D. Koopman

In this paper we present our results from a molecular dynamics study of n-octane liquids confined between planar bcc solid surfaces. The systems studied were wide enough to develop a bulklike region throughout the middle portion of the film and two well-separated interfaces. Our work focused on segmental dynamics and relaxation of ‘‘adsorbed’’ octane molecules. In particular, we investigated the role of architectural and dynamical features peculiar to short chain molecules (almost fixed bend angles and restricted torsional rotations) on the dynamics of ‘‘adsorbed’’ chains. We found that the relaxation of octane molecules exhibits the same qualitative trends as those observed in molecular simulations of generic ‘‘bead-spring’’ oligomer films. The most important effect is the dramatic slow down of rotational motions (up to a factor of 1000) for chains adsorbed on strongly physisorbing surfaces (adhesion energy per segment of 1–2 kT). Despite the qualitative similarities with bead-spring chains, the dynamics of realistic short hydrocarbon chains are affected much more strongly by the interfacial environment than their bead-spring counterparts. These stronger effects originate largely from the suppression of torsional angle transitions inside the extremely dense first layer (in cases of strong physisorption). The frequency of torsional transitions was found to be correlated directly with the amount of ‘‘free volume’’ available inside the crowded first layer.

Due to higher than expected hydrogen generation during the Tank 51-Sludge Batch 4 (SB4) qualification run, DWPF engineering requested the Savannah River National Laboratory (SRNL) to expand the ongoing catalytic hydrogen generation program. The work presented in this Technical Report was identified as part of SRNL/Liquid Waste Organization (LWO) meetings to define potential causes of catalytic hydrogen generation as well as from an external technical review panel commissioned to evaluate SRNL hydrogen related data and programs. New scope included improving the understanding of SRAT/SME process chemistry, particularly as it related to acid consumption and hydrogen generation. The expanded hydrogen program scope was covered under the technical task request (TTR): HLW-DWPF-TTR-2007-0016. A task technical and quality assurance plan (TT&QAP) was issued to cover focus areas raised in meetings with LWO plus a portion of the recommendations made by the review panel. A supporting analytical study plan was issued. It was also noted in the review of catalytic hydrogen generation that control of the DWPF acid stoichiometry was an important element in controlling hydrogen generation. A separate TTR was issued to investigate ways of improving the determination of the acid requirement during processing: HLWDWPF-TTR-0015. A separate TT&QAP was prepared for this task request. This report discusses some progress on this task related to developing alternative acid equations and to performing experimental work to supplement the existing database. Simulant preparation and preliminary flowsheet studies were already documented. The prior work produced a sufficient quantity of simulant for the hydrogen program and melter feed rheology testing. It also defined a suitable acid addition stoichiometry. The results presented in this report come from samples and process data obtained during sixteen 22-L SRAT/SME simulations that were performed in the second half of 2007 to produce eight SME products with frit 418 and a matching set of eight SME products with spherically beaded frit 418. The requirement to produce two 25 gallon batches of melter feed for the melter feed rheology modifier program fell under a separate task plan. One supporting 4-L SRAT simulation was performed with mercury, since the 22-L melter feed preparation runs had no mercury due to melter off-gas constraints. As a result of this work, a timeline of reactions has been developed showing the sequence of major reactions occurring during and shortly after acid addition. The traditional-style simulant used in this testing had fairly well defined speciation which enabled the reactions being observed to be related to acid consumption. The new coprecipitated simulants have somewhat different speciation, and it will be necessary to validate some of the conclusions from this testing using sample data from SRAT simulations with coprecipitated simulant. Noble metal dissolution data on timing and concentration were presented in a separate report discussing hydrogen generation. A few of those results will be brought into this report as part of the description of the SRAT chemistry timeline. The noble metal and mercury concentrations used in the preliminary flowsheet studies are summarized in Table 1 along with the ranges covered in this study.

