Graduation Year


Document Type




Degree Name

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

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Sarina J. Ergas, Ph.D.

Co-Major Professor

Kebreab Ghebremichael, Ph.D.

Committee Member

Katherine Alfredo, Ph.D.


Biochar, Drinking Water Treatment, Full-Scale, Point of Use, Pumice


In 2010, the UN formally stated that access to safe drinking water and sanitation is a human right. Yet, one in three people globally do not have access to safe drinking water. Contaminated drinking water can negatively impact the health of consumers. This is a challenge that disproportionally affects developing regions, where limited access to resources exacerbates the problem. One drinking water contaminant of concern is fluoride. Fluoride is often added in low concentrations to drinking water to enhance consumers’ dental health. However, when water with high concentrations of fluoride is consumed frequently, it can have negative health effects, such as dental and skeletal fluorosis.

Biosand filters (BSFs) are a point of use drinking water treatment system that is widely used in developing regions. BSFs utilize both mechanical and biological filtration to achieve excellent removal of bacteria and organic matter. Fluoride, however, is not removed effectively by BSFs. The goal of this research was to modify the already successful technology of the BSF with a new design, filter media, and media coating to make it able to remove fluoride.

Along with the traditional sand, pumice and biochar were used as BSF media. Batch fluoride adsorption experiments were carried out on aluminum oxide coated sand, pumice, and biochar. Coated sand removed 26% of fluoride, while coated pumice removed 54.4%. Neither sand nor pumice were able to achieve fluoride concentrations below the World Health Organization (WHO) limit of safe consumption (<1.5 mg/L). Aluminum oxide coated biochar, on the other hand, achieved concentrations well below the WHO limit and removed 97% of fluoride.

The new BSF design consisted of a two-stage system, where the first stage provided removal of organics and suspended solids and the second stage modified for fluoride removal. Sand and pumice were coated with aluminum oxide, a known adsorbent of fluoride. Five BSFs were constructed in the University of South Florida (USF) Botanical Gardens. Three contained uncoated sand, pumice, or biochar, in both the first and second stage. The remaining two filters contained uncoated media in the first stage and coated media in the second stage for sand and pumice. A new in-situ coating method was tested to coat the pumice and sand, resulting in the production of aluminum oxide gel in the filters.

The BSFs were operated five days per week for 38 days. Turbidity, UV254, flow rate, and pH were measured daily while E. coli and aluminum concentrations were measured once per week. Fluoride was measured every other day. The coating method significantly reduced the flow rate of the sand from an average of 144 mL/min uncoated to an average of 83 mL/min coated, while the uncoated and coated pumice filters averaged 278 mL/min and 268 mL/min, respectively. The biochar filter and the uncoated sand filter had similar average flow rates of 152 mL/min and 144 mL/min, respectively. All filters had similar removal of turbidity (88-90%), except the coated pumice, which removed only 79%, possibly due to export of fine particles from the coating process. The product water of the coated sand and pumice filters had a higher pH than the influent water, likely due to the presence of aluminum. However, the pH has been showing a downward trend as of this writing. Effluent aluminum from the coated pumice filter was 0.66 mg/L during the first week of operation, then decreased to 0.07 mg/L in the third week. The coated sand showed a similar trend but had much lower concentrations of aluminum in the effluent. This evidence, along with the increased effluent turbidity and pH, showed that not all of the aluminum was effectively coated on the media, especially in the pumice filter.

The in-situ coating method consistently doubled the removal of UV254 in both the sand and pumice filters (60% and 62%, respectively). However, the biochar filter still achieved the best removal of UV254 (87%), displaying the high capacity of biochar to adsorb organic matter. As expected, the uncoated filters did not remove a significant amount of fluoride, while the aluminum oxide coated filters did. The coated sand filter initially achieved an effluent fluoride concentration of 0.86 mg/L, but had concentrations above the WHO limit of 1.5 mg/L after Day 10. The coated pumice filter showed a similar trend, but started at 0.61 mg/L and remained below the WHO limit until Day 31.