Isotopic fractionation associated to nitrate attenuation by ferrous iron containing minerals

. Biotic and abiotic laboratory experiments of nitrate and nitrite reduction by Fe-containing minerals were performed and the isotopic fractionation of the different reactions was calculated in order to determine whether it is possible to distinguish biotic and abiotic reactions involving N compounds. Results of biotic experiments showed nitrate reduction up to 96 % with transient NO 2- accumulation and no significant N 2 O production. No significant nitrate attenuation was observed in abiotic nitrate reduction experiments. Abiotic experiments of nitrite reduction showed a rapid decrease in nitrite concentrations in those experiments with added Fe 2+ coupled with a significant N 2 O production. Preliminary results of the N and O isotopic fractionation of the biotic experiments of nitrate reduction show differences in the ε 15 N NO3 and ε 18 O NO3 when different minerals were added. The abiotic experiments of nitrite reduction contrarily, showed similar ε 15 N NO2 in all the experiments.


Introduction
During the last years, numerous studies have pointed to the coexistence of biotic and abiotic reactions involving the N and Fe biogeochemical cycles in natural environments [1,2]. Among the abiotic reactions, the NO2reduction by Fe 2+ oxidation has been well documented [3,4]. On the other hand, Fe 2+ minerals have been suggested to enhance biotic denitrification through the NO3dependent Fe 2+ oxidation (NDFO) process, which is still not well constrained [5]. As the enzymatic Fe 2+ oxidation by NDFO microorganisms has not yet been proven, it is unknown whether the Fe 2+ oxidation is caused by an enzymatic activity, due to an abiotic reaction mediated by N intermediates from denitrification or a combination of both [1,5,6].
To trace N transformation processes, chemical and isotopic characterization have proved to be powerful tools. The enzymatic NO3reduction provokes an enrichment in the heavy isotopes 15 N and 18 O of the unreacted substrate [7], unlike processes such as dilution that could lead to a concentration decrease without influencing the isotopic signature. The same pattern is expected throughout the reduction of all N intermediate products (e.g. NO2 -or N2O), which will be initially depleted in 15 N and 18 O with respect to the substrate. Information on the dual isotope systematics of NO2and N2O throughout the abiotic reduction by Fe 2+ is still scarce [4,8,9]. Therefore, it is unknown in which extent the isotopic characterization of NO3 -, NO2and N2O might help in distinguishing biotic and abiotic reactions involving the N compounds.
The aim of this work was to determine at laboratory-scale the N and O isotopic fractionation (ε 15 N and ε 18 O, respectively) during biotic and abiotic NO3and/or NO2reduction by Fe 2+ containing minerals. The selected Fe 2+ -containing minerals were magnetite (Mag), siderite (Sd) and olivine (Ol), and Mag was also tested nano-sized (Mag-NP) to quantify changes in reactivity. Special attention was directed on the generation of by-products (NO2and N2O) throughout the biotic process. Furthermore, the possible abiotic reactivity between the Fe 2+ -containing minerals and NO3or NO2and the possible N2O generation through both the biotic and abiotic pathways were evaluated.

Methods
To accomplish the goal, five series of experiments were performed ( Table 1). All series included replicates, a control without mineral and a blank (groundwater or synthetic water alone). The microcosms were set inside a glove box, using 20 mL serum bottles, crimp sealed with butyl rubber stoppers under an Ar headspace. Incubations were performed at 23 ºC and constant shaking in the darkness. The microcosms were sacrificed by turns at time intervals depending on denitrification dynamics. Control and blank microcosms were sacrificed at the end of the experiment.  All samples were filtered through 0.2 µm Millipore® immediately when obtained and stored at 4 ºC until analysis. Chemical parameters were determined by standard analytical techniques. The δ 15 N-NO3 -, δ 18 O-NO3 -, δ 15 N-NO2and δ 18 O-NO2were determined following the cadmium and azide reduction method [10,11]. N2O was analyzed using a Pre-Con (Thermo Scientific) coupled to a Finnigan MAT 253 Isotope Ratio Mass Spectrometer (IRMS, Thermo Scientific). Notation is expressed in terms of δ ‰ relative to the international standards AIR (Atmospheric N2) for δ 15 N and V-SMOW (Vienna Standard Mean Oceanic Water) for δ 18 O.

Results and discussion
Biotic experiments (BioSedGw and BioSedSw) showed a decrease in NO3concentration during the first week of incubation (Figure 1), up to 42 % in the groundwater batch experiments and up to 64 %, in the synthetic water batch experiments. Attenuation was also observed in the control microcosm containing groundwater plus sediment with no mineral addition (BioSedGw-C up to 45 %). Transient NO2accumulation (not shown) was much higher in the microcosms containing synthetic water (up 50 % of the initial NO3 -) compared to groundwater (up to 10 % of the initial NO3 -). After the first week, no significant differences were observed in the microcosms except for the batch containing nano-sized Mag, where about 96 % of NO3reduction was achieved in 91 days (Figure 1). No significant NH4 + concentration was detected, discarding the occurrence of Dissimilatory Nitrate Reduction to Ammonium (DNRA). Headspace N2O was below 1 % of the total reduced N, suggesting that the final gaseous product was N2. Abiotic experiments of NO3reduction (AbFeNO3) showed no significant decrease in NO3concentration after 220 days (results not shown). With regards to abiotic experiments of NO2reduction also no significant reduction was observed in microcosms containing synthetic groundwater and micro-sized minerals (AbNO2). However, a rapid NO2reduction was observed in the experiments in which microcosms contained synthetic groundwater with NO2and micro-sized minerals plus dissolved Fe 2+ (AbFeNO2, Figure 2). The beginning of the reaction seemed to be immediate, NO2complete removal was achieved in microcosms containing mineral plus dissolved Fe 2+ or dissolved Fe 2+ alone. An increased reduction rate (~ 50 hours) was observed in the microcosms containing micro-Sd compared to the others (~ 175 hours). A possible explanation could be an increased dissolution rate of Sd that increased dissolved Fe 2+ availability. N2O accumulated at the headspace of the AbFeNO2 microcosms, mass balance results suggested that N2O was the end product, other authors [8] previously observed a large N2O accumulation during NO2abiotic reduction by Fe 2+ oxidation. Abiotic experiments of nitrite reduction (AbFeNO2) showed a similar isotopic fractionation in all the batch experiments, with an ε 15 NNO2 between -14.1 ‰ and -17.8 ‰, and no oxygen fractionation (data not shown).