Simulations of the Defense Waste Processing Facility (DWPF) Chemical Processing Cell vessels were performed with the primary purpose of producing melter feeds for the beaded frit program plus obtaining samples of simulated slurries containing high concentrations of noble metals for off-site analytical studies for the hydrogen program. Eight pairs of 22-L simulations were performed of the Sludge Receipt and Adjustment Tank (SRAT) and Slurry Mix Evaporator (SME) cycles. These sixteen simulations did not contain mercury. Six pairs were trimmed with a single noble metal (Ag, Pd, Rh, or Ru). One pair had all four noble metals, and one pair had no noble metals. One supporting 4-L simulation was completed with Ru and Hg. Several other 4-L supporting tests with mercury have not yet been performed. This report covers the calculations performed on SRNL analytical and process data related to the noble metals and hydrogen generation. It was originally envisioned as a supporting document for the off-site analytical studies. Significant new findings were made, and many previous hypotheses and findings were given additional support as summarized below. The timing of hydrogen generation events was reproduced very well within each of the eight pairs of runs, e.g. the onset of hydrogen, peak in hydrogen, etc. occurred at nearly identical times. Peak generation rates and total SRAT masses of CO{sub 2} and oxides of nitrogen were reproduced well. Comparable measures for hydrogen were reproduced with more variability, but still reasonably well. The extent of the reproducibility of the results validates the conclusions that were drawn from the data.

The Savannah River National Laboratory (SRNL) accepted a technical task request (TTR) from Waste Solidification Engineering to perform simulant tests to support the qualification of Sludge Batch 8 (SB8) and to develop the flowsheet for SB8 in the Defense Waste Processing Facility (DWPF). These efforts pertained to the DWPF Chemical Process Cell (CPC). Separate studies were conducted for frit development and glass properties (including REDOX). The SRNL CPC effort had two primary phases divided by the decision to drop Tank 12 from the SB8 constituents. This report focuses on the second phase with SB8 compositions that do not contain the Tank 12 piece. A separate report will document the initial phase of SB8 testing that included Tank 12. The second phase of SB8 studies consisted of two sets of CPC studies. The first study involved CPC testing of an SB8 simulant for Tank 51 to support the CPC demonstration of the washed Tank 51 qualification sample in the SRNL Shielded Cells facility. SB8-Tank 51 was a high iron-low aluminum waste with fairly high mercury and moderate noble metal concentrations. Tank 51 was ultimately washed to about 1.5 M sodium which is the highest wash endpoint since SB3-Tank 51. This study included three simulations of the DWPF Sludge Receipt and Adjustment Tank (SRAT) cycle and Slurry Mix Evaporator (SME) cycle with the sludge-only flowsheet at nominal DWPF processing conditions and three different acid stoichiometries. These runs produced a set of recommendations that were used to guide the successful SRNL qualification SRAT/SME demonstration with actual Tank 51 washed waste. The second study involved five SRAT/SME runs with SB8-Tank 40 simulant. Four of the runs were designed to define the acid requirements for sludge-only processing in DWPF with respect to nitrite destruction and hydrogen generation. The fifth run was an intermediate acid stoichiometry demonstration of the coupled flowsheet for SB8. These runs produced a set of processing recommendations for DWPF along with some data related to Safety Class documentation at DWPF. Some significant observations regarding SB8 follow: Reduced washing in Tank 51 led to an increase in the wt.% soluble solids of the DWPF feed. If wt.% total solids for the SRAT and SME product weren’t adjusted upward to maintain insoluble solids levels similar to past sludge batches, then the rheological properties of the slurry went below the low end of the DWPF design bases for the SRAT and SME. Much higher levels of dissolved manganese were found in the SRAT and SME products than in recent sludge batches. Closed crucible melts were more reduced than expected. The working hypothesis is that the soluble Mn is less oxidizing than assumed in the REDOX calculations. A change in the coefficient for Mn in the REDOX equation was recommended in a separate report. The DWPF (Hsu) stoichiometric acid equation was examined in detail to better evaluate how to control acid in DWPF. The existing DWPF equation can likely be improved without changing the required sample analyses through a paper study using existing data. The recommended acid stoichiometry for initial SB8 SRAT batches is 115-120% stoichiometry until some processing experience is gained. The conservative range (based on feed properties) of stoichiometric factors derived in this study was from 110-147%, but SRNL recommends using only the lower half of this range, 110-126% even after initial batches provide processing experience. The stoichiometric range for sludge-only processing appears to be suitable for coupled operation based on results from the run in the middle of the range. Catalytic hydrogen was detectable (>0.005 vol%) in all SRAT and SME cycles. Hydrogen reached 30-35% of the SRAT and SME limits at the mid-point of the stoichiometry window (bounding noble metals and acid demand).