and then brought to 25% acid by adding 15C
RESULTS AND DISCUSSION
50% H2SO4 followed by sonicating again for six
hours. Hydrolysates were neutralized to pH 5.0
Solvent extraction
with Na2HPO4. The 3,5-DNA surrogate was added
The exhaustive, acetonitrile extractions of the
immediately before solid-phase extraction. An
time series showed that the concentration of
aqueous base solution with no added solids was
solvent-extractable (free) TNT was rapidly re-
also treated with the same procedure.
duced in compost treatments from 2370 mg/kg
on Day 1 to 28.6 mg/kg on Day 10, and to <5 mg/
kg by Day 20. These results agree with previous
Analysis for triaminotoluene
analyses of the composts (Caton et al. 1994,
in anaerobic cultures
One of the transformation products of TNT is
Pennington et al. 1995). ADNTs decreased with
triaminotoluene (TAT). It is known to exist in
time, while the DANTs increased for ten days be-
anaerobic systems (Preuss et al. 1993, Ahmad and
fore decreasing (Table 2, Fig. 1). Traces levels of
Roberts 1995); however, its instability and high
the azoxy transformation products of TNT and the
degree of polarity have made analysis very chal-
nitroso transformation products of RDX were
lenging. A novel separation using a CN column
found in a few of the samples. No hydroxyl-
with either an isocratic or gradient elution with a
aminodinitrotoluenes were detected.
When CaCl2/acetonitrile was decanted from
buffered eluent was developed and tested on
samples being extracted, about 10% of the solvent
bench-scale anaerobic cultures.
The isocratic separation was run on a CN col-
was retained by the residues. Each repeat decan-
umn (Supelco 4.6 mm 250 mm) at 1.5 mL/minute
tation with fresh solvent removed about 90% of
of 10 mM phosphate (pH 3.2)/acetonitrile (99 vol-
the previously retained analytes, but there was no
ume %/1 volume %). The retention time of the
evidence of further dissolution from the solids.
TAT was 7.4 minutes, well away from more polar
Thus, repeat decantations serve only to purify the
interferences that occur in digester solutions. TAT
residue, removing TNT, RDX, and "free" ADNTs
is extremely unstable and begins to degrade into
and DANTs. This finding agrees with earlier work
multiple compounds that disappear within an
from Myers.*
hour or two. Standards and samples have to be
The precision of acetonitrile extraction, based
prepared immediately before analysis. Getting
on percent relative standard deviation (% rsd) es-
solid TAT completely into solution requires an
timates for triplicate determinations showed no
acetonitrile/buffer mixture. The aqueous samples
clear dependence on either analyte or concentra-
tion. Values ranged from <1% to 73%. The high
from digesters must be mixed 50/50 with aceto-
value was an isolated case because the two next
nitrile before injection. A gradient elution that al-
highest values were 37% and 25%. Excluding the
lows the separation of TAT, both diaminoNTs, and
one high value and the several cases where <1%
the two aminoDNTs as a single merged peak was
was found, the remaining 42 estimates gave a
developed (Table 1).
pooled rsd of 11.9% with 84 degrees of freedom.
Acid hydrolysis
Table 1. Gradient elution, LC-CN conditions for
On the first day after initiation of composting,
triaminotoluene. Flow rate 1.5 mL/min.
there were already conjugated ADNTs and DANTs
that were released by the acid hydrolysis (Fig. 1).
Time
Milli-Q
ACN
Buffer
The quantities of conjugated transformation prod-
0
0
1
99
ucts increased for ten days, then gradually de-
2
99
1
0
creased to levels that were approximately equal
8.5
60
40
0
to, or slightly higher than, the Day-1 level. Samples
15
0
1
99
of this compost that had been stored in large plas-
Restart 20.0
tic bags at ambient temperature for several months
Retention time (minutes)
were also analyzed. After this aging, the quanti-
Triaminotoluene
4.4
2,6DANT
9
2,4DANT
10.8
*Personal communication, K.F. Myers, U.S. Army
2ADNT
12.9
Engineer Waterways Experiment Station, Vicksburg,
4ADNT
13.1
Mississippi, 1995.
4
Previous Page
