Demonstration and Commercialization of ARoNite™, A Novel Hydrogen-based

Membrane Biofilm Biological Reduction Process

CA-NV AWWA Fall Conference Sacramento, California

October 3, 2013

David Friese, P.E. Technology Director, ARoNite Development

Co-authors:

Rob Hills, CVWD Ryan Overstreet, APTwater

Dr. Bruce Rittmann, Arizona State University

The use of naturally-occurring biological mechanisms in an engineered system to accomplish a water treatment goal Aerobic systems (oxidation) Removal of biodegradable organics for TOC reduction Examples: Riverbank filtration, unchlorinated media filters

Anaerobic/anoxic systems (reduction) Removal of “oxyanions” and halogenated compounds Nitrate, chromate (hexavalent chromium), perchlorate, selenate

(selenium), DBCP, TCE, PCE, TCP….

If it’s natural, why is it “new?” A growing realization that not all bacteria are bad

2

Biological Treatment 101

Denitrification is a dissimilatory biological process that sequentially reduces NO3

- to N2 (a 5-electron reduction). NO3

- + 2e- --> NO2-

+ e- --> NO + e- --> 0.5N2O + e- --> 0.5N2

We chose H2 gas as the electron donor H2 --> 2H+ + 2e-

(2 electrons per H2 molecule)

3

What is Denitrification Using Hydrogen?

• Anoxia, or dissolved oxygen removed

• Excellent biomass retention, usually as a biofilm

• Satisfactory pH, near neutral

• Availability of nutrients, particularly P

• Sufficient electron donor – Organic compound (for heterotrophs)

– Hydrogen gas (H2) (for autotrophs)

What are the proper conditions?

4

Major advantages over organic e- donors (heterotrophic) • H2 supports autotrophic bacteria, which use bicarbonate as

their Carbon source; thus, no organic Carbon source must be added.

• Relatively low-cost electron donor that is commercially available as a bulk chemical or can be generated on-site.

• No dosing challenge, easy to control • Non-toxic, no residual • Low biomass yield, producing less biomass that must be

wasted. • Based on growing microbiology research, H2 should be

available as donor for virtually all oxidized contaminants.

5

Why Bacterial Reduction Using Hydrogen?

A novel natural partnership between technology and biofilms. For the first time, we can deliver H2

to microorganisms as the electron donor in a safe and efficient manner. o Gaining all those advantages discussed in

previous slide!

6

What is at the core of the process?

• Membrane Biofilm Reactor Concept - ~ 1996 – by Dr. Bruce Rittmann & Dr. K-C Lee • Demonstration for denitrification (late 1990s) –Dr. Rittmann & Dr. Lee • First publication (2000); now > 40 publications • Demonstration for perchlorate reduction (early 2000s) -- Dr. Rittmann & Dr. Lee • Field demonstration (early 2000s) – w/ MWH • First patent (2002) • APTwater licenses the technology and partners (2004) • Since 2005 APTwater has been expanding the scope of oxidized contaminants,

conducting field testing, developing commercial system • The concept was awarded the 2011 Environmental Engineering Excellence Award from

the American Academy of Environmental Engineers! • ARoNite received NSF 61 certification in Dec 2011 • The first commercial ARoNite system went on line in Jan 2012 • CDPH testing and review - 2012

• ARoNite will be the one of the first permitted and operating biological reduction process in the US for drinking water treatment

7

ARoNite (MBfR) Brief History

ARo Technologies – Autotrophic – an organism capable of making

organic molecules from inorganic sources. Examples are plants, algae, some bacteria

Reduction – Chemical reaction in which an electron is gained

of - Nitrates, Chromium, Selenium and other compounds

Targeting Autotrophic Reduction of Nitrates to innocuous compounds, using hydrogen, carbon dioxide

and electricity as consumables. In pilot and field testing are:

•ARoChrome targeting Cr6+ in water supplies •ARoPerc targeting specific chlorinated compounds in water i.e. ClO4

In Development: •ARoSel targeting Selenium, naturally occurring and from flue gas desulfurization, oil production and refining

What does ARoNite stand for?

