brighton ed testing with asahi monovalent membranes
Post on 03-Jun-2018
219 Views
Preview:
TRANSCRIPT
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
1/41
BRIGHTON ED TESTING WITHASAHI MONOVALENT -SELECTIVE MEMBRANES
bYDave Williams
D N ENGINEERING
2108 S. Coors Cir.
Lakewood, Colorado 80228
Contract No. 1425-5-PG-81-08300Final Report
September 1995
Water Treatment Technology Report No. 14
U.S. Department of the Interior
Bureau of Reclamation _Denver Off ice
Technical Service Center
Environmental Resources Team
Water Treatment Engineering and Research Group
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
2/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
3/41
BRIGHTON ED TESTING WITHASAHI MONOVALENT -SELECTIVE MEMBRANES
-
Dave Williams
D N ENGINEERING2108 S. Coors Cir.
Lakewood, Colorado 80228
Contract No. 1425-5-PG-81-08300Final Report
September 1995
Water Treatment Technology Report No. 14
U.S. Department of the Interior
Bureau of Reclamation Denver Office
Technical Service Center
Environmental Resources Team
Water Treatment Engineering and Research Group
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
4/41
Bureazc of ReclamaiionMission Statement
The mission of the Bureau of Reclamation is to manage, develop, and protect
water and related resources in an environmentally and economically sound
manner in the interest of the American public. -
U.S. Department of the interiorMission Statement
Asthe Nations principal conservation agency, the Department of the Interiorhas responsibility for most of our nationally-owned public lands and natural
resources. This includes fostering sound use of our land and water resources;
protecting our fish, wildlife, and biological diversity; preserving the
environmental and cultural values of our national parks and historical places;and providing for the enjoyment of life through outdoor.recreation. TheDepartment assesses our energy and mineral resources and works to ensure
that their development is in the best interests of all people by encouraging
stewardship and citizen participation in their care. The Department also has
a major responsibility for American Indian reservation communities and for
people who live in island territories under U.S. Administration.
DisdaimerThe information contained in this report regarding commercial products or
f3ms may not be used for advertising or promotional purposes and is not tobe construed as an endorsement of any product or firmby the Bureau ofReclamation.
The information contained in this report was developed for the Bureau ofReclamation: no warranty as to the accuracy, usefulness, or compkeness isexpressed or implied.
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
5/41
Acknowledgements
Larry HaugsethTom Bunnelle
Chemical Engineetig Supervisor, BORChemical Engineering Technician, BOR
M MacDonaldHank Schmidt
Ed Burke
Research 8z Laboratory Division Electrician, BORSupervisor, City of Brighton -Director of Utility Operations, City of Brighton
Special recognition belongs to the City of Brighton for their donation of pilot plant space andinvaluable assistance on the project.
1
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
6/41
Table of Contents
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1 Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2 Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3 DesignofPilotPlant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53.1 PrinciplesofFeedandBleedandAsahiEDInformation . ..-. . . . . . . . . . . . . 5
3.1.1 AsahiED StackSpecifications . . . . . . . . . . . . . . . . . . . . . . . . . . . 53.1.2 FeedandBleedProcess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53.1.3 Understanding the Process Flow Diagram and the ED Process . . . . . . . . 7
3.2 Calculations and Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2.1 Establishing Known and Unknown Quantities . . . . . . . . . . . . . . . . . . 8
3.2.2 Calculating Diluate Flowrate . . . . . . . . . . . . . . . . . . . . . . . . . . 93.2.3 Current Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .lO3.2.4 Calculating Concentrate Flowrates and Concentrations Qc, C,,and Cc,, . . 113.25 Calculation Prediction Summary . . . . . . . . . . . . . . . . . . . . . . . . . .12
4 PilotPlant...................................................134.1 DescriptionofEquipment.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134.2 Control and Operation Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134.3 Acid Injection at the Cathode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .144.4 Power Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14
5 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..17
5.1 Compilation of Chemistry Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .175.2 Chemistry Data Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17
6 Comparison of Data to Predictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .186.1 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .196.2 Current Efficiency Spreadsheet for Concentrate Compartment (Lotus 123) . . . . . . 2 0
6.3 Current Efficiency Spreadsheet for Diluate Compartment (Lotus 123) . . . . . . . . . 2 1
7 Conclusions and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..2 3
_AppendixA....................................................2 4AppendixB....................................................2 5AppendixC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..2 6AppendixD....................................................2 7
2
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
7/41
All concentrations (mg/L of nitrate as nitrogen)CPCPCW Sk.
