1	Filtron Water Filter: Disinfection Properties of Colloidal Silver Friday, August 6^th Heather Wright 4^th Year Undergraduate, Michigan Technological University Advisors: Angela Bielefeldt Scott Summers
2	Outline Background Objectives Methodology Results Conclusions Future Work
3	Background
4	Background Ceramic filter made locally from clay and sawdust with a colloidal silver coating Works by mechanical filtering and potentially bacterial inactivation by the colloidal silver Sole criterion that filter must meet: filtration rate between 1-2 L/hr Silver concentration applied to filter less than amount that causes argyria (deposition of silver salts in the skin from excessive oral intake)
5	Objectives Determine how certain environmental conditions impact disinfection of colloidal silver pH Silver concentration Determine inactivation ability of colloidal silver compared to silver ions and “small” colloidal silver
6	Methodology Effectiveness of colloidal silver tested using fecal coliform in solution with de-chlorinated tap water Bacteria started from Boulder wastewater, grown in media at elevated temperature and absorbance measured using a spectrophotometer to roughly determine concentration Different parameters (pH or silver concentration) were varied for each experiment Parameters selected to represent conditions that have been reported for Filtron field users
7	Methodology Performed batch tests Quantified fecal bacteria concentrations using the most-probable number (MPN) technique First two sets of experiments used 6 time points, last set used 2 time points More time points to calculate an inactivation rate constant Each time point had 3 dilutions (1,000x, 10,000x, and 100,000x) Each dilution had 5 replicates
8	Methodology A-1 broth used for direct isolation of fecal coliforms (Std Methods) Once inoculated, vials are incubated for 3 hours at 35^oC than transferred to 44.5^oC incubator for 21 hours Gas production in any vial indicates a positive result and the presence of fecal coliforms Bacterial growth or turbidity typically counted as positive result MPN statistically calculated from the number of positive vials
9	Test Conditions
10	Results: 1^st batch tests
11	Results: 1^st batch tests First set of experiments typically had minimal downward trend of decreasing bacteria concentration with time Generally low R² values (best example shown) Colloidal silver stock solution about half the expected concentration Measured by ICP Mass Spectroscopy Method, Geology Lab Evidence that colloidal silver not a strong bactericide Altered procedure of experiments
12	Results: 2^nd batch tests Second set of experiments: Increased test time, but experiments failed Problem of inconsistent initial bacteria concentration continued Bacteria stock left over weekend and fecal coliform food source used up, so bacteria were no longer alive and/or other bacteria took over Stopped tests and designed different experiment to test colloidal silver against other silver forms
13	Methods: Silver comparisons Batch tests with 4 different treatments used, all at 50 ug/L: Colloidal silver Silver ion (from silver nitrate) Colloidal silver permeable through 0.22 µm membrane (likely includes both ions and small colloids) Control (zero silver) All 4 treatments run at same time to achieve similar initial bacteria concentrations 4 replicates of the test conducted on different dates MPN technique used with two time points
14	Silver Colloidal silver concentration was 55% of expected, so manufacturer reported value of 3.2% was in error Calculated silver ion concentration vs measured If measured lower than calculated, indicates silver sorption to glassware, etc ~ 26% sorption of colloidal silver and 57% ion sorption indicated 77 % of the colloidal silver was <0.22 um in size
15	Example Results: Silver comparisons
16	Results: Silver comparisons
17	Conclusions Colloidal silver least effective silver treatment The inactivation of bacteria in the filtron may be due to: silver ions dissolved into the water inside the filtron (and longer contact time for the water) Future tests should explore higher silver concentrations, longer contact times, and the filtron ceramic itself with and without silver in column tests
18	Future Work Conduct more batch tests using four treatment methods, varying silver concentrations Ensure rigorous glassware washing between experiments and measure aqueous silver concentrations at the end of each experiment Perform column tests to determine if Filtron behaves as predicted by the batch tests Membrane Filter method for fecal coliform bacteria Using total coliform bacteria instead of fecal coliform bacteria
19	Questions or Comments?