8

o Hydrogen is either generated or delivered on site

o Carbon Dioxide is delivered on site and used to control pH

o Carbon Dioxide & Hydrogen are fed to the ARoNite Reactor creating the optimum environment for the biofilm

o Contaminated water is fed to the ARo Reactor allowing the biofilm to grow and reduce the target contaminant from the water

o Treated water is filtered and disinfected using conventional technologies before discharge to distribution system

H2 Onsite Generation

APT

Reac

tor

Water Source

Treated Water

Hydrogen

9

How the ARoNite System Works

Carbon Dioxide

Polypropylene Hollow Fiber Membrane

Biofilm consumes H2 on demand from hollow fiber surface, driving more

diffusion 30 µm

10

100- 300 µm

Water O2, NO3

-, ClO4-

Biofilm supported on a bubble-less gas-transfer fiber

How the ARoNite Reactor Works

• Naturally occurring biology grows on the surface of a gas diffusion membrane sheet woven of a proprietary hollow fibers

• Several of these sheets are spiral wound around a water feed tube with a water channel spaced between the sheets

• Hydrogen and Carbon Dioxide are fed to the lumen of the hollow fibers

• Water passes between the sheets making contact with the biofilm

• Nitrates are reduced to Nitrogen and Water

• Over time, cells fall off into water and are filtered

• Automated purge cycle prevents fouling

11

12

What does ARo Technologies stand for? View of the modules in operation

Location Contaminant Dates System configuration Significant Outcome

La Puente, CA Groundwater ClO4- ~2003 Pilot module AwwaRF Report by MWH, early system

design info.

Modesto (Grayson) Groundwater NO3

- 9/06 – 6/11 Pilot module Multiple fiber and module construction improvements, CDPH data collected

Lake Arrowhead Tertiary effluent NO3- 3/07-11/07 Pilot module WateReuse Report by Trussell

San Bernardino Groundwater NO3- ,

ClO4- 3/08 - 1/09 Pilot module Flow maldistribution limit performance,

ESTCP with CDM reauthorized

Glendale, AZ Groundwater NO3- 4/08 - 2/09 Pilot module

WRF Report by CH2MHill, positive comparison to IX and heterotrophic systems

Rancho Cordova Groundwater ClO4- 9/08 – 11/10 Pilot module Successfully treat 14 ppm to <4 ppb

Rancho Cordova Groundwater ClO4- 10/08 – 11/10 Commercial module Develop and test larger modules

Ojai, CA Tertiary effluent NO3- 2/10 – 12/10 Commercial module Tested multitude of large modules in one

system

Rialto, CA Groundwater NO3- ,

ClO4- 5/11 – 2/12 Commercial module ESTCP project with CDM based on

improvements in commercial module

Burbank, CA Groundwater NO3- ,

Cr(VI) 6/11 - 11/12 Commercial module

Tested low ppb Cr(VI) removal

Rancho Cucamonga Groundwater NO3

- 11/11 - current Commercial module 1St commercial, approved and operating system for biological nitrate removal

13

ARoNite Projects – Several Years in Development

• Sampling and analysis protocol developed w/CDPH for continuous 30-day test

• October 29, 2008, to December 1, 2008

• Positive results on all analyses – met DW standards

Analyte averages Inlet Final Product

Nitrate-N (mgN/L) 14.1 0.3

Nitrite-N (mgN/L) <0.01 <0.01

Chromium (μg/L) 18.3 3.5

Selenium (μg/L) 10.5 6.5

Copper (μg/L) 9.5 <2

14

ARoNite Case Study: City of Modesto

• All DW requirements met – APTwater encouraged to apply for operating permit on a specific

treatment system for approval

• Nitrate Conditional Acceptance of the treatment technology via Cucamonga Valley Water District project with supporting data from Modesto (Grayson) pilot

• Operating Permit approval – at a specific site

• Key CDPH requirements for post-treatment after any biological treatment process are:

– Turbidity <0.3 NTU

– Disinfection to 4-log virus deactivation/removal

15

CDPH Regulatory Outcome

ARoNite Case Study: Rancho Cucamonga (CVWD)

Drinking Water Well #23 Water-by-the-Gallon (DBOOM) Denitrification

• CVWD has 12 wells shut down out of 30 for nitrates • All high in nitrate, well over MCL • Target allowed nitrate <31 ppm • Two step contract

– Step 1: Start up and run 125 gpm – Step 2: Expand plant to 650 gpm,

• Plant construction; Aug-October, started up Nov 2011 • CDPH testing protocols defined and completed

Project Details:

16

ARoNite Installation Cucamonga Valley Water District

17

Show Consistent Treatment to produce Potable Water Nitrate Removal (1 mg-N/L or 4.5 mg/L) and Nitrite (< 0.5 mg-N/L) Turbidity (<0.3 NTU) Disinfection (Residual Chlorine >0.3 and < 2.0 mg/L) e. coli < 2.2 MPN/100 ml HPC < 500 cfu/ml pH ≥ 6.5 and ≤ 8.5 Aeration – Restore Dissolved Oxygen THM Formation Potential < 64 µg/L (80% of MCL) HAA5 Formation Potential < 48 µg/L (80% of MCL) Reliability ≥ 95% uptime