iGOiGO
Feed steam concentration mg/L of nitrate as nitrogenProduct stream concentration .-Waste stream concentration
(same as anode wash outlet stream)Concentrate inlet stream concentrationConcentrate outlet stream concentrationDiluate inlet stream concentrationDiluate outlet stream concentration
F l o w r a t e s
QFD Flowrate of feed stream to diluate tankQFC Flowrate of feed stream to concentrate tankQP Flowrate of total feed stream Flowrate of product streamQwntc Flowrate of waste streamQDi Flowrate of diluate inlet streamQ O Flowrate of diluate outlet streamQci Flowrate of concentrate inlet streamQ O Flowrate of concentrate outlet stream
-
Nomenclature
F B Feed and Bleed process for electrodialysisPFD Process Flow DiagramE D ElectrodialysisP&ID Piping and Instrumentation DiagramNIST National Institute of Standards and Technology
3
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
8/41
1 Executive Summary
This report summarizes work done by the Bureau of Reclamation (Reclamation) in pilottesting an electrodialysis (ED) water treatment system with special membranes tailored fornitrate removal from water. The pilot unit was tested at well no. 7 at Brighton, Colorado, a
municipal water well with nitrate concentration ranging from 10 - 20 mg/L nitrate asnitrogen. The testing took place between late 1993 and early 1994.
Product water from the ED unit had a nitrate concentration of 7.5 mg/L as nitrogen, wellunder the MCL of 10 mg/L. Water recovery was 91 percent. Feed water was fibered witha 10 micron cartridge filter prior to the ED system. Small quantities of sulfuric acid wereadded to control pH of the cathode wash streams to prevent scaling. No other scaleinhibitors or chemicals were added. The unit was run on an intermittent basis during themonth of December 1993, but was run continuously for approximately one week (betweenl/18/94 and l/24/94) before samples were taken.The City of Brighton donated space in their Reverse Osmosis (RO) Pilot Test Building,adjacent to well no. 7, for testing of the ED process. The city currently treats water using aRO system that achieves 80% water recovery using a scale inhibitor. Since Brighton waterwas also hard, there were additional water quality issues the city considered in choosing theRO process. The ED process used did not achieve the same level of mineral removal as RO,
but the ED process is optimized for nitrate removal. The ED stack and membranes werebuilt by the Asahi Glass Co. of Japan. The stack was run on a modified feed and bleedprocess. It has one electrical stage that does not reverse polarity.This report details how initial flowrates and process settings were established, how thesystem was controlled automatically, operational issues, and chemistry results from the
sampling done on l/24/94. The engineering work was done by Dave Williams whileemployed as a chemical engineer at the Bureau of Reclamation. It is the authors opinionthat this process should be considered as a competitive technology with other treatmentswhen the principal contaminate is nitrate.
2 Objective
Reclamations initial objective for this project was to establish field experience with this typeof water treatment, and to compare the process with the other nitrate removal technologiessuch as RO and ion exchange. If this technology proves promising, Reclamation wouldconsider Asahi ED when searching for treatment alternatives for other small communities.
