Materion Corporation logo

Coating Materials News Volume 11 Issue 1


March, 2001

Titanium Dioxide Structural Studies

Titanium dioxide is a commonly used high-index material for constructing bandpass filters, long-and short-wave pass edge filters, and AR coatings for the visible range because it has the highest index of all visible-range oxide compounds. It also finds application in gas sensors, capacitors, and applications requiring blood compatibility.

Titania is relatively easy to evaporate by e-beam, once the correct evaporation parameters are established. Previous CMN issues have provided starting material, deposition parameter guidance, and film behavior [V5, Issue 4, Oct - Dec. 1995; V8, Issue 3, Jul – Sept 1998]. The challenge with titania is to deposit dense layers having low optical absorption. Because of the tendency to grow with a columnar microstructure, titania layers are porous and exhibit mechanical and optical instabilities under varying atmospheric environments. We have discussed these problems, common to many metal oxide compounds, in past issues. In the case of titania, several crystal phases, not all of which are stable, can be present simultaneously in a deposited layer. The mixture and relative concentrations depend on deposition parameters. The integrity of titania layers has been demonstrated with films sputtered using the many techniques and variations of the sputtering process and with ion-based processes. The control of titania crystal structure is important because each of the four possible phases possesses different properties.

The structural phases possible in a titania layer are: amorphous, the metastable crystalline forms brookite and anatase, and the high temperature stable phase rutile. Transformation from amorphous to anatase form requires temperatures near 300ºC. Rutile exhibits the highest refractive indices (and is birefringent) and from previous studies requires transformation temperatures in excess of 800ºC to form. With optical applications employing e-beam evaporation, substrate temperatures in excess of ~250ºC are rarely achieved because of chamber temperature and other operation limitations as well as substrate limitations. Therefore, the films generally produced are a mixed composition with little or no rutile component. Post-deposition annealing at high temperature is required to produce the transformation. We shall discuss work describing how the rutile form can be grown at lower temperatures. These studies include energy, pressure, and rate variations applied to several deposition processes. We believe that these reviews enhance the technology base of our readers, enabling them to improve their own processes and consequently their products.

Thickness Dependence of the Relative Amounts of Crystalline Forms

Temperature is an important parameter involved in determining composition. A study involving rf sputter deposition of titania films on unheated substrates found that the relative amounts of the crystalline forms varied with film thickness. At thicknesses 256 nm, 506 nm, and 705 nm deposited under fixed conditions, amorphous titania co-exists with anatase and rutile phases, anatase being the dominant constituent. As thickness increased, anatase forms at the expense of the amorphous phase, while the rutile concentration remained constant. What’s going on here?

When the films were annealed at 300ºC, no changes were observed. Annealing at 500ºC also does not change the relative concentrations, but does improve the crystalline perfection of both the anatase and the rutile phases through diffusion of point defects. Annealing at 700ºC causes the amount of rutile to increase and that of anatase to decrease. After one hour the anatase disappears entirely. The conclusion of the study is that the thickness dependence of crystalline state concentration is an artifact of the temperature increase associated with the longer deposition time required to build thicker layers. Surface diffusion of adsorbed species at the growth interface controls the crystallization under these conditions. [1].

Back to Top

Ion-Beam Sputtering

High-energy techniques permit deposition at lower substrate temperatures with improved optical and mechanical properties compared with conventional evaporation. Here we discuss some of the recently developed processes. Sputtering has been examined previously in CMN.

One such process for depositing rutile at temperatures lower than historically reported is ion beam sputtering. The rutile structure was produced by IBS using Ar+ at 1.2 kV to sputter a Ti target at 45 deg. incidence. The ion current at the target was 1.2 to 1.8 mÅ/cm2. The rutile form was grown on MgO substrates that were maintained at 630ºC, well below the previously required 800ºC for growing the rutile structure. The deposition rate of Ti atoms and the partial pressure of O2 determine whether anatase or rutile is deposited. It was found that only the anatase structure was produced when the oxygen impingement rate was very high or the Ti impingement rate was very low. In this work, the arrival rate for Ti atoms was 0.1 nm / min., anatase was produced at 1.1 E-02 Pa and rutile was produced when the partial pressure was increased to 3.1 E-03 Pa for the same Ti rate. The arrival rates of O2 molecules and Ti atoms need to be equal, i.e., supplied simultaneously, in order to grow the rutile structure. [3]

Back to Top

Deposition of Rutile by E-beam and IBED

A method for growing rutile epitaxially at relatively low temperatures uses high energy ion beams. Epitaxial films grow with a structure that is oriented to the crystal direction of the substrate. In this work, titanium was evaporated by e-beam and oxygen was supplied. Xenon ions of energies 40 – 80 keV at current densities 20 – 40 uA/cm2 bombarded the growing film (Ion Beam Enhanced Deposition) at 45 deg. incidence. MgO substrates were maintained at 132º - 550ºC. For 40 keV energy and 20 uA/cm2, all films for substrate temp.132º – 250ºC were nearly stiochiometric, but as the temperature was increased, the oxidation increased. As the density of Xe+ was increased, the oxidation decreased because the sputtering rate of oxygen is greater than that of Ti. At 250ºC and above, all films exhibited rutile structure. The only effect of the higher Xe+ energy was to increase the perfection of these highly oriented crystallites. The explanation offered is that the different crystal planes have different sputtering rates, with that of rutile being lowest. This technique is able to deposit the rutile structure at a relatively low temperature. [2]

