EPA/600/R-05/119S
September 2005
THE ENVIRONMENTAL TECHNOLOGY VERIFICATION
PROGRAM
-------
ABSTRACT
Verification testing of the ORCA Water Technologies KemLoop 1000 Coagulation and Filtration Water
Treatment System for arsenic removal was conducted at the St. Louis Center located in Washtenaw
County, Michigan from March 23 through April 6, 2005. The source water was groundwater from two
supply wells, and the raw water for the verification test was withdrawn from the pressure tank at the site.
Verification testing was conducted at the operating conditions specified by the manufacturer. The raw
water, with a pH in the range of 7.0 to 7.6, was treated with chlorine bleach to oxidize arsenic (III) to
arsenic (V), as well as iron to coagulate the arsenic. When operated under the manufacturer's specified
conditions at this site, at an average flow rate of 9.9 gallons per minute (gpm), the KemLoop System
reduced the total arsenic concentration from an average of 22 micrograms per liter (ug/L) in the feed
water (raw water after chemical addition) to 3 ug/L in the filtrate (treated) water.
TECHNOLOGY DESCRIPTION
The following technology description was provided by the manufacturer and has not been verified.
The ORCA process is based on chemical addition with mixing in a proprietary mixing loop to optimize
coagulation, and granular media filtration with no intermediate solids separation process. The KemLoop
System includes pretreatment with sodium hypochlorite to oxidize any arsenic (III) to arsenic (V), and
iron present in the water supply. Ferric chloride is added to augment any natural occurring iron and
optimize the iron dose. The chemically treated water (feed water) enters the mixing loop where
coagulation of arsenic and iron occurs. The water exits the mixing loop and is applied directly to one of
the two granular media filter modules. The water enters the top of the operating filter and flows through
the granular media filter, exiting at the bottom of the module. The granular media filter removes the
precipitate, including arsenic, iron, and any other precipitated constituents. The two-filter module system
operates with the filters in parallel, one filter module is in active operation and one unit is in standby
mode. When backwash of a filter module is required, the standby filter is brought online and the
backwash cycle for the "dirty' filter module is initiated. Once the backwash cycle is complete, the clean
filter module becomes the standby unit.
The KemLoop System is fully automated and programmed to control all aspects of the filter operation.
The control system automatically initiates backwash cycles based on four criteria: differential pressure
across the media filter, treated water turbidity compared to raw water turbidity, time, and volume, as set
by the operator. The backwash frequency is dependent on the water quality conditions and the amount of
solids generated in the coagulation process. The control system is a programmable logic control and
personal computer (PLC/PC) based controller with data logging, trend display graphs, and a remote
monitoring modem connection for off-site technical support. All the information is available to the on-
site operator and to remote users
VERIFICATION TESTING DESCRIPTION
Test Site
The verification test site was the St. Louis Center, a residential community for people with developmental
disabilities, located in Washtenaw County Michigan. The source water was groundwater from two wells
located at this site, which pumped water to a common pressure tank that served as the raw water supply to
the KemLoop System. Water quality data from historical information and the characterization test showed
the wells had similar water quality. Total arsenic in the combined well water ranged from 14 to 32 ug/L
and total iron ranged from 0.39 to 1.6 milligrams per liter (mg/L). The pH was in the 7.4 to 7.6 range with
alkalinity of 250 to 260 mg/L as CaCO3. Raw water turbidity was found to be <1 nephelometric turbidity
unit (MTU) in 2004 and 1.2NTU in the 2005 characterization test.
05/20/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2005
vs-ii
-------
Methods and Procedures
Operations, sampling, and analyses were performed in accordance with the Product Specific Test Plan
(PSTP) developed and approved for this verification test. The PSTP included a Quality Assurance Project
Plan (QAPP) designed to assure the quality of the data collected and to provide an accurate evaluation of
the treatment system under the field conditions. Testing included characterization of the raw water, an
arsenic loss test (no chemical fed to the system), and a 14-day verification test.