In 1993, as part of the Small Communities Desalting program, Reclamation funded a smallproject to purchase and test an ED unit from Asahi. One goal was to acquire enoughinformation and experience to allow Reclamation to estimate treatment costs for smallcommunity ED systems. Therefore, much of the work done at Brighton was geared toward
process modeling, predicting operation, and verifying these predictions. Both the author andsupervisor Larry Haugseth agreed that it was important to develop design understanding and
4
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
9/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
10/41
PRODUCT
WASTE
FIGURE 1 PROCESS FLOW DIAGRAM
I -
ASAHI DB-0 F E E D Ic
QP
CPED STACK
Qcci Rscci
Q d iCdi D
J
DILUATE PUMP QCi Qc oCC i cc0
CONCENTRATE
CONCENTRATE PUMP
QdoCd o
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
11/41
good places to start reading about ED.
Figure 1 illustrates the feed and bleed configuration used by Asahi, and adopted by theauthor for pilot studies at Brighton. Feed and bleed (FB) typicaIly refers to using a singlestage stack with continuous recycle of product water to achieve a desired salinity. Asahiutilizes a modified FB process where both concentrate and diluate are recycled. Someauthors in [l]and [3] recycle only the product. Recycling both streams allows more controlof the process. This can become especially important at high water recoveries (> 80%) tocontrol scaling and polarization [2]. Other advantages of FB are that the stack operates at alow current density and the membranes operate under steady-state conditions [ 11 .Disadvantages of FB are higher electrical requirements than a multistage stack, morecomplicated piping and control for the various recycle streams [l], and high recirculationrates can cause water heating. It is important to note that when comparing powerrequirements between FB and multistage configurations, the multistage process should havelower power consumption than a FB process. Mintz [3] does a sample calculation in his
paper comparing the power requirements for continuous process, batch recirculation, feed-and-bleed and internally staged stack. He concludes that the feed-and-bleed processconsumes significantly more power than any of the other processes (up to 5 times as muchpower consumption). FB is used for pilot testing because it simulates staging or multiplestacks. It is especially important when scaling up a pilot study to a large commercial systemnot to confuse pilot FB power consumption to that of a large multistage stack design.However, data such as polarization and ultimate water recovery are quantities that can beused in designing large scale systems.
The process flow used for pilot studies was developed from drawing M2001-1 provided byA i. This drawing, included in the appendix of this report, specified major equipment andequipment capacities. The equipment data provided by Asahi was for maximum capacities
only, so it was necessary to perform preliminary calculations before pilot testing to establishoperating parameters for the specific system flow rates. The process flow diagram (PFD)shows the flows that were used while testing in the Reclamation Water TreatmentEngineering and Research Pilot Plant Lab. The process stream nomenclature used here ismaintained throughout the project.
3.1.3 Understanding the Process Flow Diagram and the ED Process
Various assumptions were made in determining the process flow settings.
b No significant amount of water transfers between compartments. Therefore, theflowrate of is equivalent to Qt . . Likewise, the flow rate ,is equivalent to Q&.Only minerals transfer from d&rate compartment to the concentrate compartment.
b No significant ion transport occurs in the electrode wash streams. Although the pHcan change, the majority of demineralization occurs through the diluat oncentratecompartments.
7
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
12/41
Both of these assumptions were verified in the field. To verify the first assumption, waterwas pumped through the concentrate compartment while the diluate lines were monitored forleakage. Then the same was done with the diluate compartment. A very small amount ofwater was observed to have leaked from one compartment to the other. Chemistry data wasused to verify that insignificant amounts of demineralization occurs in the electrode cavity.
By inspection of the PFD it can be seen that two flowrates are equal,
QFc=Q+=Qwntc -Qm= Q
The feed stream is split into two streams, and the feed stream to the concentrate tank is set tohave a flowrate equal to the waste stream. The feed stream flowing to the dihrate tank is setto have a flowrate equal to the product stream flowrate. If these conditions are not met,either tank will overflow.
Water from the concentrate tank is used to flush both electrodes (cathode - and anode +).Water from both diluate and concentrate compartments are recycled back to their respectivetanks. To set flowrates from either the diluate pump or the concentrate pump, streams fromthe discharge side of each pump were returned back to the tank. This simple techniqueavoids costly variable speed motors and controllers. However, this method does skew the
parasitic power measurements. These streams are not shown on the PFD but are shown inthe process and instrumentation diagram (P&ID).3.2 Calculations and AssumntionsThe following is the procedure used by the author to set initial conditions for the DB-0 EDsystem. The material balances and concentrations refer to mg/L of nitrate as nitrogen.These equations and design decisions were made in Denver before testing began.
