Effect of bias enhanced nucleation on the nucleation density of diamond in microwave plasma CVD

The variation of diamond nucleation density as a function of the conditions of bias enhanced nucleation (BEN) were studied. The nucleation density increased with microwave power, but decreased with the substrate temperature. The nucleation density also increased with bias voltage above 60 V, and had a maximum around 100 V. The crystal growth of diamond took place when either the bias voltage was high or the deposition time was long. The shift of C_1s energy measured by X-ray photoelectron spectroscopy indicated that the ratio of carbon sp³ bonds in the amorphous carbon and/or SiC phases formed before the nucleation of diamond, increased around the bias voltage of 100 V, which seemed to be the reason for enhancement of diamond nucleation by bias voltage. A simple computer simulation was performed in order to understand the effect of BEN conditions on the nucleation of diamond. The simulation reproduced the experimentally observed changes of nucleation density and particle size.     

The orientation effect of silicon grains on diamond deposition

Diamond deposition on mirror-polished polycrystalline silicon substrates which have grains in various orientations has been investigated using electron backscatter diffraction (EBSD) method with scanning electron microscopy (SEM). Diamond was deposited by microwave plasma chemical vapor deposition with application of a negative bias voltage on the substrate. The evidence from systematic SEM observations shows that silicon orientation determined by EBSD has a strong effect on diamond nucleation. In general, the diamond nucleation density on Si grains oriented close to <100> is the highest, while it is the lowest for those grains close to <111>, under the same experimental conditions for deposition. The same phenomena have been observed in the range of methane concentration from 2% to 4% in hydrogen.     

Thickness of a modified surface layer formed in a silsesquioxane-based low-k material during etching in a fluorocarbon plasma

This paper explains the origin of our previous observation that, when a silsesquioxane-based low-k film is etched in fluorocarbon plasmas, the thickness of a surface modified layer, in which cage-like Si-O bonds are dissociated to extents greater than a specified level, changes linearly with log[F]²/[CF₂], where [F] and [CF₂] denote concentrations of F and CF₂ radicals in the bulk plasma. During the etching process, the substrate consists of three distinct layers: a fluorocarbon layer, a modified surface layer, and an unmodified layer. F density at the interface between the fluorocarbon and the modified surface layers, denoted as F₀ in this study, is determined in proportional to [F]²/[CF₂], and the density decreases exponentially with the film depth. As a result, the thickness of the modified surface layer changes in proportion to a parameter, log[F]²/[CF₂].     

Enhanced performance of HFCVD nanocrystalline diamond self-mated tribosystems by plasma pretreatments on silicon nitride substrates

Nanocrystalline diamond (NCD) coatings were grown by the hot-filament chemical vapour deposition (HFCVD) method on hydrogen plasma pretreated silicon nitride (Si₃N₄) substrates. The friction and wear behaviour of self-mated NCD films, submitted to unlubricated sliding and high applied loads (up to 90 N), was assessed using an oscillating ball-on-flat configuration in ambient atmosphere. The reciprocating tests revealed an initially high friction coefficient peak, associated to the starting surface roughness of NCD coatings (R_q = 50 nm). Subsequently, a steady-state regime with low friction coefficient values (0.01–0.04) sets in, related to a smoother (R_q = 17 nm) tribologically modified surface. A polishing wear mechanism governing the material loss was responsible for mild wear coefficients (k  10^− 7 mm³ N^− 1 m^− 1). The hydrogen etching procedure notably increased the film adhesion with respect to untreated surfaces as demonstrated by the high threshold loads (60 N; 3.5 GPa) prior to film delamination.     

Fabrication of submicron scale vertically aligned diamond rods by mask-free oxygen plasma etching

A mask-free plasma etching process is described to fabricate 6 μm long submicron diamond rods (SDRs) in long conical shape. Polished polycrystalline diamond is etched in oxygen plasma ignited at a pressure of 10 mTorr by radio-frequency power of 100 W at 13.56 MHz. Each SDR is a bi-crystal, consisting of two diamond crystallites of micron size. The SDR is coated with a Fe₂O₃ layer, as characterized by Auger electron spectroscopy, X-ray photoemission microscopy, and transmission electron microscopy. We propose that a “self-forming” mask of Fe₂O₃ is generated during the etching process in which iron atoms sputtered from the substrate holder are deposited and oxidized on the diamond surface forming “micromask” that protects the underlying diamond and promotes the formation of SDRs.     

