H2O SystemsInitial Prototype

Paulo JacobJennifer LiangJonathan TejadaAmi YamamotoJoy Yuan

Thursday April 6, 2006

Prototype Design Parameters

Biology

Water flow Electricity

Bacterial Culture PrepWater Sample PrepBacterial Quantification

Flow RateResidence Time

Breakdown of waterAnodizing TiCircuit Diagram

Prototype

Electrical Behavior• So far, we’ve shown that our proposed electric

field of 1-5 x 105 V/m is lower than the dielectricstrengths of water and air (100-300 x 105 and30 x 105 V/m respectively)

• Here we describe our efforts in preventingelectrolysis of water by insulating the electrodes.Also, we provide a comprehensive electricaldescription of our device through a circuit modelwhich guides our materials selection for thedevice.

Electrolysis of Water• Not to confuse with dielectric

breakdown potential of water,which pertains to the threshold atwhich “electron avalanches” occur.

Ref. Jones et. al. J. Appl. Phys. 77, 795, 1995• Electrolysis of water involves

electron transfer between watermolecules and electrically chargedmetal electrodes.

• Hydrogen and oxygen gas (thebubbles) are produced in thecathode (-) and anode (+)respectively. OH- and H+ are left inthe solution.

• GOAL: Prevent electrolysis ofwater

- +

+-

At the Cathode (-):2H2O + 2e- H2 +

2OH-

At the Anode (+)2OH- H2O + 1/2O2 + 2e-

Total Reaction:H2O H2 + 1/2O2

H2O +elecrolytes

Ti

Circuit Diagram

Wat

er

Ti

TiO2 or Polyester shimstock

Polyestershimstock

Ti

-

The device can be electrically represented by a circuit diagram in which one capacitoris in series with two other in parallel to oneanother.

Circuit Diagram• This treatment assumes the ideal

non-conducting behavior (noelectrolysis) and is useful toestimate the electric field andpotential in each component,especially in the water.

• κ is dielectric constant and ε ispermittivity, which are materialsproperties

• The values for the dielectricconstants are:

• TiO2; κ 1=14-170 or Polyester κ 1=κ 3 = 3.2

• H2O; κ 2= 80-

ε1, κ1ε3, κ3

ε2, κ2

Vapp

Ref. www.matweb.com

Solving the Circuit

-

C1

C3

C2

The equivalent capacitance for thiscircuit is:

Ceq=C1(C2 + C3)C1 + C2 + C3

Capacitance can be expressed as:

Ci = εiAd

, where ε0 = 8.854x10-12 C2 /Nm2 is the permittivityof free space, εi is permittivity of materialA is area and d distance

κi = εi / ε0

Q=CV

Vapp

Solving the Circuit

-

C1

C3

C2

Under these assumptions, thevoltage across each component is:

The electric field in each component canbe obtained through:

V1=

Vapp

Vapp (C2+C3)C1+C2+C3

V2= V3=VappC1

C1+C2+C3

But each capacitor modelsone material. The voltage dropcan be treated as linear

Ei=Vi/di

Solving the Circuit

-

C1

C3

C2

Therefore, the electric field acrosseach component can be expressedas:

E1=

Vapp

Vapp (C2+C3)(C1+C2+C3)d1

E2= E3=VappC1

(C1+C2+C3)d2,3

This enables us to predict theeffects of the insulating (electrolysisprotection) on the actual electricfield on water. As depicted earlier,capacitor 2 (or 3) stands for water. Ifthe capacitance of capacitor 1 is low(as in polymers) the field throughwater will decrease. Morequantitatively…

Quantitatively

1 cm

10 cm

5 cm

1 - Surface Area of insulating layer (TiO2 or polyester shimstock): 5x10-3 m2 , d~1x10-9 m forTiO2 or d= 12.5x10-6 m for shimstock.

2- Surface area of water layer: 3x10-3 m2 , d=127x10-6 m.

3- Surface area of shimstock window: 2x10-3 m2

d= 127x10-6 m.

Therefore, the capacitances are:1.1 - C1,TiO2 = 6.2x10-6 to 7.5x10-5 Fand C1,ST = 1.13x10-8 F

2.1 - C2,H2O = 2.8x10-8 F

3.1 - C3,ST= 1.12x10-9 F (Shimstock window)

Expected E-Fields• Therefore, the expected E-Field across the water (capacitor 2), for

the different insulating options under 25V are:– Using Shimstock: E2 = 359.23 V/m– Using TiO2: E2 = 1.9x105 to 23.8x105 V/m which falls in our targeted

range• There is a clear advantage in using anodized Ti under these

considerations, as it allows for lower voltages in order to obtain thelysing electric field. The changes in electrical field in each capacitoris caused by the distribution on potential between each capacitor,that in series must add to the total voltage. However, this distributiondepends on the dielectric constants, which can guide our materialsselection.

Insulating Attempts• As shown above, due to dielectric constraints, anodized Ti makes a

better suited insulator to minimize the electrolysis of water.

• Although the ohmeter measured TiO2 and polyester shimstockresistances cannot be resolved (too high), their calculatedresistances differ substantially. Through R=ρl/A, where ρ isresistivity, l length and A cross-secional area, we obtain:– TiO2 ρ=1011 to 1016 Ωm thus, RTiO2= 2MΩ to 2x105 MΩ– Polyester coating ρ=1013 Ωm thus R = 524.3 MΩ

• Therefore, anodized Ti provides both high resistance and gooddielectric properties. However, we observed poor prevention ofelectrolysis with anodized Ti when compared to polyester shimstockcovered electrodes.

