Composition and property measurements for PHA Phase 4 glasses

Edwards, T.B.

Savannah River Site (S.C.)

关键词: Sludges;    Radioactive Waste Processing;    Waste Forms;    36 Materials Science;    Phase Studies;   

DOI  :  10.2172/750899
RP-ID  :  WSRC-TR-99-00294/Rev.0
RP-ID  :  AC09-96SR18500
RP-ID  :  750899

美国|英语

来源: UNT Digital Library

【 摘 要 】

The results presented in this report are for nine Precipitate Hydrolysis Aqueous (PHA) Phase 4 glasses. Three of the glasses contained HM sludge at 22, 26, and 30 wt% respectively, 10 wt% PHA and 1.25 wt% monosodium titanate (MST), all on an oxide basis. The remaining six glasses were selected from the Phase 1 and Phase 2 studies (Purex sludge) but with an increased amount of MST. The high-end target for MST of 2.5 wt% oxide was missed in Phases 1 and 2 due to {approximately}30 wt% water content of the MST. A goal of this Phase 4 study was to determine whether this increase in titanium concentration from the MST had any impact on glass quality or processibility. Two of the glasses, pha14c and pha15c, were rebatched and melted due to apparent batching errors with pha14 and pha15. The models currently in the Defense Waste Processing Facility's (DWPF) Product Composition Control System (PCCS) were used to predict durability, homogeneity, liquidus, and viscosity for these nine glasses. All of the HM glasses and half of the Purex glasses were predicted to be phase separated, and consequently prediction of glass durability is precluded with the cument models for those glasses that failed the homogeneity constraint. If one may ignore the homogeneity constraint, the measured durabilities were within the 95% prediction limits of the model. Further efforts will be required to resolve this issue on phase separation (inhomogeneity). The liquidus model predicted unacceptable liquidus temperatures for four of the nine glasses. The approximate, bounding liquidus temperatures measured for all had upper limits of 1,000 C or less. Given the fact that liquidus temperatures were only approximated, the 30 wt% loading of Purex may be near or at the edge of acceptability for liquidus. The measured viscosities were close to the predictions of the model. For the Purex glasses, pha12c and pha15c, the measured viscosities of 28 and 23 poise, respectively, indicate that DWPF processing may be compromised at the low end of the viscosity range (20 poise). Although the HM sludge glass examined (10 wt% PHA) had a measured viscosity of {approximately}90 poise, the HM glasses at 7wt% PHA are predicted to be higher than the 100 poise limit for DWPF. Further work will be required to resolve these issues.