The verification test was performed from March 23, 2004 through April 6, 2005. The KemLoop System
was operated continuously for the 14-day verification test, independent of the well operations, by using
water supplied from the pressurized supply tank. Flow rate(s), production volume, water temperature, and
system pressure(s) were monitored and recorded daily. Raw, feed (after chlorine and iron addition), and
filtrate (treated) water samples were analyzed on-site for pH, temperature, turbidity, free and total residual
chlorine, color, and dissolved oxygen by the field operator. Grab samples were collected and delivered to
the NSF Drinking Water Laboratory to be analyzed for alkalinity, calcium, magnesium, iron, manganese,
sulfate, chloride, total organic carbon (TOC), total suspended solids (TSS), and fluoride. Samples for
total arsenic were collected daily, plus 14 samples were collected during a 48-hour intensive survey. In
addition to the 25 sets of samples for total arsenic, a total of four sets of arsenic samples were speciated
during the test to determine the soluble arsenic concentration and the concentrations of arsenic (III) and
the arsenic (V) present in the soluble fraction. Samples of backwash water were collected and analyzed to
characterize the backwash wastewater.
Complete descriptions of the verification testing results and quality assurance/quality control (QA/QC)
procedures are included in the verification report.
VERIFICATION OF PERFORMANCE
System Operation
ORCA performed the system startup and shakedown testing, which included optimization of the chemical
feed rates, and determination of backwash frequency. The verification test was conducted under the
manufacturer's specified operating conditions. Chemical feeds were established to feed 1.0 mg/L of total
chlorine. The ferric chloride feed rate was set to deliver 1.5 to 2.5 mg/L (as Fe) of iron to augment the
naturally occurring iron of 0.5 mg/L. The flow rate for filtrate was set at 10 gpm to give a targeted
surface-loading rate of 2940 gallons per day per square foot (gfd). The backwash system was set to
backwash once per day or if the pressure differential across the filter exceeded 8 pounds per square inch
(psi) or if turbidity of the filtrate exceeded the raw water for ten minutes. The backwash cycle used
treated water, which was pumped at 50 gpm through the filter in an up flow mode to flush out 1he
accumulated solids.
System pressure was monitored at three locations, raw water (from pressure tank), feed water (inlet to the
filters), and filtrate (exit from the filters). There was very little change in head loss through the filter over
each 24-hour operating period. The maximum pressure differential observed was 5.0 psi, with the filter
inlet side averaging 5.8 psi and the filter outlet side averaging 2.0 psi. The automatic backwash cycle was
not triggered due to pressure differential (head loss) or an increase in filtrate turbidity at any time during
the verification test.
The filtrate flow rate remained steady for most days during the test yielding an average flow rate of 9.9
gpm over the 14 days. The total filtrate volume produced each day was also consistent, except for April 2
through 4 when volumes and flow rates were somewhat lower. It appears the pressure on the raw water
supply tank at the St. Louis Center was periodically dropping below 40 psi (the setting on the pressure
regulator). This caused periodic lower flow rates and lower volumes of filtrate to be produced over the
05/20/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2005
vs-iii
-------
24-hour period. The average hydraulic loading through the filter was 2,890 gallons per square foot, based
on the 24-hour filter run time between backwashes and the average daily filtrate production of 14,630
gallons.
Water Quality Results
The results of total arsenic analyses are shown in Figure VS-1. The raw water total arsenic averaged 23
ug/L with most of the arsenic as arsenic (III). Following chemical treatment, the feed water total arsenic
concentration averaged 22 ug/L. While the soluble arsenic and arsenic speciation data showed some
variability, the data indicate that pretreatment completely converted the raw water arsenic (III) to the
arsenic (V). The filtrate water total arsenic concentration averaged 3 ug/L with the concentration being
below the detection limit (1 ug/L) on six of 14 days. The filtrate exceeded 10 ug/L on the first and last
day of the verification test. On the first day the total arsenic concentration was 12 ug/L with dissolved
arsenic of <1 ug/L. It appears the high arsenic concentration was caused by an overdose of ferric chloride
resulting in solids passing through the filter. After adjusting the iron feed rate, the turbidity in the filtrate
dropped from 1.7 NTU to 0.10 NTU and the arsenic on Day 2 was 1 ug/L. The cause of the higher filtrate
arsenic concentration (11 ug/L) on the last day is not known, as the iron feed rate, and iron concentration
and turbidity level in the filtrate were low. The data collected during the 48-hour intensive survey were
consistent with the data collected each day during the verification test. There was no indication of any
transient or short time changes in the arsenic concentration or in any other monitored parameters.