Goals of ARoNite Demonstration

18

H2

ARo-Nite

CO2 AirMicro

Nutrients

Aeration

Air

Filtration

BackwashWater

NaOCl

DisinfectionTank

Well 23

Reservoir #3

SanitarySewer

Cl2 analyzer

Compliance Nitrate analyzer

Compliance turbidity analyzer

Process Nitrate analyzer to monitor several points within process

19

ARoNite Process Overview and Monitoring Points

ARoNite Demonstration

• System Installation Started September 2011 • System Installation Completed November 2011 • System startup and inoculation November to Jan. 2012 • 30 Day Demonstration Started January 25, 2012 • 30 Day Demonstration Completed on February 24, 2012 • Challenge Tests – March 2012 • Extended Demonstration Testing – October and December 2012

During Demonstration

• 187 Field Analysis conducted each week • On-line Measurements taken every minute • 137 Lab Analysis Taken and Analyzed each week.

Project Time Line:

20

21

Analyses Conducted

Nitrate-N Heterotrophic Plate Counts (CFU/ml)Nitrite-N Total ColiformAmmonia (mg/L) e.Coli (MPN/100 ml)TKN (mg/L) TSS (mg/L)Perchlorate (µg/L) TDSSulfate (mg/L) Total Settleable SolidsSulfide (mg/L) Turbidity (NTU)Total Phosphorus as P (mg/L) pH Alkalinity Group in CaCO3 units THM (µg/L)Calcium (mg/L) THMFP (µg/L)Iron (mg/L) HAA (µg/L)Magnesium (µg/L) HAAFP (µg/L)Potasium (mg/L) VOC Sodium (mg/L) Total Chlorine ResidualHeavy Metals Chlorine DemandTOC (mg/L)DOC (mg/L)

Nitrate-N Field Nitrite-N Field ARoNite (N as NOx)ARoNite TurbidityDissolve Oxygen (DO) -Field Temperature-Field ⁰CpH-Field ARoNite pHARoNite Residual ChlorineResidual Chlorine-Field

OFF-SITE LAB FIELD AND ONLINE

October 1, 2013 22

Nitrate as NO3 Influent v. Effluent - Rancho

Note values are as Nitrate, not N MCL – 45 as nitrate, 10 as N Product below 1 mg N/L (4.5 as nitrate)

23

Comparison Summary – Lab, On-Line and Field

• Good agreement of online nitrate sensor with lab data (IC)

• Poor agreement of field nitrate analyses (Hach Cd-reduction colorimeter) with lab and online data, variable results on same sample

• Good agreement field (Hach colorimeter) and lab nitrite

• Good agreement online, field, lab turbidity – after air bubble trap installed

• Good agreement online, field, lab chlorine residual

(Online instruments – all from Endress+Hauser)

24

Disinfection Final Product

Location MPN/100 ml Average High Low STDEV

SP1 Raw 1 <1.1 45 93 21 28SP3 Aeration Effluent 2 <1.1 >5,700 -- -- --SP4 Filtration Effluent 2 <1.1 2875 5600 1700 2031

SP5 Final Effluent 2 <1.1 9 32 < 1 14SP6 Backwash Effluent 1 <1.1 >5,700 -- -- --

1 5 Samples Taken 2 8 Samples Taken

HPC (CFU/ml)e. coli

25

TOC and DOC Results

Location Average High Low STDEV Average High Low STDEV Average High Low STDEVSP1 Raw 0.34 0.43 0.30 0.08 <0.3 -- -- 0.00 0.04 0.32 -0.40 0.30

SP2 Lag Effluent 0.57 0.69 0.45 0.09 0.92 1.20 0.52 0.26 -- -- -- --SP3 Aeration Effluent 0.60 0.74 0.46 0.10 0.94 1.40 0.50 0.36 -- -- -- --SP4 Filtration Effluent 0.52 0.63 0.37 0.13 0.54 0.74 0.37 0.16 -0.52 0.48 -1.60 0.69

SP5 Final Effluent 0.56 0.65 0.47 0.08 0.57 0.76 0.39 0.15 0.77 1.10 0.00 0.451 5 Samples Taken2 BDOC results are the difference between and initial and final DOC, both positive and negative values uses in summary statistics

TOC, DOC and BDOC 1

DOC (mg/L) TOC (mg/L) BDOC (mg/L) 2

• Approximately 0.2 mg/L increase in DOC/TOC across treatment train, consistent the Modesto/Grayson and other pilot data – from soluble microbial products (SMP).