Influence of silicon carbide interlayer evolution on diamond heteroepitaxy during bias enhanced nucleation on silicon substrates

In order to improve the crystalline quality of diamond films produced by microwave plasma assisted chemical vapour deposition (MPCVD), the structural evolution of the silicon carbide interlayer during the bias nucleation step has been investigated by reflection high energy electron diffraction (RHEED). Here we highlight the fact that the carbonisation pre-treatment induces a strong extension of the silicon carbide lattice in the direction perpendicular to the surface. This extension gives a lattice constant close to that of silicon. Then, during bias enhanced nucleation, the carbide lattice relaxes. At the same time, this modification is accompanied by an increase of the surface roughness and by a progressive polar misorientation of the silicon carbide. All these transformations could be responsible for the observed drop of the diamond epitaxial ratio when the duration of the bias step is extended. Finally, we found that a lower methane concentration in the plasma slows down this carbide transformation, allowing us to obtain a promising 37% epitaxial ratio.     

Influence of surface etching pretreatment on PEO process of eutectic Al-Si alloy☆

To solve the problems generally encountered during the plasma electrolytic oxidation (PEO) of Al alloys with high Si content, a pretreatment of chemical etching was applied before the process. The influence of such pre-treatment was studied by SEM, EDS and XRD. The pretreatment presents a significant effect on positive voltage at the beginning stage of PEO, leading to higher voltage over the whole process. The difference between the pos-itive voltages of non-etched and etched specimens decreases gradual y with the increase of processing time. The pretreatment exhibits much less influence on the negative voltage. For the sample with surface pretreatment, the average growth rate of PEO coating is increased from 0.50 to 0.84μm·min?1 and the energy consumption is de-creased from 6.30 to 4.36 kW·h·μm?1·m?2. At the same time, both mullite and amorphous SiO2 contents are decreased in the coating.     

Comparative study of diamond films grown on silicon substrate using microwave plasma chemical vapor deposition and hot-filament chemical vapor deposition technique

Diamond films on the p-type Si(111) and p-type(100) substrates were prepared by microwave plasma chemical vapor deposition (MWCVD) and hot-filament chemical vapor deposition (HFCVD) by using a mixture of methane CH₄ and hydrogen H₂ as gas feed. The structure and composition of the films have been investigated by X-ray Diffraction, Raman Spectroscopy and Scanning Electron Microscopy methods. A high quality diamond crystalline structure of the obtained films by using HFCVD method was confirmed by clear XRD-pattern. SEM images show that the prepared films are poly crystalline diamond films consisting of diamond single crystallites (111)-orientation perpendicular to the substrate. Diamond films grown on silicon substrates by using HFCVD show good quality diamond and fewer non-diamond components.     

F2, H2O, and O2 etching rates of diamond and the effects of F2, HF and H2O on the molecular O2 etching of (110) diamond

Oxidation kinetics of natural (110) diamond by oxygen and water were investigated using in situ Fizeau interferometry. Apparent activation energies of 53 and 26 kcal mol⁻¹ were obtained for the etching of (110) type Ia diamond by O₂ and H₂O respectively. The etch rate was found to follow second-order kinetics with respect to O₂ pressure in the pressure range 0.04–10 Torr. For water over the vapour pressure range 0.1–2 Torr, the reaction has a reaction order near unity. The diamond (110) surface was impervious to etching by molecular fluorine at all temperatures up to 1300 °C. Fluorine, hydrogen fluoride and water were found to inhibit the molecular oxygen etching of diamond. Below 900 °C, oxidation is inhibited by the addition of F₂ and HF presumably by blocking reactive sites on the diamond surface through formation of C---F bonds. Above 900 °C, the fluorine is thought to desorb from the diamond (110) surface, rendering the surface susceptible to further oxidation. Addition of water below 800 °C was found to retard etching by molecular oxygen. This is attributed to the formation of C---OH bonds, analogous to C---F.     

A unified global self-consistent model of a capacitively and inductively coupled plasma etching system

Based on the concept of independent control of ion flux and ion-bombardment energy, a global selfconsistent model was proposed for etching in a high-density plasma reactor. This model takes account of the effect on the plasma behavior of separate rf chuck power in an Inductively Coupled Plasma etching system. Model predictions showed that the chuck power controls the ion bombardment energy but also slightly increases the ion density entering the sheath layer, resulting in an increase in etch rate (or etch yield) with increasing this rf chuck power. The contribution of the capacitive discharge to total ion flux in the ICP etching process is less than about 6% at rf chuck powers lower than 250 W. As a model system, etching of InN was investigated. The etch yield increased monotonically with increasing the rf chuck power, and was substantially affected by the ICP source power and pressure. The ion flux increased monotonically with increasing the source power, while the dc-bias voltage showed the reverse trend.     