Back to Top

Plasma Ion-Assisted Deposition

This is a set of deposition technologies that includes IAD, plasma IAD, arc discharge, and ion plating. In plasma assisted IAD, a hot cathode, large-area source creates a high plasma current density in the volume of the coating chamber and a high degree of ionization of the reactive gas and the evaporant is produced. Evaporation can be by e-gun or resistance-heated source. The plasma sustains a positive charge relative to the substrate holder, therefore ions from the plasma are accelerated and bombard the substrates during film growth. Oxides, fluorides, sulfides as well as metals can be deposited with this technique. An ‘advanced plasma source’ introduced by Leybold AG has demonstrated significant improvements in density, index, and durability (scratch resistance) over conventional techniques. Film microstructure can be controlled as a function of ion current density. Thus the typical columnar structure of TiO2 can be made to disappear at intermediate plasma densities or transformed to a dense rutile structure at high plasma densities, all without substrate heating. [4]

Back to Top

Ion Plating

This technology is also based on activated reaction of species by an energetic plasma, and in fact its development precedes the plasma ion assisted process described previously. A variation used for the production of tribological as well as optical coatings at low substrate temperatures is called ‘reactive low-voltage ion plating’. A high current beam of electrons ionizes and activates the evaporant and reactive gas, which are then accelerated toward the negatively biased substrate holder. The bias is <20 v,="" and="" eliminates="" the="" possibility="" of="" sputtering.="" the="" high="" powers="" consumed="" in="" creating="" the="" plasma="" do="" result="" in="" substrate="" heating,="" however.="" crn,="" ticn,="" and="" tin="" among="" other="" tribological="" coatings="" are="" deposited="" at="" substrate="" temps.="" below="" 500º="" c.="" optical="" films="" consisting="" of="" metal="" oxides="" are="" produced="" with="" full="" oxidation,="" high="" density,="" and="" with="" an="" amorphous="" structure.="" refractive="" index="" approaches="" bulk="">

Back to Top

Achieving Adhesion of Gold to Glass: Another Solution

Inert metals, gold being probably the extreme example, do not form strong bonds with substrate materials, and therefore present adhesion problems. We have discussed the well-known method of depositing a precursor layer to promote adhesion for difficult cases. The adhesor layer is often chromium or titanium, but bismuth oxide (slightly reduced) has been used successfully. The metals form oxides that bond to both substrate surface and to the gold layer. In the case of semiconductors that require subsequent high temperature processing, a barrier layer of nickel or palladium might be interposed to prevent interdiffusion. An oxygen plasma has also been used to condition the surface of glass for enhanced adhesion. Similarly, IAD has been successful. We review another approach for obtaining adhesion of gold films to glass surfaces.

Gold was ion beam sputtered onto glass substrates held at room temp. An Ar+ beam of 1 keV energy sputtered a 1600 Å gold layer. The deposition rate was 0.4 Å/s. After deposition, the films were irradiated with 1 keV Ar+ at 15 uA/cm2 current density. The fluence of the beam was varied and the films examined for roughness, thickness, and adhesion. Adhesion was tested by the scratch method and adhesive tape, thickness by Rutherford back scatter, and roughness by atomic force microscopy. As the ion dose was increased, sputtering by the Ar+ caused the roughness of the gold to increase and the thickness to decrease. Between approximately 5 E16 and 2 E17 ions/cm2, the thickness decreased to about half and the roughness increased six times. Simultaneously, the adhesion increased three times. Compared with un-irradiated films, the adhesion is increased nine times. A reasonable compromise between the three performance parameters is to limit the dose to ~15 E16 ions/ cm2. The increased adhesion is explained as a relaxation (lowering) of the energy at the glass – gold interface. [6].

Back to Top


  1. J. D. DeLoach and C. R. Aita, J. Vac. Sci. Technol. A, Vol. 16, No. 3, May/Jun 1998, 1963.
  2. Feng Zhang, Zhihong Zheng, Xingzhao Ding, Yingjun Mao, Yu Chen, Zhuyao Zhou, Shiqi Yang, and Zianghuai Liu, J. Vac. Sci. Technol. A, 15(4), Jul/Aug 1997, 1824.
  3. Toshihiko Aoki and Kunisuke Maki, J. Vac. Sci. Technol. A, Vol. 15, No. 5, Sept/Oct 1997, 2485.
  4. S. Pongratz and A. Zoller, J. Vac. Sci. Technol. A, Vol. 10, No. 4, Jul/Aug 1992, 1897.
  5. H. K. Pulker, J. Vac. Sci. Technol. A, Vol. 10, No. 4, Jul/Aug 1992, 1669.
  6. H. G. Jang, K. H. Kim, S. Han, W. K. Choi, H. J. Jung, and S. K. Koh, J. Vac. Sci. Technol. A, Vol. 15, No. 4, Jul/Aug 1997, 2234.

 *Newsletters beginning with Vol. 6 are available from the CMN archives page of this web site.  Contact CERAC for printed copies of referenced newsletters prior to Vol. 6.

If you have a question or a topic you would like us to consider for a future issue of CMN, e-mail your requests to or fax them to 414-289-9804. 

(S.F. Pellicori is available for private consulting on matters concerning optical thin films. Please contact him directly for more information)

Russ DeLong
CERAC, inc.
P.O. Box 1178 | Milwaukee, WI 53201
Phone: 414-289-9800 | FAX: 414-289-9805

Principal Contributor:
Samuel Pellicori
Pellicori Optical Consulting
P.O. Box 60723 | Santa Barbara, CA 93160
Phone/FAX: 805-682-1922

back to top All printed, graphic and pictorial materials made available on this website are owned by CERAC and protected by Federal Copyright laws. None of the materials, in whole or in part, may be reprinted and distributed or otherwise made available to others for any purpose without CERAC's prior written consent.

Phone:  414-289-9800 /  FAX: 414-289-9805  /