• BDOC data highly variable and not consistent with BDOC being a subset of DOC, which is a subset of TOC.

26

Disinfection By-Products Initial/Product Comparison

Location Average High Low STDEV Average High Low STDEV

SP1 Raw <0.05 -- -- -- 2.35 2 2.60 2.10 1.38SP5 Final Effluent 2.2 3.5 1.2 0.82 23.20 30.00 14.00 5.89

1 5 Samples Taken 2 2 Samples Taken

Location Average High Low STDEV Average High Low STDEV

SP1 Raw <2 <2 <2 0.0 1 2 1.00 1.00 0.00SP5 Final Effluent <2 <2 <2 0.00 13.0 21.0 8.3 5.23

1 5 Samples Taken 2 2 Samples Taken

THM (µg/L) and THM Formation Potential (µg/L)1

THM (µg/L) THM Formation Potential (µg/L)

HAA5 (µg/L) and HAA5 Formation Potential (µg/L)1

HAA5 (µg/L) HAA5 Formation Potential (µg/L)

27

Challenge Testing

Challenge Tests Results: 1-hour H2 off, continue water feeding 1-hour power suddenly off 24-hour power suddenly off

All had instant nitrate reduction and back to normal after two hydraulic residence times

10-day “well outage”

Kept system “alive” on dissolved oxygen, recycled aerated water Instant nitrate reduction, approximately 5HRT back to normal

28

Miscellaneous Data

• One detection of Perchlorate at 4.5 µg/L in the Raw water , ND in Final Product

• One detection of Total Cr at 3 µg/L in the Raw water on Day 1, ND in Final Product

• One detection of Cu at 2.3 µg/L in Filtrate on Day 1, ND in Final Product

• Manganese was detected in the Raw Water 2.2 µg/L, ND in Final Product

• Dissolved Oxygen in Final Product averaged 6.6 ± 0.8 mg/L (Raw Water Ave. DO = 8.7)

• Temperature of the Final Product averaged 72.4 ± 1.3⁰F (Raw Water Temp 68.4±1.2)

All other Water Quality parameters were unchanged

29

First Demonstration Outcome

• Nitrate removal to target - the “easy part”

• CDPH requested below improvements and then a second, less extensive demonstration

Process control – lack of tuning – Fixed

Filtration system – easily upset, inconsistent performance

Residual chlorine control – Fixed, improved flow-paced dosing

30

Filtration System Improvements

• Filtration system – easily upset, inconsistent performance

• Entrained air in filter feed, bubble traps before turbidimeter

• Dose filter aid farther upstream (not enough mix/coag time)

• Jar and filter testing with alternate coagulants (found ACH/polymer best)

• Discovered leaking o-rings in filter valves (Kinetico Hydrus)

Underside of top filter valve

• O-ring to seal internal riser pipe

• 2/3rd of o-ring groove

(Not supposed to be loose)

31

So, now what….

• Extended Demonstration completed Dec. 2012. • System mothballed to await regulatory approval. • Report to CDPH submitted 2/4/13. • CDPH Water Treatment Committee reviewed and issued

Conditional Acceptance Letter 7/26/13. • Now installing filtration improvements to restart system. • Then will apply to local CDPH district for operating permit.

What to walk away with

• Utilizes naturally-occurring indigenous bacteria - no seeding required • Very low biomass compared to carbon based system • Adds no organic residual, no dosing challenges • Self regulated demand of hydrogen • Hydrogen can be generated on-site as needed • Meets and exceeds state and federal drinking water standards for drinking

water • NSF-61 certified • California DPH approval for drinking water applications • No brine or concentrated waste as Nitrate is destroyed versus concentrated • Remote monitoring, troubleshooting and control of equipment

Drinking Water

32

Acknowledgements

CVWD: Rob Hills, Ed Diggs, J.R. Rivas, operations team CDM Smith: Dave Cary, Jesse Aguilar, Luis Leon

APTwater: Steve Gagnon, Rich Buday, Ryan Overstreet, Reid Bowman

ASU: Bruce E. Rittmann, Aura Ontiveros-Valencia, He-Ping Zhang, Youneng Tang, Bi-O Kim

Thanks for your Time and Interest David Friese, P.E. dfriese@aptwater.com

• ARo system installed, no post aeration and filtration steps

• Ran 205 days on unspiked feed • Began Cr(VI) spike to feed, ran

248 days • No use of coagulant/filtration

steps that could remove additional Cr(III)