O)
3
c
01
1/1
ra
'o
35
30
25
20
15
10
5
0
Date
Raw —H Feed
• Filtrate
Note: 48-hour intensive survey began on 3/30/05.
Figure VS-1. Total Arsenic Results
The raw water and filtrate alkalinity averaged 260 mg/L as CaCQj, indicating that the chemical addition
and filtration process had no impact on the alkalinity concentration. The pH of the raw water was steady
in the range of 7.20 to 7.48 with a mean value of 7.30. The filtrate pH ranged from 7.22 to 7.46 with a
median value of 7.30 showing that the addition of chlorine and ferric chloride had very little impact on
pH. The average raw water iron concentration was 0.47 mg/L, and the feed water averaged 1.9 mg/L of
iron after the addition of ferric chloride. The filtrate water iron concentration was 0.03 mg/L or less on ten
out of fourteen days. On March 25 and 31, the iron concentration was 0.08 and 0.07 mg/L, respectively.
The first day of the test, when the iron concentration in the feed was measured at a maximum
concentration of 4.5 mg/L (chemical feed pump subsequently adjusted downward), the filtrate
concentration was 1.7 mg/L. On March 27 the iron was 0.31 mg/L. These data show that the KemLoop
System can produce a filtrate with <0.30 mg/L of iron. The KemLoop System lowered the turbidity levels
05/20/EPADWCTR
The accompanying notice is an integral part of this verification statement.
vs-iv
September 2005
-------
with the filtrate turbidity averaging 0.30 NTU based on the bench-top turbidimeter and 0.20 NTU based
on the inline turbidimeter. The bench top turbidity meter always gave higher turbidity readings compared
to the inline units. The raw water turbidity based on the bench top unit averaged 2.4 NTU, whereas the
average turbidity based on the inline unit was 0.60 NTU. It is believed that the bench top unit data may
have been biased high due to temperature and fogging issues that can be problematic when collecting cold
samples and transferring them to the bench top vials. Based on the bench-top meter measurements, the
filtrate was below 0.5 NTU in 93% of samples, had no values between 0.5 and 1 NTU, and 7% of the
readings (1 reading) were between 1 and 2 NTU. There were no turbidity levels above 2 NTU. The inline
turbidimeter gave the same distribution of turbidity readings in the filtrate. During the 48-hour intensive
survey the turbidity levels in the filtrate did tend to increase slightly near the end of each filter run, and
then were lower again when the standby filter was brought on line. All inline turbidity measurements for
the filtrate during the 48-hour intensive survey were below 0.2 NTU, even at the end of a 24-hour run.
The backwash water was sampled on four occasions and found to have an average total arsenic
concentration of 760 (ig/L, an average iron concentration of 120 mg/L, and an average TSS concentration
of 250 mg/L. The backwash cycle occurred once every 24 hours and yielded an average of 220 gallons
per day of backwash water. This represented 1.5% of the average daily treated water production. The
backwash water was enriched in arsenic, iron, and TSS, as would be expected, given the removal of
arsenic and iron as measured in the filtrate. Local disposal requirements determine whether this water is
acceptable for discharge to a sanitary sewer system, some other discharge location, or if it will require
further treatment prior to discharge. The backwash solids are not considered a hazardous waste based on
Toxicity Characteristic Leaching Procedure (TCLP) arsenic results of 0.32 mg/L, which is below the 5.0
mg/L limit under the Resource Conservation and Recovery Act (RCRA).