3.2.1 Establishing Known and Unknown Quantities
Usually the nitrate concentration of the feed stream is known. The designer would choosethe quantity of product and waste, and choose the nitrate concentration of the product as a
performance specification. The remaining parameters would be calculated, and aresummarized below.
8
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
13/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
14/41
concentration was chosen to be a very small quantity, larger and larger electrical currentswould be required. Because the electrical power is what moves the ions, to get more changein concentration across the diluate compartment would require more electrical power. Thiscould mean a very large DC power supply. Also, concentration polarization becomes a
problem with very dilute streams. On the other hand, if C, were chosen to be close to 5,this would equate to a very large diluate pump. Since it represents a very small change inconcentration magnitude, large flowrates would be required to move a specific amount ofions. This can also be inferred by inspection of the design equation (1). After consideringthese two extremes, a midpoint value of 3 mg/L was chosen for C,. It may not be possibleto have complete control of CD, ; this concentration may not be attainable due to polarizationof the water, but the value was used for design and initial testing.
Given that was specified to be 4 gal/min (15.1 L/nun), and using design equation (l),QDi = 4(15-5)/(5-3)QDi= 20 galhnin (75.7 L/mill)
3.2.3 Current Calculations
In the ED process the relationship between electricrd current and the amount of mineralsremoved from the water is described by Faradays Law. The author used an equation from
PIE=FxF,AN/kI
where: E = current efficiency
F = Faraday constant (96500 amp-sec/equiv.)Ft = Total product flow (wsec)DA = change in feed concentration (equiv./L)k = number product stream compartmentsI = current (amperes)
Current was calculated to establish a target value. One assumption made was that the overalldemineralization would follow the nitrate demineralization. It should be understood that thisis an over simplification, since some ions are transferred more easily than others. But a
target value is required before pilot studies and to size the DC power supplies.
Using Faradays Law, the current was calculated for water sample H-3356 (appendix)
where:
assuming a 85% current efficiency and using 100 cell pairs.
1 0
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
15/41
Z=96500amp*sec*20 gal*3.785 L* mill * l3.68xlo-3eq peq min gal 6osec L 3 5 85 * 1
I = 2.6 amps
Since the Asahi ASV membranes are monovalent selective, this current would be a worstcastvalue. If no SO~zanions were transferred, the current consumption would decrease. BYassuming that sulfate was not in the feed water, the electrical current was found to be:
I = 1.7 amps
In the field, the voltage was adjusted to attempt to reach 1.7 amps. When running an ED
system one should be aware that this voltage may not be attainable due to polarization.Polarization could be occurring if the rate of increase in current with increasing voltagedrops off. In addition, there is an absolute limit to the voltage that can be applied.Excessively high voltage can damage membranes [4].For this project, no detailedpolarization studies were performed except for field monitoring. To perform a polarizationstudy for this water, a larger DC power supply would be required. In designing ED
equipment for field studies, it is important that relatively large voltages (for this system-up to400 volt) be available to measure polarization. The procedure in the field was to increase
the voltage slowly while monitoring resistance (voltage/current) to maintain a linearrelationship. Mason and Kirkham [2] indicate that due to the many factors affectingpolarization, experimental determination is the preferred method for design.
3.2.4 Calculating Concentrate Flowrates and Concentrations Qc, , and Cc0calculating c,:
A nitrate material balance done on the entire system gives:
Q P = QFCp + Qwas te Gate= Q i + Cb isolving for Cci,2) Gi = Q p-QpGYQw.efor QF = 4.5, Q -= 0.5, = 4, and Cp = CJ 3,
Gi = [ 4.5xc, - 4xcp/3 ]0.5Gi = 6.33
1 1
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
16/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
17/41
4 Pilot Plant4.1 Descrintion of EquipmentThe P&ID drawing shows major equipment. No detailed bill of materials was generated by
the author. Pumps were chosen by consideration of maximum capacities provided by Asahi.Most equipment was surplus and was salvaged from existing stock at Reclamation.