Influence of substrate temperature on formation of ultrananocrystalline diamond films deposited by HFCVD argon-rich gas mixture

The influence of the substrate temperature on the formation of ultrananocrystalline diamond (UNCD) thin films, prepared by an argon-based hot filament chemical vapor deposition (HFCVD), is discussed in this work. The gas mixture used for diamond growth was 1 vol.% methane, 9 vol.% hydrogen and 90 vol.% argon at a total flow rate of 200 sccm and at a total pressure of 30 Torr. The substrate temperature range was from 550 to 850 °C at deposition time of 8 h. Mass growth rate was determined at different deposition temperatures. The activation energy for UNCD growth, determined from the Arrhenius plot, was lower (5.7 kcal/mol) than the values found for standard diamond deposition (around 11 kcal/mol). In this work, we suggest that the activation energy was lower because the growth of these films occurs at conditions that there is a high growth competition between diamond phase and sp² phases. To support this hypothesis, systematic characterization studies based on Raman scattering spectroscopy, high-resolution X-ray diffractometry and high-resolution scanning electron microscopy were performed.     

Surface study of iridium buffer layers during the diamond bias enhanced nucleation in a HFCVD reactor

The BEN nucleation of diamond on iridium substrates has been studied in a hot filament reactor. Without a prior BEN stage, no diamond nucleation could be detected. Nucleation is promoted only if a BEN step is applied before the CVD growth with nucleation densities up to 5 10⁹ cm⁻². The present study focuses on the early stages of BEN to better understand its specific role. In this way, samples have been in situ characterized using electron spectroscopies (XPS, AES, ELS) and further investigated by HR-SEM, AFM, Nano-Auger and Raman spectroscopy. A very different behaviour in the interface formation has been observed, as compared to silicon. First, a substrate roughening takes place during the cleaning step. Second, the formation of a graphite layer was systematically observed, with or without the BEN stage, in the early stages of CVD synthesis. Its crystallinity has been studied from the Raman experiments. The study of the XPS Ir 4f peaks supports a weak chemical bonding between graphite and iridium. Finally, after the BEN stage, spatially resolved Nano-Auger and Raman measurements revealed the presence of diamond nanocrystals.     

Surface roughness of AlN films deposited on negatively biased silicon and diamond substrates

Our previous studies on AlN microstructures have shown that smooth amorphous films (a-AlN) can be grown on negatively biased Si substrates by the versatile physical vapour deposition technique under reactive magnetron sputtering. These a-AlN films are produced by energetic Ar ion bombardment under negative bias whereas those grown without bias were columnar crystallized ones (c-AlN). Here, we show first that depositing an a-AlN layer on c-AlN/Si structures by switching a suitable bias to the Si substrate can efficiently reduce their surface roughness. We then extend this smoothening method to a c-AlN/Poly-crystallized diamond (PCD) structure to reduce its high surface roughness that hampers using such structures in SAW device design. In fact, the piezoelectric c-AlN surfaces grown on rough diamond surfaces are equally rough. Effectively, the a-AlN layer deposited on the c-AlN/PCD structure brings down the latter's RMS surface roughness to one tenth of its initial RMS roughness, as confirmed here by TEM and AFM observations. The insulating property of the diamond as biased substrate doesn't impede the growth of this a-AlN layer. This smoothening method is without process interruption, where simply a negative bias is switched on to the diamond substrate once the desired piezoelectric c-AlN film thickness as monitored here by in-situ reflectometry, is attained. This as-grown smoothening method can be therefore easily and rapidly implemented and can thus replace time-consuming and costly PCD ionic and/or mechanical polishing. Hopefully, the method can be advantageously applied to c-AlN/nano-crystallized diamond structures (NCD) where the NCD films are not prepared under rigorous conditions meant to minimize their surface roughness.     

Characterization of surface modified carbon nanoparticles by low temperature plasma treatment

The carbon nanoparticles derived from polypyrrole were treated with oxygen and ammonia radio-frequency plasma. Micro-attenuated total reflection Fourier transform infrared, element analysis, and X-ray photoelectron spectroscopy analyses were performed to confirm the incorporation of polar functional groups onto the carbon nanoparticles. Moreover, the morphology of plasma treated carbon nanoparticles was retained without nanoparticle aggregation. The plasma treated carbon nanoparticles exhibited the enhanced dispersibility in aqueous solution, compared to the pristine carbon nanoparticles.     

