Experimenting with film thickness and fluorine concentration was fairly straightforward due to the consistency of the results. First, I kept the solution contents constant and varied the amount of solution to be sprayed, until the glass produced had a sheet resistance of approximately 10 ohms per square. For a 3 by 3 inch square of window glass this required 10mL of solution, but judging by the overspray on the hot plate, this amount could probably cover a 4 by 4 square just as easily. Next, I kept the solution volume constant at 10mL and varied the fluorine concentration, recording the best sheet resistance value for each concentration. A graph of that data can be seen below.

   Following the graph, tin and fluorine should be in a 2:1 atomic ratio in the solution for optimal results. Using the same ingredients as in the previous article, this means the spray solution should contain the following:

•   30mL Distilled Water
• 3384mg Stannous Chloride
•  214mg Ammonium Bifluoride
•  ???mL Hydrochloric Acid (add dropwise until the solution becomes clear)

   I then used various volumes of the above solution to produce glass with different sheet resistances, which can be seen at the top of the page. The glass is highly transparent, with little to no discoloration or haze. The conductive film is mechanically durable and waterproof; it can be scrubbed clean with soap and water with no effect on the transparency or conductivity. Overall, this glass is comparable to commercial conductive glass, with one important difference: the cost. Commercial conductive glass, at the time of writing this article, costs approximately $1 per square inch. Discounting the cost of tools (since the tools required are common and may already be available), a square inch of the type of glass described here costs approximately 1¢ to produce. The importance of this development cannot be overstated. Over the past decade, I have seen many experimenters hit a wall while working on a project that requires a transparent conductor, due to the high cost of commercial products and the inadequacy of homemade solutions. I believe this is now a solved problem. I have included a video below, which demonstrates the simplicity of the deposition process. Anyone can do this.

   Since I first wrote this article I have made multiple square feet of this glass, and have discovered a number of additional details regarding the process:

• Stannous chloride is air-sensitive, even in its crystalline form. Over time, it will lose the ability to form clear solutions, even with the addition of hydrochloric acid. Fortunately, cloudy solutions can still be used to make conductive glass, and I've noticed no significant difference in transparency or conductivity.
• There is no upper limit for the temperature of the substrate other than its melting point. I once let my hot plate reach a dull red glow on accident, but the glass being heated still ended up highly conductive.
• There is a second cause of glass breakage, other than spraying for too long. If a large piece of glass is sprayed at a right angle to one of its edges, it will break instantly. Spraying should be done from corner to corner to avoid this.
• The above is more likely to happen the lower the spray angle is, and is likely due to cold air being forced under the hot glass. A spray angle of 45 degrees is adequate. Spraying straight down would be optimal, but is incompatible with hand-spraying. Airbrushes tend to drip, and acid vapors travel straight up.
• Conductive glass is cleanest when it is freshly made. For projects requiring cleanliness, extensive solvent treatments can be avoided if the glass is used immediately.