Silicon (100) substrates are ubiquitous in microfabrication and, accordingly, their surface characteristics are important. Herein, we report the analysis of Si (100) via X-ray photoelectron spectroscopy (XPS) using monochromatic Al Kα radiation. Survey scans show that the material is primarily silicon and oxygen with small amounts of carbon, nitrogen, and fluorine contamination. The Si 2p region shows two peaks that correspond to elemental silicon and silicon dioxide. Using these peaks the thickness of the native oxide (SiO2) is estimated using the equation of Strohmeier. The oxygen peak is symmetric. These silicon wafers are used as the substrate for subsequent growth of templated carbon nanotubes in the preparation of microfabricated thin layer chromatography plates.

We apply a suite of analytical tools to characterize materials created in the production of microfabricated thin layer chromatography plates. Techniques used include X-ray photoelectron spectroscopy (XPS), valence band spectroscopy, time-of-flight secondary ion mass spectrometry (ToF-SIMS) in both positive and negative ion modes, Rutherford backscattering spectroscopy (RBS), and helium ion microscopy. Materials characterized include: the Si(100) substrate with native oxide: Si/SiO 2 , alumina (35 nm) deposited as a diffusion barrier on the Si/SiO 2 : Si/SiO 2 /Al 2 O 3 , iron (6 nm) thermally evaporated on the Al 2 O 3 : Si/SiO 2 /Al 2 O 3 /Fe, the iron film annealed in H 2 to make Fe catalyst nanoparticles: Si/SiO 2 /Al 2 O 3 /Fe(NP), and carbon nanotubes (CNTs) grown from the Fe nanoparticles: Si/SiO 2 /Al 2 O 3 /Fe(NP)/CNT. The Fe films and nanoparticles appear in an oxidized state. Some of the analyses of the CNTs/CNT forests appear to be unique: (i) the CNT forest appears to exhibit an interesting 'channeling' phenomenon by RBS, (ii) we observe an odd-even effect in the SIMS spectra of C n -species for n = 1 -6, with the n ≥ 6 ions showing a steady decrease in intensity, and (iii) valence band characterization of CNTs using X-radiation is reported. Initial analysis of the CNT forest by XPS shows that it is 100 at.% carbon. After one year, only ca. 0.25 at.% oxygen is observed. The information obtained from the combination of the different analytical tools provides a more complete understanding of our materials than a single technique, which is analogous to the story of 'The Blind Men and the Elephant'. The raw XPS and ToF-SIMS spectra from this study will be submitted to Surface Science Spectra for archiving.

No abstract

The authors report the ozonation of patterned, vertically aligned carbon nanotube (CNT) forests as a method of priming them for subsequent pseudo atomic layer deposition (ψ-ALD) (alternating layer deposition) of silica to produce microfabricated, CNT-templated thin layer chromatography (TLC) plates. Gas phase ozonation simplifies our deposition scheme by replacing two steps in our previous fabrication process: chemical vapor deposition of carbon and ALD of Al2O3, with this much more straightforward priming step. As shown by x-ray photoelectron spectroscopy (XPS), ozonation appears to prime/increase the number of nucleation sites on the CNTs by oxidizing them, thereby facilitating conformal growth of silica by ψ-ALD, where some form of priming appears to be necessary for this growth. (As shown previously, ψ-ALD of SiO2 onto unprimed CNTs is ineffective and leads to poor quality depositions.) In conjunction with a discussion of the challenges of good peak fitting of complex C 1s XPS narrow scans, the authors present an analysis of their C 1s data that suggests an increase in oxidized carbon, particularly the C=O group, with increasing oxygen content of the CNT forests. After coating with SiO2, the CNTs are removed by elevated temperature air oxidation, the SiO2 is rehydrated, and the plates are coated with 3-aminopropyltriethoxysilane (APTES). The resulting APTES-coated plates separate various fluorescent dyes giving results that are generally at least as good as those the authors reported previously with their more complicated fabrication/priming scheme. TLC plates with different geometries are microfabricated, where plates with narrower channels show longer run times (lower mobile phase velocities) and plates with narrower features appear to give higher efficiencies.

No abstract

The effects of iron catalyst thickness on the fabrication and performance of microfabricated, binder-free, carbon nanotube (CNT)-templated, thin layer chromatography (TLC) plates are demonstrated. The iron catalyst was deposited at thicknesses ranging from 4 to 18 nm in increments of 2 nm. Its thickness plays a key role in governing the integrity and separation capabilities of microfabricated TLC plates, as determined using a test dye mixture. Atomic force microscopy and scanning electron microscopy show that smaller and more numerous catalyst nanoparticles are formed from thinner Fe layers, which in turn govern the diameters and densities of the CNTs. The average diameter of the Fe nanoparticles, Dp, is approximately six times the initial Fe film thickness, tFe: Dp ≈ 6tFe. After deposition of relatively thick silicon layers on CNTs made with different Fe thicknesses, followed by oxidation, all of the resulting CNT-templated SiO2 wires had nearly the same diameter. Consequently, their surface areas were very similar, although their areal densities on the TLC plates were not because thinner catalyst layers produce denser CNT forests. For tFe = 6 nm, nanotube growth appears to be base growth, not tip growth. Best TLC separations of a test dye mixture were obtained with plates prepared with 6 or 4 nm of catalyst. Calculations suggest a loss of surface area for TLC plates made with thicker Fe layers as a result of fewer, thicker CNTs, where the density of silica nanotubes (device surface area) goes approximately as 1/tFe2. While the focus of this paper is toward a greater understanding of the processing conditions that lead to the best TLC plates, a baseline separation of three analgesics (caffeine, phenacetine, and propyphenazone) is shown on a normal phase TLC plate grown with 6 nm of iron.

No abstract

We report the XPS characterization of a thin film of Al2O3 (35 nm) deposited via e-beam evaporation onto silicon (100). The film was characterized with monochromatic Al Kα radiation. An XPS survey scan, an Al 2p narrow scan, an O 1s narrow scan, and the valence band spectrum were collected. The Al2O3 thin film is used as a diffusion barrier layer for templated carbon nanotube (CNT) growth in the preparation of microfabricated thin layer chromatography plates.