Diluate Pump PVC case l-1/12 x 1 discharge1HP 120VAC single phase _Concentrate Pump PVC case l/2 suction x l/2 discharge3/4HP 120VAC single phaseTanks 30 Gallon capacity, Plastic,
DC Power Supply Sorenson model DCS-150-7 (15Ov/7A)HP Linear supply (unknown model)
Diluate Flowmeter Pilot tube type F-300 Blue White Ind.
Concentrate Flowmeter andElectrode Wash Flowmeters Rotameter type
The power supplies were wired in series to provide a higher voltage. Many of the processlines were simply looped into the tanks to make sampling easier. Feed water pretreatmentwas a 10 micron filter. Occasionally, due to the nutrient rich water, it became necessary to
shock chlorinate with approximately 5 mg/L chlorine to kill bacteria growth. Bacteriagrowth was most prevalent when the stack was shut down for long periods of time (e.g. 1week) with concentrated water still in the membranes. The pressure drop across the stackdid not increase significantly over the period of operation. However, longer run times would
be required to determine if bio-fouling of the membranes was a problem.
Since the DB-0 is not a reversing stack, acid was injected at the cathode (-) to control thebasic reaction.
4.2 Control and C&ration IssuesAs previously mentioned, to maintain the levels in the diluate and concentrate tanks, the feedflowrate Qm, and Qpc) must match the flowrate of the product and waste streams,respectively.
Because of variation in well pressure, it was difficult to maintain tank levels over longperiods of time. To alleviate the constant adjustments, a control scheme was designed by the
13
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
18/41
author to monitor tank overflow with a flowswitch. The switch was chosen such that smalloverflows would not actuate, but large flows (greater than about 4 gal/m@would shut downthe system. System shutdown consisted of closing the 3/4 inch, normally closed (NC), inletsolenoid valve and turning off the pumps and rectifier. To monitor leaks and to protect thestack from a no flow condition, two pressure switches were installed on each pump. Alsoone pressure switch was utilized on the feed line to detect when the city well was shutdown.Loss of pressure from any of these three switches would shut down the system. Theelectrical logic schematic shows details of the shutdown and control logic. The shutdownlogic preformed well without any problems.
Pressure gauges were installed on the inlet side of the stack to monitor pressure drop acrossthe membranes. If inlet pressures were increasing, shock chlorination or acid descaling wasused to clean the membranes. Asahi claims that these membrane can survive acidity in therange of 2-10 PH. Acid descaling was used when the unit was shut down for long periods oftime and concentrated water had been allowed to stand in the membranes. During normalshutdowns, if the stack was flushed clean, no noticeable pressure drop occurred across themembranes and acid treatment was not required.
