RD 83560201 ‐ 0 Incorporating Ferrate Oxidation into Small Drinking Water Systems David A. Reckhow, Yanjun Jiang, Joseph E. Goodwill, Joshua C. Cunningham, Xuyen Mai & John E. Tobiason University of Massachusetts Amherst, MA
Ferrate Basics • Decomposes in water �2 � 5 3 4 � 2 � 2�� � 2 � 2 � → ������ 3 � ��� 4 – Forming ferric hydroxide and oxygen • Must be produced on ‐ site by either – Electrochemical method 2 ��� � � 2�� � → ��� 4 �2 � 2 �� � � 3 3 5 – Wet oxidation method ������ 3 � 2 � 2 � • Becoming more economical • An oxidant and disinfectant – Many studies showing reaction rates for a wide range of organic and inorganic solutes in water • e.g., Sharma et al.; Lee & von Gunten • Won’t produce regulated DBPs – A good alternative for pre ‐ Cl 2 and maybe pre ‐ O 3
To be used in US Water Treatment • Ferrate must: – Be given disinfection (CT) credit by the EPA – Be cost competitive – Not interfere with other treatments used in plant – Offer some advantage over existing treatment technologies; examples: • Better removal of trace contaminants in raw water • Better control of organic and inorganic disinfection byproducts • Less energy consumption, carbon footprint, etc. • Easier to use, or more reliable • Improve aesthetic qualities of the product water
Summary of 2-log Disinfection @pH7 M ‐ Fe(VI)*min M ‐ Cl 2 *min [(mg ‐ Fe(VI)/L)*min] [(mg ‐ Cl 2 /L)*min] 3×10 ‐ 5 [1.9] †* 7.1×10 ‐ 7 [0.05] E. Coli ( 5ºC) 4.7×10 ‐ 5 [2.6] ‡ MS2 ( 5ºC) 2.5×10 ‐ 6 [0.18] 3.8×10 ‐ 4 [21] † Giardia (2 5ºC) 3.8×10 ‐ 4 [27] V. cholerae ( rugose ) 6.3×10 ‐ 5 [3.5] † 3.5×10 ‐ 6 [0.25] (20 ‐ 2 5ºC) † Current EPA study; * S35150 strain ‡ Hu et al. ES&T , 2012 4
Ferrate exposure (CT product) 80 25 M, pH 6.2 25 M, pH 7.5 60 50 M, pH 6.2 CT, (mg Fe*min)/L 50 M, pH 7.5 40 2 log inactivation Giardia 20 Cholera 0 Bolton Holton Houston Palmer ReadsboroStockbridge CT values at pH 6.2 are much lower than those at pH 7.5, pH had a great effect on CT values.
But Questions Remain �2 � 5 3 4 � 2 � 2�� � 2 � 2 � → ������ 3 � ��� 4 ? • Fe(+VI) goes to Fe(+III) • Intermediate products and oxidation states? – Fe(V), Fe(IV)? – Are they reactive too? “I think you should be more explicit here in step two”
Decomposition in phosphate buffer [Fe(VI)] 2 [Fe(V)] 2 [Fe(IV)] 2 H 2 O 2 H 2 O 2 H 2 O 2 O 2 H 2 O 2 O 2 H 2 O 2 H 2 O 2 HO 2 H 2 O 2 O 2 Fe(V) Fe(VI) Fe(IV) Fe(III) Fe(II) OH H 2 O 2 HO 2 O 2 Fe ‐ OH ‐ HPO 4 ‐ PO 4 • Mechanism contributed by many groups – Especially those led by Bielski and von Gunten
With phosphate and bromide [Fe(VI)] 2 [Fe(V)] 2 [Fe(IV)] 2 H 2 O 2 H 2 O 2 H 2 O 2 O 2 H 2 O 2 O 2 H 2 O 2 H 2 O 2 HO 2 H 2 O 2 O 2 Fe(V) Fe(VI) Fe(IV) Fe(III) Fe(II) OH H 2 O 2 HO 2 O 2 Br ‐ HOBr BrO 2 ‐ BrO 3 ‐ Fe ‐ OH ‐ HPO 4 ‐ PO 4 O 2 H 2 O 2 Formation is suppressed due to high H 2 O 2
