Program Speaker
Scott Wilson on Opal Synthesis

by Nancy L. Attaway and Scott Wilson

President Scott Wilson presented his talk on “Opal Synthesis” and discussed his experiments in synthesizing opal. Born and raised in Los Alamos, Scott studied one year at New Mexico State University in Las Cruces, and he graduated from the University of New Mexico with three degrees (BS, MS, and Ph.D.). While still an undergraduate, Scott started his own company, Sandia Systems, where he developed techniques of non-contact machining of precision optics and laser instrumentation for semi-conductor metrology. The company changed hands several times and is now owned by Accent Optical Technologies, for whom Scott now works.

Scott began his talk by outlining what he intended to cover. The five divisions of his talk included a definition of opal, a description of how opal is formed in nature, an explanation of the laboratory growth of opal, and the presentation of results. Scott concluded his talk with a discussion of technology related to opals and opal synthesis. Scott provided copies of several references on opal synthesis that included: “Controlled Growth of Monodisperse Silica Spheres in the Micron Size Range” by Stober and Fink; “Origin of Precious Opal” by Darragh, et. al.; “Color of Precious Opal” by Sanders, and copy of a patent on “Opaline Materials and Method of Preparation”.

Scott described opal as a three-dimensional diffracting array composed of hydrated silica, active in the optical wave length region. Opal is a crystal-like lattice of objects in space, in this case, silica microspheres. The spacing and the uniformity of the lattice is sufficient to diffract electro-magnetic radiation (in precious opal, this is light) in accordance with Bragg’s Law. The packing of the lattice is face-centered, cubic, with faults, twins, and dislocations (Sanders, 1964), just as you might see in an actual crystal.

Scott explained that hydrated silica is nearly pure SiO2 with water entrapped at the molecular level, usually as hydroxyl ions. Natural opal shows trace elements of sodium, calcium, potassium, and aluminum. Synthetic opal may also show Ti and other metals.

Opal is active in the optical wavelength region from 400 nanometers to 700 nanometers. From Bragg’s Law, the diameter of the spheres may be calculated, something like 250-350 nanometers for red-green opal flash, and 150-250 nanometers for green-blue flash.

Opal formation in nature is an area subject to much conjecture and study. Darragh gave a description of opal formation in 1966, and it still holds true to a great extent, even in light of very recent work. Darragh described a 100 parts per million silica solution, dissolved from host rocks at about 25 degrees C in slightly acid waters. Slow evaporation of the solution over a period of about 1 million years is followed by condensation to form about 30 nanometer spheres in colloidal form. Aggregation forms about 300 nanometer spheres of uniform size, followed by sedimentation (settling) that forms the opal lattice. Hardening occurs by dessication and partial fusing by silica gel. Only high quality lattices yield precious opal. The lattice quality is controlled in later stages by the geologic stability of the growth environment.

Opal growth in the lab follows a similar trail. In general, silica spheres of the correct size must be procured that have very narrow size distribution. Sedimentation is used to form a lattice. Void filling is done, often to a partial degree, with a silica gel (although some folks have done it with epoxy), all followed by a hardening process. The trick is to try to get this done in less than a year’s time!

To obtain the silica spheres, one might buy them from any of several suppliers of technical materials. Silica spheres are very expensive now, but they were much cheaper at one time. In the past, silica spheres were used in paints, cosmetics, food, and juices. They are no longer widely available, due to a competing process (fumed silica) that forms irregular-shaped particles.
Another possibility might be to try to “mine” such particles. It may be possible to separate silica spheres from suitable soils, like those found in Australia. There is no actual evidence of this method being used in the literature, but Cram implies that it could be done (“only pennies for the raw materials”. Sounds “dirt” cheap!).

Silica spheres can be made, and there are many ways to this. All involve some process that takes place under significant acid or base conditions that usually involve a combination of nucleation, growth, and aggregation. Aggregation should probably be avoided, due to its tendency to produce non-uniform-sized particles. The most common approaches are sodium silicate ion exchange and condensation from silicic acid. Both create “sols”. Hopefully, they don’t “gel” until the end, or you would end up with “potch”. Temperature, ph, motion, and chemistry are all factors that must be carefully controlled in order to obtain particles suitable for opal synthesis.

The Stober-Fink method (1968) is a condensation approach that makes use of ammonia as a catalyst to condense silicic acid from hydrolytic breakdown of TEOS in water-ethanol solution. The sphere size is controlled by ammonia/water ratio, which is something that requires only careful measurement. This process is done at room temperature, which makes is relatively convenient.

