Table 1. Concentration (g/mL) of 2,4-DNT and
Experiments were conducted to assess the effects
2,4,6-TNT in the headspace. (After Jenkins et al.
of soil type, moisture content, and ambient tem-
1974.)
perature on the rate of breakthrough and the
qualitative and quantitative nature of the vapor
Ft. Belvoir clay
Windsor sand
signatures. At the end of the experiments, the sur-
Day
2,4-DNT
2,4,6-TNT
2,4-DNT
2,4,6-TNT
face soil was sampled to relate the vapor signa-
0
<d*
<d
<d
<d
ture in the headspace to the explosives-related
1.2 1012
t†
4
<d
<d
signature present in the surface soil.
5.0 1012
6
<d
t
<d
1.4 1011
1.8 1012
7
<d
<d
2.1 1012
<d
10
N.D.**
N.D.
EXPERIMENTAL METHODS
3.1 1011
2.8 1012
12
t
t
1.3 1011
13
N.D.
N.D.
t
Soils used for experiment
1.4 1011
t
14
N.D.
N.D.
Three soils were selected for this study because
1.6 1011
18
N.D.
N.D.
t
8.0 1011
5.8 1013
of their textural differences. A natural silica sand
19
N.D.
N.D.
7.0 1013
2.6 1011
from Ottawa, Illinois (ASTM 1999), was selected
25
N.D.
N.D.
to represent coarse-grained material with little or
* Concentration was less than an estimated detection
limit of about 1 1013 g/mL.
no organic carbon content. This material was
used as received. The second soil selected was
† Trace.
** No data.
Hitchcock silt loam, which is a native soil at our
location, and represents a medium-textured soil.
Table 2. Analysis of a soil
The third soil selected, Fort Edward clay, repre-
cover (Ft. Belvoir clay) for
sents a fine-grained material. Both the silt loam
DNT and TNT content. (After
and clay soils were passed through a no. 40 sieve
Jenkins et al. 1974.)
(420 m) to remove some of the larger stable
aggregates. The approximate grain size distribu-
Analyte
Concentration in g/g of soil
tion for these soils is presented in Table 3.
6.2 106
2,6-DNT
The three soils were each studied at three dif-
8.0 106
2,5-DNT
ferent moisture levels: air dry, low moisture, and
1.4 10-4
2,4-DNT
high moisture (Table 3). To moisten the soils, a
9.3 105
2,4,6-TNT
layer of each was spread out in a large aluminum
pan and wetted with well water (from a deep res-
Objective
idential water well in Hartland, Vermont) using a
Our major objective was to conduct a more
hand pump sprayer. The pan was shaken to mix
detailed experiment to investigate the qualitative
the soil and more water was added. This proce-
and quantitative effects of soil barriers at various
dure was continued until the approximate
temperatures on the vapor signature from buried
desired moisture level was obtained. This process
military-grade TNT.
was duplicated with a new soil sample and an
The overall strategy for this study was to set up
additional amount of water was added to raise
small-scale experiments where we buried a small
the soil to a higher moisture level. The final mois-
quantity of military-grade TNT beneath soil in an
ture content of the soils was determined by
enclosed container and to analyze the headspace
weighing two subsamples of each moist soil,
vapor periodically to determine the rate of break-
allowing them to air dry, then weighing again,
through of the various explosives-related vapors.
and obtaining the moisture by difference.
Table 3. Physical characteristics of soils used in explosives vapor transport
study.
Bulk density
(g/cm3)
Approximate grain size
Moisture based on
distribution (%)
wet weight (%)
Low
High
Soil
Sand Silt Clay
Low level High level
Air dry moisture moisture
Ottawa sand
100
2.1
3.1
1.65
1.58
1.64
Hitchcock silt
18
73
9
5.8
10
1.14
1.25
1.28
Fort Edwards clay
5
11
84
15
33
1.09
1.11
1.29
2
To Contents
Previous Page
