 2002;297:1531–3. [78] Orwa JO, Peng JL, McCallum JC, Jamieson DN, Rubanov S, Prawer S. Microstructure evolution in carbon-ion implanted sapphire. J Appl Phys 2010;107:023508.