4.3 Acid Injection at the Cathode
As mentioned previously, acid was injected at the cathode to control pH at the concentratetank to approximately 4-6. The positive displacement diaphragm pump (LMI) withfrequency and stroke adjustment, was used to maintain a consistent injection quantity. This
proved to require periodic attention and a closed loop method was considered by the authorbut not implemented. 93% sulfuric acid was diluted to about 30%) to make a more easilyinjected solution. It was difficult to control the pH with the 93% acid; also, the more diluteacid is safer to handle and minimizes the heat of dilution at the cathode. Acid injection was
found to be 52 ml/hour when sampling was taken on l/24/94.4.4 Power Measurements
Electrical power to the stack was measured by a Fluke model 806OA digital multimeter@MM) which was calibrated via a NIST traceable standard on 8/9/93 in the Reclamationelectrical power lab. All voltage and currents to the stack were measured with this meter.Parasitic electrical power from pumps was not measured. Since flowrates and pressures areknown, pump power could be estimated, but due to the large recycle rates involved in theFB process, it was decided by the author that it would not be a valid comparison to other
processes. Also the small pumps employed are not as efficient as larger pumps that would
be used in commercial systems. _
1 4
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
19/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
20/41
F I G U R E 3 E L E C T R I C A L L O G I Cmanualstart
stop
CR M CRA PSW PSWCONC DIL , 314 N C A S C I I SCILENOID V A L V El 2 O R I F I C E 20 40~ 12OVAC. CR1
N
POWER STRIP
(DC SUPPLIES ACID PUMP3/4HP CONC. VUMP 8A>
I1HP D I L U A T E P U M P 15A
NOTE 1 PSW CONC. IS A PRESSURE SWITCH MUMTED ON THE DISCHARGE f f THE CoNcENTRnTE Puw2 PSW DIL IS A PRESSlRE SWITCH MOUTED ON THE DISCHARGE OF THE DILUATE PUMP3 OVERFCOW IS A NC FLOWSWITCH MMBUED ON THE OVERLFLOW LDK FROM BOTH TANKS1 4 3/4 ASCO SOLE NOID ,WLVE IS TED p1THE FEED LINE
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
21/41
5 Data5.1 CornDilation of Chemistrv DataMajor Cations/Anions (mg/L)
BOR SamDle HA698 H-4699 H-4701 H-4695 H-4696 H4697 H-4700Process stream c, c,Flowrates 4.68 4.256NJ .UPH 6.66 6.6 3.1Conductivity 1260 988
(us/cm @25Csuspended solids 0.357 0Dissolved solids 884 653
WWSum of ions 890 717Ca+2 112 73 307b lg+ 26.2 19 66.1Na+ 127 116 173xc 4.23 3.55 10.8co; 0 0 0HC4 3 1 0 243 0so;* 210 213 781
Cl- 100 49.1 381
Nitrate Results (mg/L)
ca,, CC3 CC0 Go c0.433 2.6 2.6 20 0.67
3.82 5.55 6.57 2.17
2950 2950 3190 923 7080
0.877 0.531 0.862 0 . 6 9 6 1 . 0 5
1970 2220 2550 620 2410
1720 1790 2070 660 2640
341 394 58.7 357
72.4 82.7 16.4 75.218 8 202 10 3 194
10.6 11.9 4.42 10.9
0 0 0 0
0 94.3 226 0
787 810 215 1610
392 472 36.2 395
BOR SamDle H-4705 H-4706 H-4708 H-4707 HA703 H-4704 H-4707NO,asN 5.9 7.5 65 73.8 87.3 5.55 76.4Selected Metals (mg/L)
BOR Sample # H4714 H-4715 H-4717 H4711 H4712 H-4713 H-4716
Ba+ ug/l 253.2 62.59 241.6 253.2 294.1 54.23 262.1F e 49.67 9.112 55.62 49.67 51.97 11.49 127.4
Mn
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
22/41
6 Comparison of Data to Predictions
Table 1. Comparison of data with predicted values.
Description Quantity P r e d i c t i o n A c t u a l CommentsFeed FlowrateFeed Concentration
Dil. Flowrate InletDiluate (orProduct).Concentration Inlet
Diluate
ConcentrationOutlet
Cont. FlowrateInlet
Cont.Concentration Inlet
Cont.ConcentrationOutlet
waste FlowrateCathode WashFlowrateProduct FlowrateStack Voltage
Stack Current
Stack Power
Acid Injection
Diluate Pressure
ConcentratePressure
CD,OWL)
Cci w/L)
QPcf5dmin) 4.0 4.25v (Volts) ? 300
1 (Amps) 1.7-2.6 1.14
P (watts)
PsiSpsig
4.5
15
20
5
3
2.7 2.6 calibrated OK
9 5
110
0.5 0.433
0.5 0.67
?
??
4.68
15.9
20
7.5
5.55
73.8
87.3
342
52mVh.r11
10
calculated value
given by well
difficult to calibrate-
calibrated OK
calibrated OK
calibrated OK
maximum voltage
more voltagerequired?
