Page 340 - gas transport in porous media

Chapter 21: Gas Transport Issues in Landmine Detection
341
The Buried Chemical Model was also used to evaluate the effect of differing soil
properties, water flux conditions and sequences on the behavior of TNT, DNT and
DNB (Phelan and Webb, 1997; Phelan and Webb, 1998a; Phelan and Webb, 1998b;
Webb et al., 1998). The surface vapor flux was evaluated because this parameter was
considered to be the principal pathway for detection of buried landmines by dogs.
While the Buried Chemical Model was valuable for an initial assessment, the
assumptions of constant and uniform liquid content and temperature, as well as a
single boundary layer thickness, were obviously great simplifications. In order to
address these and other issues, a multidimensional mechanistic code was modified
for application to this problem. This code, which is based on the TOUGH code from
Lawrence Berkeley Laboratory (Pruess, 1987, 1991), considers air, water vapor, and
explosive chemical mass transport and heat flow in a porous media, and has been
named T2TNT (Webb et al., 1999).
Modifications to TOUGH2 to produce T2TNT included the following:
(1) chemical components for landmine signatures (TNT, DNT, and DNB),
(2) gas diffusion – gas diffusion can be an important transport mode for explosive
vapors in the subsurface,
(3) liquid diffusion – liquid diffusion can be a dominant transport mode for explo-
sive vapors in the subsurface, especially for moderate and high moisture content
conditions,
(4) liquid-solid sorption,
(5) vapor-solid sorption – vapor-solid sorption is significant for explosive vapors at
low soil moisture contents,
(6) biodegradation,
(7) surface boundary conditions – due to the shallow burial depth of many land-
mines, the fluid conditions surrounding the landmine are strongly influenced by
the surface conditions. The parameters necessary to adequately model the surface
boundary conditions include: solar and long-wave radiation, the surface bound-
ary layer that is a function of wind speed and soil-air temperature differences,
precipitation and evaporation at the surface, plants and their root systems, and
the diurnal and seasonal variation of these parameters.
To partially validate chemical transport in surface soil, an experimental soil column
was fitted with an atmospheric air plenum configured to collected vapor emissions
by solid phase micro extraction followed by chemical quantification with gas chro-
matography (Phelan et al., 2000). A liquid DNT source was injected into the soil and
upward water and solute gradients were established with low humidity air flowing
across the soil surface. Tests were performed to evaluate the effects of wetting and
drying phenomena on the vapor flux of DNT at low liquid saturation, which include
vapor-solid sorption phenomena (Phelan et al., 2001).
The data and model comparison for the surface flux of DNT is shown in Figure 21.2
that reflect the variation in soil column conditions. The initial relative humidity of the
air was ∼50%. At Day 35, the relative humidity was changed to 0%, which increased
the evaporation rate and the DNT vapor flux. At Day 44, a drying event was imposed
that dramatically lowered the soil saturation and the DNT vapor flux. Awetting event