Impurity leaching rates of 1000 liter growth tanks

This memo reports on the analysis of some recent measurements of solution impurity levels in the three KDP and one DKDP Pilot Production 1000 liter growth tanks (Tanks B, C, D, & F). Solution samples were taken on a weekly basis during recent crystal growth runs in each tank and were analyzed by inductively coupled plasma emission spectroscopy (ICP-ES). The solution history for five specific elements, Si, B, Al, Fe and Ca will be analyzed in detail. The first four of these elements are input into solution via slow dissolution of the glass vessel at a rate which is strongly dependent on the solution temperature. Si and B continuously accumulate in solution, since they are not incorporated into the crystal. Al and Fe by comparison are incorporated into the crystal (primarily the prismatic sectors) and present problems to inclusion-free growth (Al) and 30 damage (Fe). The level of these impurities initially increases when the crystal size is small but later decreases when the rate of incorporation into the crystal exceeds the rate of dissolution of the glass tank. The last element, Ca is of interest since it has recently been observed to be one of the elements found at the location of 3cu damage. For Si and B, the dissolution or leach rate from the glass tank is easily obtained from the results of the chemical analysis. The temperature dependent leach rates are shown to be comparable (within a factor of two) for all four tanks, with Tank B (DKDP) having the lowest rate of Si accumulation. The glass leach rates of the two incorporating elements Al and Fe require substantially more analysis as the daily variation of the crystal dimensions, the solution concentration, and the mass of KDP remaining in solution must be taken into account in order to separate the rate of impurity incorporation from the rate of dissolution of the glass. The method for accomplishing this separation is described, and the result obtained is that the leach rates of all four tanks are within a factor of three of each other. Tank B again shows the lowest leach rate for both Al and Fe. The results for Ca are less clear. From the present data, the level of Ca does not change appreciably during a run, indicating that it .is neither leaching from the glass strongly nor being incorporated into the crystal at a significant rate. It does increase with the age of the solution, however, as Ca is a small but measurable component of both the starting salt as well as the D,O. Older solutions that have successfully grown several crystals will therefore have higher accumulated levels of Ca, which will increase the driving force for Ca incorporation into the crystal.