._
18
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
23/41
6.1 Discussion
The electrical current calculation did not agree with the experimental value for two reasons.The calculation was performed making the assumption that all ions are removed with equalefficacy. Also, 300 volts was the maximum voltage attainable with the power supplies
available. Higher voltages and higher corresponding currents may have improved productconcentrations with the possibility of reaching the target of 5 mg/L nitrate as nitrogen. Ofcourse, another way to improve the product concentration would be to change flowrates.
The nitrate concentrations on the concentrate streams, both inlet and outlet, appear to havelower values than predicted. However, all concentration calculations were based on a
predicted product concentration of 5 mg/L. Since the resulting product concentration wasfound to be 7.5 mg/L, the other concentrations are expected to vary. The appendix showsthe design equations calculated with actual values (see appendix).When reviewing the design equations and material balances, it is important to keep in mindthat the ideal conditions, Qld=Q and Qp,=Qw are not always met. In practice there isalways some slight overflow of the tanks. It was found this was the preferred method ofoperation, to avoid any tank running empty. If the pressure of the well varied, the flowratewould change to each tank and eventually a tank would empty, shutting down the system onlow pressure. This slight overflow condition introduces error in the material balancesthough. To maintain tighter control, an investment in better equipment, such as flow controlvalves and pressure regulators, would be required. Qpd and QpC were measured during this
project but not recorded; in the future it would be a good idea to record these values to helpin resolving material balance errors. Other errors in the material balances could be attributedto variations in chemistry results and inaccuracies in flow measurements.
19
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
24/41
Calculation of current for Asahi ED StackCONCENTRATE COMPARTMENT
.-Ions Equiv.Wt Cci Cci C, Cc0
@-s/eq.) OWLI (meq/L) (mg/L) (meq/l)H-4695 H-4696
-a+2 20.00 341.00 17.05 394.00 19.70Mg+ 12.20 .72.40 5.93 82.70 6.78Na+ 23.00 188.00 8.17 202.00 8.78K+ A3910 1060 0.27 1190 0.30sum ofcations 31.43 35.57
co,- 30.00 0.00 0.00 0.00 0.00HCO, 61.00 0.00 0.00 94.30 1.55SOi 48.00 787.00 16.40 810.00 16.88Cl- 35.50 392.00 11.04 472.00 13.30NO,asN 73.80 87.30NO, as ION62.00 326.83 5.27 386.61 6.24sum of anions 32.71 37.95Normality(N) 32.07 36.76
TOTAL 2117.83 2453.51
AN (meq/L)= -4.69 Flow paths= 100Flow(gal/min)= 2.60 Cell pairs= 100change in equivalents = Flowrate(Lhnin)*AN @q/L)=3.785*2.60*4.69/1000=0.0462 eq./min
20
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
25/41
6.3 Current Efficiencv Sureadsheet for Diluate Comnartment (Lotus 123)Calculation of current for Asahi ED StackDILUATE COMPARTMENT
Ions Equiv.Wt Productgrams@ @x/L
H-4699
Product@w/L)Ca+2 20.00 73.00 3.65Mg+ 12.20 19.00 1.56Na+ 23.00 116.00 5.04K+ 39.10 3.55 0.09sum of cations 10.34
coc2 30.00HCO, 61.00so, 48.00Cl- 35.50NO;asNNO; as ION 62.00sum of anionsNormality(N) (average)
0.00243.00
213.0049.00
7.5033.21
0.00 0.00 0.003.98 226.00 3.70
4.44 215.00 4.481.38 6.20 1.02
0.54
10.34
GOw/L)H-4697
58.7016.40
103.004.42
5.5524.5810.34
2.941.34 -4.480.118.87
0.409.609.24
-___-_______________---------------------------------------------------------------TOTAL 749.76 684.30
AN me@ = 1.10 F low paths= 100Flow gal/min) = 20.00 Cell pairs= 100Change in equivalentbin = Flowmte L/min)*AN @q/L)
= 3.785*20.0*1.10/1000= 0.0833 eqhnin
Averaging the concentrate and diluate compartments:
Average equivalents transferred/min = 0.0462+0.0833)/2 = O.O648eq./minE = Faraday amp-sec/es *~o~~~~ *~ ~~ /c~ent amps)/flowpaths
= 96500*0.0648/60/1.13/100 . .E = 0.922
21
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
26/41
7 Conclusions and Recommendations
Other Data and SamplesWater samples were taken after initial startup in Denver, but the author has omitted discussing
results. While being tested in Denver, the stack removed nitrate as expected, but the author doesnot believe the unit was run long enough with consistent feed concentrations to provide usefulinformation. Reclamation continued to run the pilot system in Brighton after the author was nolonger employed at Reclamation, and that data has not been considered in this report.