Analyte Inlet Final Product

Nitrate-N (mgN/L) 8-9 <0.3

Chromium(VI) (μg/L) 4-6 <1

Chromium total (μg/L) 4-6 3-4

Spiked Cr(VI) (μg/L) 15-25 <1

Chromium total (μg/L) 15-25 7-12

Spiked Cr(VI) (μg/L) 40-50 <2

Chromium total (μg/L) 40-50 15-20

34

Case Study: Burbank Cr(VI)

MBfR Publications Highlights Lee, K.-C. and B. E. Rittmann (2002). Applying a novel autohydrogenotrophic hollow-fiber membrane

biofilm reactor for denitrification of drinking water. Water Research 36: 2040-2052.

Nerenberg, R., B. E. Rittmann, and I. Najm (2002). Perchlorate reduction in a hydrogen-based membrane-biofilm reactor. J. Amer. Wat. Works Assn. 94(11): 103-114.

Cowman, J. C. Torres, and B. E. Rittmann (2005). Total nitrogen removal in an aerobic/anoxic membrane biofilm reactor system. Water Sci. Technol. 52(7): 115-120.

Chung, J., R. Nerenberg, and B. E. Rittmann (2006). Bio-reduction of selenate a hydrogen-based membrane biofilm reactor. Environ. Sci. Technol. 40: 1664-1671.

Chung, J., R. Nerenberg, C. Torres, and B. E. Rittmann (2006). Bio-reduction of soluble chromate using a hydrogen-based membrane biofilm reactor. Water Res. 40: 1634-1642.

Nerenberg, R., Y. Kawagoshi, and B. E. Rittmann (2006). Kinetics of an autotrophic, hydrogen-oxidizing perchlorate-reducing bacterium. Water Res. 40: 3290-3296.

Rittmann, B. E. (2007). The membrane biofilm reactor is a versatile platform for water and wastewater treatment. Environ. Engr. Res., 12(4): 157-175.

Chung, J., C.-H. Ahn, Z. Chen, and B. E. Rittmann (2008). Bio-reduction of N-nitrosodimethylamine (NDMA) using a hydrogen-gased membrane biofilm reactor. Chemosphere 70(3): 516-520.

Nerenberg, R., Y. Kawagoshi, and B. E. Rittmann (2008). Microbial ecology of a hydrogen-based membrane biofilm reactor reducing perchlorate in the presence of nitrate or oxygen. Water Research 42: 1151-1159.

Chung, J., R. Krajmalnik-Brown, and B. E. Rittmann (2008). Bio-reduction of trichloroethene using a hydrogen-based Membrane biofilm reactor. Environ. Sci. Technol. 42(2): 477-483.

Chung, J. and B. E. Rittmann (2008). Simultaneous bio-reduction of trichloroethene, trichloroethane, and chloroform using a hydrogen-based membrane biofilm reactor. Wat. Sci. Technol. 58(30): 495-501.

Ziv-El, M. C. and B. E. Rittmann (2009). Systematic evaluation of nitrate and perchlorate bioreduction kinetics in groundwater using a hydrogen-based membrane biofilm reactor. Water Research, 43: 173-181.

Ziv-El, M. and B. E. Rittmann (2009). Water-quality assessment after treatment in a membrane biofilm reactor. J. Amer. Water Works Assn. 101 (12): 77 – 83.

Tang, Y. N., M. Ziv-El, C. Zhou, J.-H. Shin, C.-H. Ahn, K. Meyer, D. Candalaria, P. Swaim, D. Friese, R. Overstreet, R. Scott, and B. E. Rittmann (2010). Bioreduction of nitrate in groundwater using a pilot-scale hydrogen-base membrane biofilm reactor. Frontiers Environ. Sci. Engr. 4: 280-285.

Zhang, H., M. Ziv-El, B. E. Rittmann, and R. Krajmalnik-Brown (2010). The effect of dechlorination and sulfate reduction on the microbial community structure in denitrifying membrane biofilm reactors. Environ. Sci. Technol. 44: 5159-5164.

Van Ginkel, S. W., C. Zhou, and B. E. Rittmann (2010). Hydrogen-based nitrate and selenate bioreductions in flue-gas desulfurization brine. J. Environ. Engr., in press.

Rittmann, B. E., Y. Tang, K. Meyer, W. D. Bellamy, and R. Nerenberg. Biological Processes (2011). Chpt. 17 in Water Treatment Plant Design, fifth ed., S. Randtke, ed., Amer. Water Works Assn., Denver, CO.

35

Peer-reviewed publications (MBfR in academia)