Without phosphate [Fe(VI)] 2 [Fe(V)] 2 [Fe(IV)] 2 H 2 O 2 H 2 O 2 H 2 O 2 O 2 H 2 O 2 O 2 H 2 O 2 H 2 O 2 HO 2 H 2 O 2 O 2 Fe(V) Fe(VI) Fe(IV) Fe(III) Fe(II) OH H 2 O 2 HO 2 O 2 Br ‐ HOBr BrO 2 ‐ BrO 3 ‐ O 2 H 2 O 2 H 2 O 2 O 2 Fe(V) + Fe(IV) Fresh Fe(OH) 3 precipitate
Typical water with NOM [Fe(VI)] 2 [Fe(V)] 2 [Fe(IV)] 2 H 2 O 2 H 2 O 2 H 2 O 2 O 2 H 2 O 2 O 2 H 2 O 2 H 2 O 2 HO 2 H 2 O 2 O 2 Fe(V) Fe(VI) Fe(IV) Fe(III) Fe(II) OH H 2 O 2 HO 2 O 2 Br ‐ HOBr BrO 2 ‐ BrO 3 ‐ NOM NOM O 2 H 2 O 2 Fe ‐ NOM TOBr H 2 O 2 O 2 Fe(V) + Fe(IV) Fresh Fe(OH) 3 precipitate
All reactions [Fe(VI)] 2 [Fe(V)] 2 [Fe(IV)] 2 H 2 O 2 H 2 O 2 H 2 O 2 O 2 H 2 O 2 O 2 H 2 O 2 H 2 O 2 HO 2 H 2 O 2 O 2 Fe(V) Fe(VI) Fe(IV) Fe(III) Fe(II) OH H 2 O 2 HO 2 O 2 Br ‐ HOBr BrO 2 ‐ BrO 3 ‐ Fe ‐ OH ‐ HPO 4 ‐ NOM NOM PO 4 O 2 H 2 O 2 Fe ‐ NOM TOBr H 2 O 2 O 2 Fe(V) + Fe(IV) Fresh Fe(OH) 3 precipitate
How does Ferrate affect NOM reactivity with chlorine? DBP formation? • Test Protocol – Treat raw water samples with ferrate – Allow ferrate to dissipate (<60 min) – Chlorinate in the lab • pH 7 • Dose required for 1 mg/L residual after 72 hrs • 20 o C – Measure DBPs • Neutral Extractables (including THMs) • Haloacetic Acids (9 total)
Effects of Direct Ferrate Oxidation on Trihalomethane (THM) Formation Potential Substantial decrease; little pH effect Amherst, MA, pH 6.2 Amherst, MA, pH 7.5 1.2 Bolton, VT, pH 6.2 Bolton, VT, pH 7.5 Gloucester, MA, pH 6.2 1.0 Gloucester, MA, pH 7.5 Relative TTHM Formation Holton, KS, pH 6.2 Holton, KS, pH 7.5 0.8 Houston, TX, pH 6.2 Houston, TX, pH 7.5 Norwalk, CT (epi), pH 6.2 Norwalk, CT (epi), pH 7.5 0.6 Norwalk, CT (meso), pH 6.2 Norwalk, CT (meso), pH 7.5 Norwalk, CT (hypo), pH 6.2 0.4 Norwalk, CT (hypo), pH 7.5 Palmer, MA, pH 6.2 Palmer, MA, pH 7.5 0.2 Readsboro, MA, pH 6.2 Readsboro, MA, pH 7.5 South Deerfield, MA, pH 6.2 0.0 South Deerfield, MA, pH 7.5 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Stockbridge, MA, pH 6.2 Stockbridge, MA, pH 7.5 Ferrate Dose (mg Fe/mg C)
Comparison with ozone 1.2 Ferrate 1.0 Relative TTHM Formation 0.8 0.6 0.4 0.2 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Ferrate Dose (mg Fe/mg C) Data from: Reckhow et al., 1986 Data from: current study
Integration into water treatment I • Three pre ‐ ferrate scenarios – I: direct ferrate oxidation (e.g., groundwater) – III: part of conventional (e.g. surface water) • IIIA: ferrate & optimal coagulation • IIIB: ferrate & sub ‐ optimal coagulation Corrosion Control Fluoride Fe(VI) raw water Disinfectant Dist. Sys. Clear well Pre ‐ oxidation/ disinfection
Integration into water treatment II • Three pre ‐ ferrate scenarios – I: direct ferrate oxidation (e.g., groundwater) – III: part of conventional (e.g. surface water) • IIIA: ferrate & optimal coagulation • IIIB: ferrate & sub ‐ optimal coagulation – Reduce coagulant dose to account for prior addition of iron Corrosion Control Fluoride Fe(VI) Coagulant Disinfectant Dist. Sys. Clear well raw water rapid flocculation Settling Filtration Pre ‐ oxidation/ disinfection mix
