ICMCTF1999 Session B2: CVD, MOCVD and Related Techniques

Wednesday, April 14, 1999 3:30 PM in Room Golden West

Wednesday Afternoon

Time Period WeA Sessions | Abstract Timeline | Topic B Sessions | Time Periods | Topics | ICMCTF1999 Schedule

Start Invited? Item
3:30 PM B2-7 Recent Advances in CVI for Refractory Composites
T.M. Besmann (Oak Ridge National Laboratory); K.J. Probst (University of Florida); D.P. Stinton, R.A. Lowden (Oak Ridge National Laboratory); T.J. Anderson (University of Florida)

Chemical vapor infiltration (CVI) is simply chemical vapor deposition (CVD) on the internal surfaces of a porous preform and has been used to produce a variety of developmental and application materials. The greatest use of CVI is to infiltrate continuous filament preforms, taking advantage of the relatively low stress CVD process.

Chemical vapor infiltration originated in efforts to densify porous graphite bodies by infiltration with carbon. The technique has developed commercially such that half of the carbon/carbon composites currently produced are made by CVI (the remainder are fabricated by curing polymer impregnated fiber layups). The earliest report of CVI for ceramics was a 1964 patent for infiltrating fibrous alumina with chromium carbides.

In CVI, reactants are introduced in the porous preform via either diffusion or forced convection, and the CVD precursors deposit the appropriate phase(s). As infiltration proceeds, the deposit on the internal surfaces becomes thicker. Thus after some length of time the growing surfaces meet, bonding the preform and filling much of the free volume with deposited matrix.

The major advantage CVI has over competing densification processes is the low thermal and mechanical stress to which the relatively sensitive fibers are subjected. CVI can occur at temperatures much more moderate than the melting point of the deposit, and therefore usually well below the sintering temperature. In addition, the process imparts little mechanical stress to the preform as compared to more traditional techniques such as hot-pressing.

The most widely used commercial process is isothermal/isobaric CVI (ICVI), which depends only on diffusion for species transport. It generally operates at reduced pressure (1-10 kPa) for deposition rate control. This diffusion-dependent process is slow, requiring several-week-long infiltration times. It is commercially attractive, however, because large numbers of parts of varying dimensions are easily accommodated in a single reactor.

The forced-flow/thermal-gradient technique (FCVI) developed at Oak Ridge National Laboratory (ORNL) overcomes the problems of slow diffusion and restricted permeability, and has demonstrated a capability to produce thick-walled, simple-shaped SiC-matrix components in times of the order of hours. It is the current state of development of this process based on recent work at ORNL that is the subject of the presentation.

4:10 PM B2-9 Characterization and Properties of MT-CVD TiCN and MT-CVD ZrCN Coatings
S. Kudapa, K. Narasimhan, P. Boppana, B. Russell (Valenite Inc.)

Medium temperature CVD (MT-CVD) coatings on cemented carbide cutting tools have gained prominence in the machining of irons, stainless steels and other abrasive work-piece materials that require increased edge toughness of the cutting tools. Although MT-CVD TiCN is the most common coating layer used in the current popular cutting tools, other coatings like MT-CVD ZrCN used in conjunction, can offer potential performance improvements in machining.

Properties of mono- and multi-layer hard coatings of TiCN and ZrCN deposited on carbide substrates by MT-CVD will be discussed in this paper. Evaluation of residual stresses, texture of coating, along with interfacial features of these hard coatings will be characterized by XRD, SEM, and optical microscopy techniques. Failure mechanism modes of these various coating designs, involved in machining of ductile iron and 4340 steels will be discussed to delineate the wear properties of the coatings.

4:30 PM B2-10 Wedge-shaped Ceramic Membranes for Gas Sensor Applications Produced by a Variety of CVD Techniques
M. Frietsch, L.T. Dimitrakopoulos, T. Schneider (Forschungszentrum Karlsruhe GmbH, Germany); J. Goschnick (Forschungszentrum Karlsruhe, Germany)

In order to realize an inexpensive and reliable electronic nose a gas sensor microarray was developed which is based on SnO2 or WO3 segmented metal oxide layers. The basic structure of the microarrays is manufactured by HF-magnetron sputtering combined with a shadow masking technique. Rectangular SnO2 or WO3 layers, 4 x 8 mm wide and some 100 nm thick, were deposited on oxidized silicon substrates. Parallel Pt strips serving as electrodes for the conductivity measurement were placed on top of the metal oxide field. Thus the latter is divided into 40 identical sensor elements. The reverse side of the microarray is equipped with 4 Pt meanders to heat the microarray to the operating temperature of 200 - 500°C. In the final production stage the microarray is coated with a ceramic layer of gradually varying thickness by CVD. The thickness variation of this membrane across the microarray differentiates the initially identical sensor elements with respect to their gas response and results in gas characteristic conductivity patterns allowing to distinguish gases also in mixtures.