Inductively coupled plasma etching of Ta,Co, Fe,NiFe, NiFeCo,and MnNi with Cl<Subscript>2</Subscript>/Ar discharges

Dry etching of the magnetic thin films such as Ta, Fe, Co, NiFe, NiFeCo, and MnNi was carried out in inductively coupled plasmas of Cl₂/Ar mixture. All the magnetic materials went through a maximum etch rate at 25% Cl₂. The effects of the ICP source power and the rf chuck power on the etch rate and the surface roughness were quite dependent of the materials. An ion-enhanced chemical etch mechanism was important for the magnetic films. The surface roughness of the etched samples was relatively constant of the rf chuck power up to 200 W, but a rougher surface at a higher rf power was obtained. Post-etch cleaning of the etched samples in de-ionized water reduced the chlorine residues substantially.     

Effects of TiMoTa nano-crystalline interlayer on nucleation,adhesion and tribological behaviors of diamond coating

To investigate the influence of the transition layer on diamond nucleation and growth, the TiMoTa multi-alloy interlayers were firstly prepared on WC-6%Co cemented carbide by double glow plasma surface alloying technique, and diamond was then deposited using microwave plasma chemical vapor deposition system. The thickness of the TiMoTa interlayers increased from 1.2 μm to 2.7 μm with the deposition temperature increasing from 800 to 900 °C. Due to the formation of nano-crystalline structure and hard phase, the as-prepared TiMoTa interlayer exhibited high micro-hardness. In the diamond deposition process, the nano-carbide particles preferentially formed at the TiMoTa interlayers can effectively prevent the further diffusion of C, so the surface carbon concentration rapidly accumulates to the critical value required for the nucleation and growth of diamond microcrystals. Finally, a microscale wear-resistant diamond coating with good adhesion was grown on the transition layer, the crack propagation radius of the diamond coating is ～162 μm, and can reach Hf 2–3 grade. Therefore, our prepared TiMoTa nano-crystalline interlayer provides a new path for the development of a high-quality diamond coating with good adhesion on cemented carbide.     

Field emission of polycrystalline diamond films grown by microwave plasma chemical vapor deposition. II. Effect of p-type doping in diamond

The effects of boron (B) doping on the field emission (FE) of diamond films grown by a microwave plasma chemical vapor deposition technique were studied. Raman scattering spectroscopic analysis revealed that B-doping significantly suppressed formation of non-diamond components in the diamond film. The B-doped p-type diamond films had low resistivity, ranging from 0.07 to 20 Ω cm, and various volume fractions of non-diamond components in the diamond films. The turn-on electric field, F_T, was independent of the resistivity, the film thickness, and the volume fraction of the non-diamond components. The lowest F_T value of 8 V μm⁻¹ and the highest emission current of 3×10⁻² A cm⁻² were obtained in the B-doped diamond films. The high efficiency of the electron emission in the B-doped diamond films was believed to be due to the increase in volume fraction of the conductive regions in the film and the high density of emission sites on the film surface.     

Deposition of diamond film on silicon and quartz substrates located near DC plasma

Deposition of diamond films on to both Si and quartz substrates was succeeded by means of locating the substrate near plasma, and their microstructures were investigated by using SEM and Raman spectroscopy. The plasma was generated by intermittent DC discharge in H₂–CH₄ gas mixture at high gas pressure. The deposition rate of the film was remarkably increased when distance (D) between the substrate and the plasma (discharge electrodes) decreased. When the gas pressure (P_g) was increased from 100 to 250 Torr, the deposition rate was extremely increased and the crystalline quality of the film was improved. The deposition rate, when P_g = 200 Torr and D = 5 mm, was about 2.1 and 1.7 μm/h for Si and quartz substrate, respectively.     

Influence of the substrate type on CVD grown homoepitaxial diamond layer quality by cross sectional TEM and CL analysis

To assess diamond-based semiconducting devices, a reduction of point defect levels and an accurate control of doping are required as well as the control of layer thickness. Among the analyses required to improve such parameters, cross sectional studies should take importance in the near future. The present contribution shows how FIB (focused ion beam) preparations followed by electron microscopy related techniques as TEM or CL allowed to perform analysis versus depth in the layer, doping and point defect levels. Three samples grown along the same week in the same machine with identical growth conditions but on different substrates (CVD-IIIa (110) oriented, CVD-optical grade (100) oriented and a HPHT-Ib (100) oriented) are studied. Even though A-band is observed by CL, no dislocation is observed by CTEM. Point defect type and level are shown to substantially change with respect to the substrate type as well as the boron doping levels that vary within an order of magnitude. H3 present in the epilayer grown on HPHT type of substrate is replaced by T1 and NE3 point defects for epilayers grown on the CVD type one. An increase of excitonic transitions through LO phonons is also shown to take place near the surface while only TO ones are detected deeper in the epilayer. Such results highlight the importance of choosing the correct substrate.