19 Leslie Court
Whippany, New Jersey 07981
TEL. (973)386-0053
FAX (973)503-0256




These solutions have been available for some time and there has been some confusion on the use of and the difference between these products. In addition, there has been a question concerning the fact that there is usually a reference to the Co of the solutions even though they are not used as diffusion sources. With regard to the latter, the Co is specified as a simple way to indicate the phosphorous content of the glass. Since the determination of the phosphorous concentration in the solution is not a straightforward chemical analysis procedure and since all of us have diffusion facilities, it is a simple matter to specify the doping capability of the film to indicate the phosphorous content of the glass.

With this in mind we list the difference between the aforementioned formulations:

  Film Thickness in angstroms at 2000 RPM spin speed Co atom/cm3
in Si after 15 min diffusion
SILICAFILM with Phosphorous 2000 1 x 1019
PHOSPHOROSILICAFILM for Glassivation 2000 5 x 1019
GLASS FORMING SOLUTION 306 3500 5 x 1019

As to applications and the reason for three different formulations:
Silicafilm with Phosphorous, the first of this family was intended to be applied as a phosphorous containing SiO2 layer after emitter diffusion when one had problems with photo resist adherence to a highly phosphorous doped thermal oxide. One simply removed the phosphorous rich layer and deposited this film which contained less phosphorous to achieve good resist adhesion. However the practice of applying a glass layer over a metallized IC demonstrated that Silicafilm with Phosphorous was inadequate for the purpose. The film cracked over the steep steps provided by the metallization.

Phosphorosilicafilm for Glassivation was then developed which does not crack over the metallization steps. While the phosphorous content is sufficient to provide a sodium barrier, in this instance this is not a necessary requirement of this film. The film is intended to provide scratch protection for the underlying metal. However the phosphorous content is such that surface is not hygroscopic and no phosphorous etching of the aluminum occurs in high relative humidity.

The last product Glass Forming Solution 306 is similar to Phosphorosilicafilm for Glassivation in its phosphorous content. However, it was developed originally as a diffusion mask for zinc in III-V diffusion. Since it was required that the film be thicker than the first two films, several firms found that Glass Forming Solution 306 may be used in place of Phosphorosilicafilm for Glassivation and one can obtain a thicker glassivation with the same number of applications.

All these solutions are applied in the same way. Spin-densify at 400oC for 1 hour, spin a second coat and densify at 400oC for 1 to 2 hours. The layers will be very uniform in thickness, they will etch uniformly in dilute HF solution and with automatic spinners, you can achieve a faster through put than with CVD. With the volume discount, this process eliminates the high capital investment in CVD equipment. After all, the $40,000 to $100,000 needed for CVD could earn, at 9 or 10%, the cost of our solutions, and you will have no rejects due to overetched aluminum or open circuits where the glass was not completely removed. Nor will your firm be noted in the Wall Street Journal because of phosphorous problems.

Alphabetic Product Listing home.gif (1052 bytes)
mail25.gif (398 bytes)
Material Safety Data Sheet