Dendroremediation of trinitrotoluene (TNT). Part 2: fate of radio-labelled TNT in trees.

BACKGROUND, AIM AND SCOPE: Problems of long-term existence of the environmental contaminant 2,4,6-trinitrotoluene (TNT) and necessities for the use of trees ('dendroremediation') in sustainable phytoremediation strategies for TNT are described in the first part of this paper. Aims of the second part are estimation of [14C]-TNT uptake, localisation of TNT-derived radioactivity in mature tree tissues, and the determination of the degree of TNT-degradation during dendroremediation processes.
METHODS: Four-year-old trees of hybrid willow (Salix spec., clone EW-20) and of Norway spruce (Picea abies) were cultivated in sand or ammunition plant soil (AP-soil) in wick supplied growth vessels. Trees were exposed to a single pulse application with water solved [U-14C]-TNT reaching a calculated initial concentration of 5.2 mg TNT per kg dry soil. Two months after application overall radioactivity and extractability of 14C were determined in sand/soil, roots, stem-wood, stem-bark, branches, leaves, needles, and Picea May sprouts. Root extracts were analysed by radio TLC.
RESULTS: 60 days after [14C]-TNT application, recovered 14C is accumulated in roots (70% for sand variants, 34% for AP-soil variant). 15-28% of 14C remained in sand and 61% in AP-soil. 3.3 to 14.4% of 14C were located in aboveground tree portions. Above-ground distribution of 14C differed considerably between the angiosperm Salix and the gymnosperm Picea. In Salix, nearly half of above-ground-14C was detected in bark-free wood, whereas in Picea older needles contained most of the above-ground-14C (54-69%). TNT was readily transformed in tree tissue. Approximately 80% of 14C was non-extractably bound in roots, stems, wood, and leaves or needles. Only quantitatively less important stem-bark of Salix and Picea and May shoots of Picea showed higher extraction yields (up to 56%). DISCUSSION: Pulse application of [14C]-TNT provided evidence for the first time that after TNT-exposure, in tree root extracts, no TNT and none of the known metabolites, mono-amino-dinitrotoluenes (ADNT), diaminonitrotoluenes (DANT), trinitrobenzene (TNB) and no dinitrotoluenes (DNTs) were present. Extractable portions of 14C were small and contained at least three unknown metabolites (or groups) for Salix. In Picea, four extractable metabolites (or groups) were detected, where only one metabolite (or group) seemed to be identical for Salix and Picea. All unknown extractables were of a very polar nature.
CONCLUSIONS: Results of complete TNT-transformation in trees explain some of our previous findings with 'cold analytics', where no TNT and no ADNT-metabolites could be found in tissues of TNT-exposed Salix and Populus clones. It is concluded that 'cold' tissue analysis of tree organs is not suited for quantitative success control of phytoremediation in situ. RECOMMENDATIONS AND OUTLOOK: Both short rotation Salicaceae trees and conifer forests possess a dendroremediation potential for TNT polluted soils. The degradation capacity and the large biomass of adult forest trees with their woody compartments of roots and stems may be utilized for detoxification of soil xenobiotics.