TEOS, tetraethylorthosiloxane or tetra-ester of silicic acid, is a chemical used in large quantities in semiconductor fabrication to make high quality SiO2 (glass) for integrated circuits. TEOS can be considered to be ethyl alcohol with silicic acid bonded in a molecular structure. It is highly flammable and “burns” to form glass, so it absolutely must be handled carefully using the proper equipment, alway being mindful of the safety precautions.

Scott’s procedure for opal synthesis is, first, select the ammonia/water ratio to get desired sphere size. After scaling to the size of the “batch”, mix the TEOS and ethanol. Mix ammonia and water. Rapidly add ammonia/water mixture to TEOS/ethanol while stirring rapidly (rapid stirring helps to maintain a narrow size distribution). Stir for an hour and cover to keep dust out and also to reduce evaporation. This results in production of a milky colloidal suspension of the silica spheres. To centrifuge and separate the spheres, pour off the excess liquid. Re-disperse in ethanol (wash). Centrifuge again. Re-disperse to the desired density in ethanol. The trick to the re-dispersal is to partially submerge the solution container in an ultrasonic cleaner, as the acoustic cavitation helps break up any aggregates that may have formed.

The sedimentation process follows next. The cake produced during centrifuge processing shows definite color, but it is not good enough yet. It is still weak and diffuse, so a slow and controlled sedimentation process must be carried out. Gravitational sedimentation takes a very long time (It is suspected that Gilson does it this way). The trick to doing this quickly is to perform the sedimentation under silicone oil (Philipse, 1989). This procedure involves adjusting the specific gravity of the solution to match the oil, then pouring the solution on top of the oil and covering it with a plastic membrane to keep out dirt and dust. Store for about four weeks in a quite, draft-free, constant-temperature location. The solution slowly loses the ethanol by evaporation through the oil. It sinks to the bottom of the oil as a “blob”. Good opal color shows as the blob shrinks.

The results are good pinfire color, but the material is very soft. Better results are often obtained if a 10% methanol in ethanol solution is used during sedimentation, due to chemical reasons and electrochemical forces. Even with all of this work, the awesome color saturation that you look for in opal is just not quite there.

To improve the color, something must be done to increase the quality of the microsphere lattice. Many things can affect the quality of the lattice. A very important one involves electrostatic forces. The spheres tend to charge negative, due to the disassociation of surface silanol groups or the absorption of hydroxyl ions. Charged particles rarely behave nicely. The surface charging tends to cause the microspheres to resist efficient packing in the lattice, thus forming lattice defects. This problem can be reduced by modifying the microsphere surfaces with a silane coupling agent (SCA), which reduces surface charging and allows the spheres to more easily form a high quality lattice arrangement.

It is interesting to look at the costs of doing this work. The cost for one gram of silica, produced by the above process is: TEOS $3; ammonia $0.50; EtoH $3; and TPM (SCA) $10, for a total of $26.50 per gram. The silicon oil can be reused many times. Silicone brake fluid appears to work well, too, but it may produce a discolored opal.

The synthesized opal needs to be hardened. This can be done in lots of ways. Heating is an obvious approach, but Scott found that furnace-fusing at between 200 degrees C to 600 degrees C caused the opal to lose color and mechanical strength, possibly due to thermal breakdown of the silicon oil. Scott is experimenting with solutions to wash the oil out of the opal. Adding some weak silica solution to the soft opal (by soaking) may assist in fusing the microspheres during heating.

Scott reported that the current state of this process takes about three hours for preparation of the microspheres and the initial solution and about six weeks for sedimentation, a scalable but expensive approach. Work needs to be done to find a cheaper SCA. It will certainly be worthwhile looking at the silicate route to microsphere preparation. More work on hardening also remains to be done.

There are areas of optical technology that are related to the growth and synthesis of opal. Optical bandgap materials are specially constructed opals, just a few lattice layers thick. These materials typically have multiple sphere types in the opal and are arranged in a complex, computer-designed array. These materials exhibit special optical properties, much like a semiconductor for electrons but for photons (light). They can exhibit complex (and possibly useful) wavelength dependent behavior. The technology of optical bandgap materials is currently where semiconductor technology was about 40 years ago. Perhaps, one day we will be using what amounts to a computer built with optical transistors to surf the Internet.

Scott closed with a word on safety. These chemicals represent a most serious fire danger due to their flammability and combustion by-products. The chemicals and vapors can and will cause serious burns, eye damage, respiratory problems, etc. Special equipment and a thorough understanding of the hazards is required by a person who handles these materials. Serious ventilation is required. Scott concluded with the recommendation that you “Don’t Try This At Home,” which spoiled all the fun! Thank you, Scott.