Operation and Maintenance Results
The KemLoop System was found to be easy to operate and required little time for daily maintenance. The
field staff was on-site for two to three hours per day. Most of the time on-site was spent performing field
activities, including daily chemical analyses, flow checks, calibrations, etc. In a normal operation, the
inline pH meters and turbidimeters would be used for system checks. The KemLoop System has a
PLC/PC that records data for all key operating parameters, including flow data, pressure information,
backwash cycles, etc. It is estimated that the time to check the system on-site would be minimal, possibly
less than 30 minutes, except when chemical feedstocks needed to be replenished or inline instruments
calibrated. The PLC can be setup for remote access; so main system parameters can be monitored without
a site visit.
The ORCA operation and maintenance (O&M) manual provides a detailed description of the system,
appropriate safety precautions, and detailed descriptions of operating procedures, capability and operation
of the computer control system, and specific instructions for utility operators. The maintenance section of
the manual includes some descriptions of required maintenance, but refers the reader to the individual
equipment literature supplied by the various pump and instrument manufacturers. These manuals were
provided in a notebook. The draft O&M manual did not contain specific checklists for routine site visits.
The review of the O&M manual shows that the manual is well organized and easy to read.
Consumables and Membrane Chemical Cleaning
The KemLoop System used a 6% sodium hypochlorite (bleach) solution, made on site from a 12% stock
solution. A total of 28 liters of 6% bleach solution was used to treat 204,870 gallons of raw water. This
equates to an average concentration added to the raw water of 2.2 mg/L. The average total residual
chlorine in the feed water after chlorine addition was 1.0 mg/L, indicating a chlorine demand in the water
of 1.2 mg/L. Iron was added to the raw water using a 4.8% as iron (Fe) ferric chloride solution. A total of
05/20/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2005
VS-V
-------
23.9 L (6.3 gallons) was used to treat 204,870 gallons of raw water, yielding an average concentration of
iron added to the water of 1.5 mg/L. The feed water concentration averaged 1.9 mg/L and the raw water
concentration averaged 0.47 mg/L, indicating 1.43 mg/L of iron addition, which was close to the
calculated 1.5 mg/L fed based on chemical use.
Electrical power consumption was estimated based on the raw water pump (not used at this site) and
backwash pump horsepower. With miscellaneous electrical use by chemical feed pumps and the PLC/PC,
power consumption is estimated to be 0.5 kilowatt-hr.
Quality Assurance/Quality Control
NSF provided technical and QA oversight of the verification testing as described in the verification
report, including an audit of nearly 100% of the data. The NSF QA department conducted a technical
systems audit during testing to ensure the testing was in compliance with the test plan and performed a
QA review of the analytical data. A complete description of the QA/QC procedures is provided in the
verification report.
Original Signed by Original Signed by
Sally Gutierrez 10/3/05 Robert Ferguson 10/5/05
Sally Gutierrez Date Robert Ferguson Date
Director Vice President
National Risk Management Research Laboratory Water Systems
Office of Research and Development NSF International
United States Environmental Protection Agency
NOTICE: Verifications are based on an evaluation of technology performance under specific,
predetermined criteria and the appropriate quality assurance procedures. EPA and NSF make no
expressed or implied warranties as to the performance of the technology and do not certify that a
technology will always operate as verified. The end-user is solely responsible for complying with
any and all applicable federal, state, and local requirements. Mention of corporate names, trade
names, or commercial products does not constitute endorsement or recommendation for use of
specific products. This report is not an NSF Certification of the specific product mentioned
herein.
Availability of Supporting Documents
Copies of the ETV Protocol for Equipment Verification Testing for Arsenic Removal
dated April 2002, the verification statement, and the verification report (NSF Report
#04/10/EPADWCTR) are available from the following sources:
(NOTE: Appendices are not included in the verification report. Appendices are available
from NSF upon request.)
1. ETV Drinking Water Systems Center Manager (order hard copy)
NSF International
P.O. Box 130140
Ann Arbor, Michigan 48113-0140
2. NSF web site: http://www.nsf. org/etv (electronic copy)
3. EPA web site: https://www.epa.gov/etv (electronic copy)
05/20/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2005
vs-vi
------- |