A remotely operated, field deployable tritium analysis system for surface and groundwater measurement Page: 7 of 11
This report is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
The commercially available "tritium columns" (EIChroM Industries, Inc.), containing about 2 cc
of three distinct types of absorbing material, were tested for their applicability to automated use in the
proposed autosampler. The columns are designed to remove the matrix elements from solution,
resulting in a column effluent similar to a sample that had been prepared for tritium analysis by
distillation. Capacity studies showed that column performance did not significantly degrade with
sample volumes up to 50 mL, even when starting with a seawater matrix saturated with chlorinated
hydrocarbons, under gravity flow conditions (about 0.5 mL/minute). At flow rates greater than 10
mL/minute, however, some break through of matrix ions was evident. A flow rate of 5 mL/minute was
thus chosen as optimum for column performance and sample throughput. As the resin columns are
shipped from the manufacturer in water, the first several milliliters of sample eluting from the column is
somewhat diluted. Laboratory studies showed that an equilibrium was established after 15 mL of
sample had passed through the column. It was therefore decided to discard the initial 20 mL of sample
through the column to avoid the dilution effect. A typical tritium analysis operation requires 5 mL of
sample so that several duplicate analyses can be run from a single 50 mL sample.
A variety of LS cocktails, including Packard Ultima Gold LLT, Ultima Gold AB, Ultima Gold
XR, and Ultima Flo M were evaluated for use with the FDTAS. In addition to the parameters of
background and counting efficiency, the viscosity of the cocktail also required consideration. Sample
backgrounds were determined using a Packard 2550 LS counter. The cocktails were spiked with a
known aliquot of known tritium activity and then counted to determine counting efficiency. The Ultima
Gold LLT, normally a clear choice for tritium counting due to its low background and high efficiency,
was found to be too viscous to be readily washed from the sample vial during the cleaning cycle. Two
other viscous cocktails, Ultima Gold XR and Ultima Flo M were found to have a significantly higher
background and low counting efficiency when compared with Ultima Gold AB. Although the Ultima
Gold AB has a high viscosity, it could be washed out of the vial using methanol and was chosen as the
best cocktail for this application. Using Ultima Gold AB, a series of tests were performed to determine
the optimum water to cocktail ratio. Cocktail:water mixtures of 40:60, 50:50, 60:40, and 70:30 were
made up and counted to determine the backgrounds. The mixtures were then spiked with a know tritium
activity and then recounted to determine the tritium detection efficiency. The 50:50 cocktail:water ratio
was found to produce the best overall results for the FDTAS measurements.
Many tests were conducted to maximize the efficiency of the autosampler and flow-cell flushing
operations to optimize the cleaning cycle, minimize the amount and type of waste, and minimize cross
contamination of samples. Cleaning of the autosampler is accomplished with non-tritiated water flushes
and air purges of the lines. For the tritium analyzer, various alcohol solvents were tested with methanol
found to produce the best cleaning results. Three washes of pure methanol, followed by three washes of
non-tritiated water serve to completely eliminate any "memory" in the system. Bubbling air through the
cell was found to be a necessary step in the cleaning operation.
TESTING AND EVALUATION
The remote testing of the tritium analyzer capabilities was first conducted in the CAIS
laboratory in March 1996. The remote computer was set up in one laboratory and the tritium analyzer
and local computer, which included the valve control interface, were located in a separate laboratory.
Using a telephone/modem linkage, the sample collection/analysis/cleaning cycle was successfully
operated from the remote computer. A background of 1.35 cpm and a counting efficiency of 24% were
demonstrated in this laboratory setting. These data yield a minimum detectable activity (MDA) of
7.2 Bq/L at the 95% confidence level for a 100-minute count via the method of Currie (5).
A preliminary field test was conducted at the SRS in May 1996 which was similarly successful.
The tritium analyzer was placed in a small trailer equipped with a motor generator and located near L-
Lake, a man-made lake constructed in 1985 to receive the heated secondary coolant from one of the five
production reactors at the SRS. The L-Lake sampling location is about 17 km from the central
Here’s what’s next.
This report can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Report.
Cable, P.R.; Hofstetter, K.J.; Beals, D.M.; Jones, J.D.; Collins, S.L.; Noakes, J.E. et al. A remotely operated, field deployable tritium analysis system for surface and groundwater measurement, report, December 31, 1996; Aiken, South Carolina. (https://digital.library.unt.edu/ark:/67531/metadc676402/m1/7/: accessed May 26, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.