The Electroceramics Group is for the first time conducting research in the field of microwave ceramics.  The project will be divided into studies of bulk ceramics and thin films for wireless communications.  The approach is to study the loss mechanisms of paraelectric ceramics in the range of 0.5 – 40 GHz and to correlate the materials’ properties to extrinsic defects in the materials, and then to relate the data with processing/fabrication techniques.  The underlying idea being to identify the defects that limit the performance of microwave devices and to devise new processing procedures to minimize the defect concentration and reduce, as much as possible, the extrinsic contribution to the loss to obtain high quality microwave dielectrics.

             A useful comparison of the basis for this study can be made with the optical fiber research effort in the late 1970s through the early 1980s.  Pure silica and silica-based glasses exhibited attenuation values several orders of magnitude higher than theory predicted, but with the invention of CVD processes the loss gradually decreased until the fundamental, or intrinsic limit was reached.  This is analogous to the problem at hand with the properties, particularly the high loss, of paraelectric thin films in the microwave regime.

            The problem at hand is that microwave ceramics exhibit losses that are a factor of 10²-10³ times higher than theory would predict.  The figure below shows the relation between the dielectric constant and the loss as a function of frequency.  In the microwave region there are no relaxation or resonance phenomena that contribute to the intrinsic loss, and therefore it must be assumed that the relatively high losses that are observed are due to extrinsic defects in the materials.  The relaxation time constants of these extrinsic mechanisms may occur anywhere in the spectrum depending on the particular defects present in the material.  It is these extrinsic contributions that must be identified.

Systems of Interest

SrTiO₃

            Bulk strontium titanate (ST) in an incipient ferroelectric with a highly nonlinear dielectric response.  The high dielectric constant of the bulk single crystal (i.e. at 4 K in the low gigahertz regime, e_r > 20,000) combined with a low loss of 10^-4 (near intrinsic levels) and the ability to tune the dielectric properties make it a very attractive material for use in tunable microwave components.  In addition its crystal structure and chemical compatibility with YBCO superconductors adds to its potential.  However, in thin film form the dielectric constant drops by about two orders of magnitude due to size effects and, more importantly, the loss increases to around 10^-3 to 10^-2.  The increase in the loss (tan d) is believed to be the result of extrinsic defects in the material. It is these phenomena that must be eliminated to drive these materials into the industrial arena.

 Ba_(1-x)Sr_xTiO₃

                The solid solution of barium titanate and strontium titanate gives the perovskite BST.  This material is ferroelectric and tetragonal at room temperature for x~0.7 and higher.  All other compositions are paraelectric and cubic.  The entire range of stoichiometry has been explored with the most interest on 60/40 and 50/50 Ba:Sr for microwave applications.  The properties of BST in thin film form at room temperature are similar to SrTiO3 at cryogenic temperatures, in that the dielectric constant is on the order of 1000 and the loss tangent about 0.005 in the low GHz region.  The same problem exists with this material in that the loss is affected greatly by extrinsic defects in the material generated during processing.

Thin Film Fabrication

       The thin film research effort will be conducted using a new state-of-the-art Pulsed Laser Deposition (PLD) system.  This system allows the deposition of up to six different target materials without breaking vacuum.  Deposition can be carried out at any temperature up to 950C, under a controlled oxygen atmosphere in the pressure range of 1mTorr to 1Torr.  The ablation of the target material is achieved with a new KrF (248 nm) excimer laser.

The PLD process is simple in design, but complicated in terms of the many variable process parameters.  A basic description of the process is as follows: A laser beam is focused through an optical system into a vacuum chamber where it hits a rotating ceramic target.  The energy of the laser beam is absorbed by the target, resulting in instantaneous vaporization of the target surface.  The ablated material expands in a plume away from the target toward a heated substrate.  As the atomic species, ionic species, and/or molecules of the target travel toward the substrate they react with a background gas of oxygen (for oxide ceramics).  Upon contact with the heated substrate, the various species deposit on the surface, and this material undergoes rearrangement and diffusion to form a thin film of the same composition as the target material.

PLD was chosen as the fabrication method because of several factors.  Of the many other chemical and physical vapor deposition techniques used by industry to make thin films, PLD is one of the least expensive and quickest ways to make high quality thin films for research.  It allows the fabrication of multi-material thin films with relative ease (by changing the target) without the use of hazardous gases or solutions typically needed in processes such as MOCVD and Sol-gel, and provides excellent target stoichiometry transfer (even of complex oxides such as superconductors).  In addition, our group has used PLD for the deposition of films of BST, PMN-PT, and BST/BT multilayers.  The only major disadvantage of PLD is that the area of uniformity of the films will be about 1cm² (without a substrate rotator), which is relatively small in comparison to other thin film deposition techniques.