The Intermediate Ferrate Scenario • Point of Addition – After clarification (settling) – Before Filtration Corrosion Control Fluoride Fe(VI) Coagulant Disinfectant Dist. Sys. Clear well Intermediate raw water rapid flocculation Settling Filtration oxidation/ mix disinfection
Intermediate Fe(VI) and THMs 1.2 Relative TTHM Formation • d Houston, pH 6.2 1.0 Compare with Houston, pH 7.5 Palmer, pH 6.2 Palmer, pH 7.5 Pre ‐ Fe(VI) Readsboro, pH 6.2 0.8 Readsboro, pH 7.5 Atkins, pH 6.2 Atkins, pH 7.5 Amherst, pH 6.2 0.6 Amherst, pH 7.5 Stockbridge, pH 6.2 1.1 Stockbridge, pH 7.5 0.4 1.0 South Deerfield Relative TTHM Formation 0 10 20 30 40 50 Norwalk Ferrate Dose ( M) Norwalk_50 ft 0.9 Babson Norwalk_5 ft 0.8 Intermediate 0.7 Fe(VI) 0.6 0.5 0.4 0 10 20 30 40 50 60 Ferrate Dose ( M)
Some Conclusions I Ferrate decomposition is more complicated than previously recognized. Natural waters have a stabilizing effect on ferrate. Some bromide oxidation occurs Phosphate suppresses decomposition and oxidation of Br Ferrate is capable of oxidizing regulated DBP precursors with an effectiveness similar to ozone. • At mass doses 1-2x those for ozone • Bromine incorporation is less with ferrate • Little bromate formation. • Exact nature of “effective” Fe oxidant is uncertain
Some Conclusions II When introduced at an intermediate stage, ferrate seems to be much more effective at destroying DBP precursors than when applied as a pre-oxidant Early data show ferrate to be effective at inactivating many bacteria, viruses and protozoans Ferrate in a pre-oxidant mode does not adversely affect filtration performance (filtered water turbidities, headloss buildup and filter run length) Ferrate seem to be an especially interested alternative for small systems that have water quality challenges
Acknowledgments • WINSSS Center • US EPA Star program • UMass water research group – Especially: Yun Yu, Sherrie Webb ‐ Yagodzinski, Arianne Bazilio • Personnel from participating Utilities – Amherst, Stockbridge, Palmer, Readsboro, etc. • Carole Tomlinson (Haskell Indian Nations Univ.)
The UMass Ferrate Group Dave Reckhow John Tobiason Yanjun Jiang Joe Goodwill Josh Cunningham Xuyen Mai
Universities of Massachusetts (Amherst), Texas (Austin), Nebraska, Florida, Illinois, South Florida, and Carollo Engineers RD 83560201 ‐ 0
Kinetics of Ferrate with contaminants • Prodigious literature – Sharma & others
Kinetic Analysis, high dose • 50 µM dose, Houston Water • Alkyl alcohols 1.0 • Alkyl amines 0.8 Fraction Remaining ethynlestradiol sulfamethoxazole 0.6 bromide Sulfide Nitrite Phenol 0.4 Analine 0.2 • sulfides 0.0 6.0 6.5 7.0 7.5 8.0 8.5 pH 25
Recommend
More recommend
