Graduation Year


Document Type




Degree Name

MS in Environmental Engr. (M.S.E.V.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Jeffrey A. Cunningham, Ph.D.

Co-Major Professor

Sarina Ergas, Ph.D.

Committee Member

Matthew Pasek, Ph.D.


Microcosms, Pyrite, Small Community Drinking Water Systems, Sphalerite, Upflow Packed-Bed Reactors


Nitrate contamination in groundwater resources poses a serious threat, not only to the health of those who would drink it, but also as a potential nutrient for eutrophication. Current methods of nitrate removal can efficiently remove nitrate below 10 mg/L as nitrogen, but may require resources unavailable to small community water systems (SCWS). Sulfur-oxidizing biological denitrification is a promising alternative that may be more suitable for SCWS. Sulfur-oxidizing denitrification has previously been shown to have a high nitrate removal efficiency; however, it produces water high in sulfate. Utilization of sulfide-bearing minerals, rather than elemental sulfur as an electron donor, may yield similar nitrate removal efficiencies without the production of unwanted byproducts.

In this study, five minerals were examined through batch and column experiments to investigate suitable electron donors for autotrophic denitrification: sphalerite ((Zn,Fe)S), pyrite (FeS2), pyrrhotite (Fe1-xS2), iron (II) sulfide, and molybdenite (MoS2). Throughout the batch experiments, MoS2 performed poorly, removing less total inorganic nitrogen (TIN) than the non-inoculated control and was thus not considered for future tests. Fe1-xS2 tracked with the positive S0 pastille control during the first trial, though underperformed during the second trial and was therefore eliminated due to inconsistent performance. Iron(II) sulfide underwent dissimilatory nitrate reduction to ammonium (DNRA), effectively removing no TIN, and was thus not considered for the column experiments. Pyrite and sphalerite both had similar nitrogen species profiles throughout all trials and were therefore considered for the column experiment.

Two side-by-side upflow packed-bed reactors were operated over the course of 312 days, one with pyrite and pumice, and the other with sphalerite and pumice. Phases 1-3 of the experiment (Days 0-70) were operated as an acclimation phase where the procedure for the columns was altered to obtain consistent results. Phases 4-5were operated with an approximate HRT of 24 hours and target influent concentrations of 100 mg/L NO3--N, 10 mg/L NO2--N, 10 mg/L PO43--P, and 300 mg/L alkalinity as HCO3-. Throughout phase four (days 71-221), the pyrite and sphalerite column product waters had average nitrate concentrations of 48 mg/L and 43 mg/L respectively (as N). Though no significant (P>0.05) sulfate production was observed in the product water, 24 and 23 mg/L of sulfur oxyanions, i.e. sulfur oxidation intermediates, were produced in the pyrite and sphalerite columns, respectively. Additionally, the pyrite and sphalerite column product waters had average nitrite-nitrogen concentrations of 5.8 and 5.3 mg/L respectively, which greatly exceeds the 1 mg/L NO2--N primary drinking water standard. Phase 5 (days 222-312) of the column study included 4.1 mg/L of yeast extract in the feed stock in order to account for organic carbon required for denitrifiers for biosynthesis. During this phase, the average nitrate-nitrogen concentrations in the influent, pyrite product water, and sphalerite product water were 90 mg/L, 69 mg/L and 68 mg/L respectively. Furthermore, the average sulfate concentrations in the influent, pyrite product water, and sphalerite product water were 116 mg/L, 118 mg/L and 122 mg/L respectively. Additionally, the pyrite and sphalerite columns produced about 21 and 27 mg/L, respectively, of intermediate sulfur oxyanions. The low sulfate formation during phases four and five of the column experiment suggest that chemolithotrophic denitrification via sulfur oxidation was not the sole pathway for denitrification. Ultimately, pyrite and sphalerite did not achieve safe levels of nitrate or nitrite under the given reactor configuration, though improving surface reactivity may yield a greater removal efficiency.