The ceramic membranes have to be thin enough to be highly gas permeable while providing a selective access of gases to the metal oxide layer. Moreover, they have to be heat-resistant due to the operating temperature of metal oxide gas sensors. Al2O3 and SiO2 coatings were found to be suitable to adjust the selectivity of metal oxide sensors. Different CVD techniques were used to produce coatings of controlled inhomogeneity with a thickness from 1 nm to about 100 nm on gas sensor microarrays. The thermal decomposition of aluminum-tri-isopropylate and phenyl-triethoxy-silane, at the surface of the heated metal oxide (250-450°C) yielded Al2O3 and SiO2 layers, respectively. Thickness gradients of the membranes were either achieved by a temperature gradient applied across the sensor array using the 4 heaters on the reverse side or by directing a molecular beam of the precursor vapor to the metal oxide surface. Furthermore IBAD was used to produce SiO2 membranes using phenyl-triethoxy-silane as the precursor. The ion beam profile was shaped to gradually alter the ion current density across the microarray leading to a laterally different SiO2 deposition rate. The thickness and the chemical composition of the layers, especially in the interface region, were analyzed with SNMS.

4:50 PM B2-11 The PROXI Project: Cost-effective Fabrication of Oxidic Multi-Layer Structures to be Used as Gas Sensors
A. Synowczyk, J. Goschnick (Forschungszentrum Karlsruhe, Germany)

Monitoring gaseous components of the atmosphere has seen an enormously increased demand for the last few years. Especially consumer applications, such as exhaust gas control or air-conditioning of cars, domestic food processing, household appliances and private care diagnostics, require a novel type of gas sensor devices offering high-quality analytical power with long-term reliability at very low cost. These requirements stimulated the development of a gas sensor microarray of thumbnail size at the Forschungszentrum Karlsruhe based on semiconducting metal oxide thin layers.

In a two-year program - the PROXI project - novel, economical production technologies will be developed for oxidicmulti layer systems. These production techniques will be applied to manufacture low-cost mass production gas sensor arrays with a high discrimination power. The special technological novelty of this project is to set up a gas sensor array by simple segmentation of a metal oxid layer using a shadow-mask technique to obtain a high discrimination of the gas sensor array towards a high number of sensor elements in only one process step. The selectivity of the elements is differentiated by the gradient technique. A temperature variation across the array is obtained by operating the 4 rear heaters of the chip with a different power. This provides one way to alter the selectivity of the sensor segments. A thickness gradient of the topmost gas-permeable coating (gradient membrane) serves as an additional tool to continuously change the gas selectivity of the sensor segments across the array. Typically the thickness is changed from 2 to 50 nm. Again only one fabrication step is needed to differentiate the sensors of the array in order to obtain a high discrimination power for gases.

Moreover, for the mass production of the gas sensor microarray, novel PVD and CVD techniques are already under test. For depositing the metal oxide on oxidized 6" Si wafers, a highly effective medium-frequency sputtering cathode, called TWIN-MAG, is used. A broad band ion source with a beam width of 1 m is developed. This allows the formation of gradient membranes on several chips simultaneous with the IBAD (Ion Beam Assisted Chemical Vapor Deposition) technique. An adjustable beam profile permits a precise control of the thickness gradient which should be of high uniformity from chip to chip to obtain a good reproducibility in the gas discrimination power of the sensor chips.

5:10 PM B2-12 Effect of Al Addition on Adhesive Strength of Al2O3 to Base Layer of Ti(C,O)
S. Kiriyama (Toshiba Tungaloy Co.,Ltd., Japan)

As a coating material for cutting tools, Al2O3 has very good properties in chemical stability, welding resistance and high temperature hardness. Therefore Al2O3 layer is indispensable to multilayer coatings by CVD for cutting inserts used for high speed turning. In severe cutting conditions, however, Al2O3 layers often peel away at the beginning of cutting. It is caused by the poor adhesive strength of Al2O3 layer to the base layer. As the base layer, Ti base compounds having B1 structure such as TiC, Ti(C,N), TiN, Ti(C,O) are used in commercial coated inserts.

In this study, an attempt is made to improve the adhesive strength of Al2O3 layer by adding Al to the base layer of Ti(C,O). The results obtained are as follows: 1) In case of low Al content, (Ti,Al)(C,O) single phase layer is obtained. In high Al content, two phase layer containing dispersed Al2O3 is formed. 2) The adhesive strength of Al2O3 layer to the base one is improved with increasing Al content in the base layer. In particular, the excellent adhesive strength is accomplished when Al2O3 disperse in the base layer.

Time Period WeA Sessions | Abstract Timeline | Topic B Sessions | Time Periods | Topics | ICMCTF1999 Schedule