Insulating the Electrodes:Anodized Ti

• Titanium anodization occurs byplacing a piece of titanium asthe anode in an electrolytic cell.TiO2 is formed on the surface ofthe sample, which providescorrosion resistance andelectrical insulation.

• Anodization took place with 30Vfor 1h. The current raised from~0.32A to 0.80A. The solutionemployed was 0.1M NaOH, withpH~13, as specified by SAEAMS 2488.

Color change observed in early stagesof anodization.

Insulating the Electrodes:Anodized Ti

• The resistances of theanodized layers andthe titanium weremeasured on thesurface by placing thevoltmeter probes onthe surface, separatedby 1cm

0.22Bare Ti

Resistance[Ω]Part

1013-1018 ΩcmResistivityof TiO2(anatase)

Not resolvedGrey region

50Blue layer

Ref. www.matweb.com

Insulating the Electrodes:Anodized Ti

Current Build Up in Anodized Ti Sample

-200

0

200

400

600

800

1000

1200

0 5 10 15 20 25

Applied Voltage (V)

Ob

served

Cu

rren

t (m

A)

The plot depicts the change in current as voltage was applied across the Ti electrodes, one of them anodized (the cathode in this case). Below we see the set up used.

A- +

- +

AnodizedElectrode

Water flow

TiO2 film thickness0nm ~200nm

Insulating the Electrodes:Anodized Ti

• Currently, we observe bubbles when running waterthrough the anodized device, which means electrontransfer occurs between water molecules and theelectrode plates.

• This can be due to the small films formed, because ofvoltage limitations

currentrange

Our Goal

Insulating the Electrodes:Anodized Ti

• Ideally, the sharp increase in current should beobserved at voltages higher than our targetoperating potential of 25V. This way electrolysisof water would be avoided.

• Power supply capable of higher voltages isnecessary, in order to assess the feasibility ofanodized titanium as means to preventelectrolysis of water.

• Possibility of TiO2 interacting with ions in water?

Insulating the Electrodes:Polyester Shimstock

• We also tested theperformance of theshimstock coating thesteel prototypeelectrodes underincreasing voltages.

• The shimstock proved tobe an efficient barrier forelectrolysis at thevoltages analysed:

• No bubbles wereobserved throughout theexperiment

Steel Coated with Polyester Shimstock

00.511.522.533.5

0 10 20 30 40

Voltage (V)

Cu

rren

t (

mA

)

Insulating the Electrodes:Summary

• As shown above, anodized Ti hasgreat potential as an insulating layer.However new attempts to createbetter films must be performed. Theuse of polyester shimstock has beenproven to be effective, however highervoltage supplies will be necessary inorder to create the target electric fieldthrough water.

Biology: Bacterial Culture

www.qiagen.com

Want bacteria to be inexponential growthphase duringexperiment

For every experimental run:1. Inoculate overnight culture day before2. Dilute o/n culture in fresh media

morning of run3. Prepare water sample from culture and

run experiment in afternoon

Biology: Water SampleTarget concentration ~ 272 cfu/ml

– Average E. coli concentration in Charles River– Concentration controlled by dilutions performed the

morning of the experiment

Stainless Steel system:– 40 ml total volume (20 ml/syringe)– prepare 45 ml sample = 9 ml culture + 36 ml water

Ti system:– 30 ml total volume– prepare 35 ml sample = 7 ml culture + 28 ml water

Biology: Bacterial Quantification

LB Amp plates1. Plate 100µL water sample

– Perform dilutionbeforehand if necessary

2. Allow 1 day to grow3. Count colonies

Spectrophotometer1. Transfer 1ml sample to

cuvette– Perform dilution

beforehand if necessary2. Measure OD6603. Calculate concentration by

Beer’s Law: A = εcl

• To be performed before and after each experimental run• 2 methods of bacterial concentration quantification:

– LB Amp plates– Spectrophotometer

• Decrease in bacterial concentration = indication of cell lysis

Experiment Set Up

Water In

Water Out

Power Supply

Weight

To control flow rate, will apply aconstant load » constant pressure

Water Flow: Controlling flow rateConstant Load

Constant PressureConstant Flow Rate

Flow Rate vs Applied Mass

y = 3.4586x - 2.9095R2 = 0.9593

0.00

1.00

2.00

3.00

4.00

5.00

6.00

0 0.5 1 1.5 2 2.5

Mass (kg)

Flo

w R

ate

(ml/s

)

Couette Flow

P = =3µLQ2Wd3Area

mg

Water Flow: Setting flow rate

Flow Rate [ml/s] = 3.46(Applied Mass [kg]) – 2.91Target Flow Rate = 1 liter/hour = 0.278 ml/s

Residence Time = Volumewater flow region / Flow RateTarget Residence Time >> 17 ms

Taking the target flow rate and required residencetime into consideration, we conclude that:

a ~1.5 kg weight will yield our desired results at a flowrate of ~2.3 ml/s and a residence time of ~134 ms

Initial PrototypeTesting procedure:1) Setup system design and attach

onto a stand.2) Prepare water sample for testing.3) Determine bacterial concentration

of pre-treatment water sample.4) Load water sample

(20mL/syringe in 2-syringe SSsystem; 30mL in 1-syringe Tisystem) into syringe(s) andremove air bubbles.

5) Clamp syringes onto stand tostabilize.

6) Set voltage at 25 V.7) Apply load onto syringe(s) and

allow entire volume of sample torun through system.

8) Determine bacterial concentrationof post-treatment water sample.

Gantt Chart

Modification

Testing

FinalPresentationPreparation

Construction

MaterialAcquisition

Design

Research

5/11-5/184/27-5/104/6-4/263/16-4/52/23-3/152/9-2/22

Questions?

h2o@mit.edu

Titanium Electrode Setup