Current Efficiencv -
The current efficiency, calculated in section 6.2, was found to be 92%. This value was derivedby averaging the current efficiencies between the concentrate and diluate compartments. Thismethod was chosen since the efficiency is sensitive to slight variations in chemical results anderrors in flowrates. Typical commercial ED systems have current efficiencies between N-95 .Power Consumotion and PostTreatmentDC power consumption was found to be 1.5 KWH/1000 Gallons of Product. Although no anti-scalanb were required for the process, due to the low pH of the waste water, post treatment may
be required before disposal.
Recommendations for Further Studv
One of the disadvantages of ED is that numerous process streams samples are required toverify predications. Of course this increases the cost of pilot studying ED systems. A hand-held
nitrate probe was used at Brighton, but it was difficult to obtain consistent readings. Perhapswith more field experience, consistent readings are possible or, better field type measurementequipment could be purchased.
Since only one sample point was taken, longer testing duration and more frequent samples wererequired. The City of Brighton at one time offered space in their large RO treatment plant,located east of town, for continued testing of the process. The City eventually sold the RO PilotTesting facility on Jesup Street and Reclamation had to cease testing and remove equipment.
It is premature to recommend this process for commercial water treatment processes withoutfurther study. But some conclusions can be made. High recoveries (> 80%) are possible with
this technology, and with ion selectivity, overall power consumption can be lower thanconventional ED.
22
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
27/41
-
References
1. Shaffer L.H. and Mintz M.S., principles of Lksuli~on, Chap. 6, pp 257-357, AcademyPress 1980, ISBN o-12-656701-8
2 . Mason E.A. and K i r k h am T.A., D esign of Ekx t r od ia l y s i s EquQment, Chem. Eng.Progress Symposium Series,vol. 55, no.24 pp 173-189, 1959
3 . M i n t z M.S., Electrodialysis Principles of Process Design, (complete text included inappendix)
4. Meller Floyd H., Electmdia lys is (ED)& Electrodi alysis Reversal EDR Technology. IonicsInc. 1984.
23
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
28/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
29/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
30/41
,
r--------------L:K - 1 2 0I-J f
I
\s .v
C.l\
70
-n 1L l j1x11 No, fR-loo IRECW-loo M - 1 1 0 I M - 1 2 0
tOUT NAME RECIWWd RECTWEA k DKUAlENma PAM1
CAPACIN W-0 ( IDD DClODVnJA 50MAlClUAl OlJANllrY 1 i I
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
31/41
APPENDIXD
-
fELE TRODI LYSISPRINCIPLES OF PROCESS DESIGN
CONCENTRATE STREAM
PRODUCT STREAM CONCENTRATE S1
0
q
I JPK NT~FLOW
Tl t NPERMEABLE
MEMBRANE
RE M
I Vf ddrRP E
Figure 7. dDeminerali@ion Yelect r i a~sis i na sir e m h nr pair. Basicperformance equations of such a pair cari be extended to complex engirieering processes
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
32/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
33/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
34/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
35/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
36/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
37/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
38/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
39/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
40/41
-
8/13/2019 Brighton ED Testing With Asahi Monovalent Membranes
41/41
top related