Forensic Applications for 1,4-Dioxane and Solvent Stabilizers - Environmental Investigation and Remediation - page 419

The interpolation analysis is made difi cult by the heterogeneity of the subsurface. The percentages
and interconnected porosity of gravels, sands, silts, and clays can be estimated from soil borings;
however, the solution to this inverse problem lies more in the domain of probabilistic conditional
simulation than in the more available “back of the envelope” approach using spreadsheet software.
Once an estimate of percentages of different geologic materials and their respective porosities
and permeabilities is made, the investigator must also infer the propensity for the solvent to migrate
into these media. The ability of dissolved solvent to migrate into the pore space of the analyzed soil
core depends not only on advective l ow and diffusion, but also on sorption of the solvent to soil
organic matter and mineral surfaces and its abiotic and biologic breakdown.
Where 1,4-dioxane is present, the uncertainty due to sorption and abiotic and biologic transfor-
mation is reduced because 1,4-dioxane is presumed to be resistant to these fates. If the starting
concentration of 1,4-dioxane in the release is known or estimated, the 1,4-dioxane concentration
measured in down-gradient wells can be useful for inferring the mass released and the mass remain-
ing in the source area. For example, 1,4-dioxane could be used to establish an order-of-magnitude
estimate to the upper limit of the volume of the subsurface contaminated by the release, as 1,4-
dioxane is expected to migrate farthest among the contaminants released at facilities using methyl
chloroform for vapor degreasing.
The range of 1,4-dioxane concentrations present in the solvent wastes released at a vapor degreas-
ing site can be estimated from the vapor degreaser's operating history or from measurements made
at other facilities that used methyl chloroform for vapor degreasing in a similar fashion. As dis-
cussed in Section 1.2.7.1, the boiling-point difference between 1,4-dioxane and methyl chloroform
(101°C versus 78°C) causes 1,4-dioxane to become concentrated in vapor degreaser still bottoms.
Experiments have measured the 1,4-dioxane liquid-vapor partitioning factor at methyl chloroform's
boiling point. In the operating vapor degreaser, 27% of the 1,4-dioxane will partition to the vapor
phase, while 73% will remain in the liquid phase (Spencer and Archer, 1981). Through continued
use, a vapor degreaser iteratively partitions 1,4-dioxane such that the proportion of 1,4-dioxane in
the sump will increase over time. In several weeks of daily use, the composition of the liquid solvent
can evolve to contain as much as 10-20% of 1,4-dioxane. For example, the measured 1,4-dioxane
concentration accumulated in the solvent sump in a laboratory vapor degreaser used for 24 days was
7.5%, whereas the originally supplied methyl chloroform had been stabilized with only 2.8% 1,4-
dioxane (Spencer and Archer, 1981). Laboratory analysis of new and spent methyl chloroform sam-
pled from a vapor degreaser at the Hayes International Corporation showed a 68% increase in
1,4-dioxane concentration, from 1.7% to 2.9% (by mass) (Tarrer et al., 1989).
It is possible to estimate the probable concentration of 1,4-dioxane in vapor degreaser still
bottoms if the operating practices are known from records or personnel interviews. For example, a
Model MLW-120 stainless steel Baron-Blakeslee vapor-immersion degreaser unit has a 45.4 L
solvent-sump capacity. In this estimation, 45.4 L of methyl chloroform solvent with 3% 1,4-dioxane
i lls the sump, and 73% of the 1,4-dioxane remains in the liquid phase. Moreover, once in each i ve-
day work week, 6 L of makeup solvent was added to the boiling sump. Solvent vapor carryout,
1,4-dioxane removal in the water trap (see Section 1.1.1.3), and accumulation of dirt and oils in the
sump are ignored. The calculations using these assumptions indicate that the 1,4-dioxane content of
the solvent waste could build up to 10% by mass within 20 weeks.
Calculating the 1,4-dioxane content of still bottoms in this manner produces lower estimates
than the measured data, most likely because of the signii cant role that vapor drag-out and liquid sol-
vent carryout plays in removing about three times more solvent from the degreaser than 1,4-dioxane.
As discussed in Section 1.1.1.3, solvent drag-out rates have been estimated at 1-2 gallons of solvent
per ton of small parts cleaned when a covered, in-line conveyor system is used, whereas small parts
cleaned at a similar rate in manually operated open-top degreasers lost 20% more of the solvent
through vapor drag-out (ASTM, 1962). The average annual solvent losses of three methyl chloro-
form vapor degreasing operations at southeastern U.S. military installations (summarized in Table
1.8) was 65%.