crc
Concurrent
Technologies
Corporation
U.S. Environmental Protection Agency
Environmental Technology Verification Program
For Metal Finishing Pollution Prevention Technologies
Verification Test Plan
Evaluation of USFilter Membralox® Silverback™ Model
900 Alkaline Cleaner Recycling System
Concurrent Technologies Corporation is the Verification Partner for the EPA ETV Metal
Finishing Pollution Prevention Technologies Pilot under EPA Cooperative Agreement No.
CR826492-01-0.
Revision 0
April 4, 2000
-------
Revision 0 - 4/4/00
etV
crc
Concurrent
Technologies
Corporation
U.S. Environmental Protection Agency
Environmental Technology Verification Program
For Metal Finishing Pollution Prevention Technologies
Verification Test Plan
Evaluation of USFilter Membralox® Silverback™ Model 900
Alkaline Cleaner Recycling System
April 4, 2000
Prepared by: ETV-MF Program and
USFilter Corporation
i
-------
Revision 0 - 4/4/00
TITLE: EVALUATION OF USFILTER MEMBRALOX® SILVERBACK™
MODEL 900 ALKALINE CLEANER RECYCLING SYSTEMS
ISSUE DATE: April 4, 2000
DOCUMENT CONTROL
This document will be maintained by Concurrent Technologies Corporation in accordance with
the EPA Environmental Technology Verification Program Quality and Management Plan for the
Pilot Period 1995-2000 (EPA/600/R-98/064). Document control elements include unique issue
numbers, document identification, numbered pages, document distribution records, tracking of
revisions, a document MASTER filing and retrieval system, and a document archiving system.
ACKNOWLEDGMENT
This is to acknowledge Marion Rideout, Jim Totter and Valerie Whitman for their help in
preparing this document.
Concurrent Technologies Corporation is the Verification Partner for the EPA ETV Metal
Finishing Pollution Prevention Technologies Pilot under EPA Cooperative Agreement No.
CR826492-01-0.
11
-------
Revision 0 -
Environmental Technology Verification Program for Metal Finishing Pollution Prevention
Technologies (ETV-MF) Verification Test Plan for the Evaluation of USFilter Membralox®
Silverback™ Model 900 Alkaline Cleaner Recycling System.
PREPARED BY:
V- // - 2-ooi
George Cushnie, CAI ResourcesT
ETV-MF Project Manager Date
APPROVED BY:
Clinton TwiUey
CTC QA Manager Dmte
t-f/ /Cf /
-------
Revision 0 - 4/4/00
TABLE OF CONTENTS
1.0 INTRODUCTION 1
2.0 TECHNOLOGY DESCRIPTION 2
2.1 Theory of Operation 2
2.2 Description of Membralox® Silverback™ Unit 2
2.3 Commercial Status 4
2.4 Pollution Prevention Classification 4
2.5 Environmental Significance 4
2.6 Local Installation 5
3.0 EXPERIMENTAL DESIGN 7
3.1 Test Goals and Obj ectives 7
3.2 Critical and Non-Critical Measurements 8
3.3 Test Matrix 8
3.4 Testing and Operating Procedures 9
3.4.1 Set - Up and System Initialization Procedures 9
3.4.2 System Operation 11
3.4.3 Sampling and Process Measurements 11
3.4.4 Production Measurements 14
3.5 Analytical Procedures 14
3.5.1 Alkaline Cleaner Constituents 14
3.5.2 Oil 15
3.5.3 Total Solids and Suspended Solids 17
4.0 QUALITY ASSURANCE/QUALITY CONTROL REQUIREMENTS 18
4.1 Quality Assurance Obj ectives 18
4.2 Data Reduction, Validation, and Reporting 18
iv
-------
Revision 0 - 4/4/00
4.2.1 Internal Quality Control Checks 18
4.2.2 Calculation of Data Quality Indicators 19
4.2.2.1 Precision 20
4.2.2.2 Accuracy 20
4.2.2.3 Completeness 22
4.2.2.4 Comparability 22
4.2.2.5 Representativeness 22
4.2.2.6 Sensitivity 22
4.2.2.7 Method Detection Limit 22
4.2.2.8 Cleaner Recovery Efficiency 23
4.2.2.9 Mass Balance 23
4.2.2.10 Energy Use 24
4.2.2.11 Cost Analysis 24
4.2.2.12 Waste Generation Analysis 24
4.3 Quality Audits 24
5.0 PROJECT MANAGEMENT 25
5.1 Organization/Personnel Responsibilities 25
6.0 EQUIPMENT AND UTILITY REQUIREMENTS 26
7.0 HEALTH AND SAFETY PLAN 26
7.1 Hazard Communication 26
7.2 Emergency Response Plan 26
7.3 Hazard Controls Including Personal Protective Equipment 27
7.4 Lockout/T agout Program 27
7.5 Material Storage 27
7.6 Safe Handling Procedures 27
v
-------
Revision 0 - 4/4/00
8.0 WASTE MANAGEMENT 27
9.0 TRAINING 27
10.0 REFERENCES 28
11.0 DISTRIBUTION 28
LIST OF APPENDICES
PAVCO Clean-R-120-GR Material Safety Data Sheet
USFilter Membralox® Silverback™ Model 900 Operation and
Maintenance Manual (Selected Sections)
ETV-MF Operation Planning Checklist
Job Training Analysis Form
ETV-MF Project Training Attendance Form
Oil Analysis - EPA Method 8015, Modified (NW Total Petroleum
Hydrocarbon - Extended Diesel) (NWTPH-DX)
LIST OF FIGURES
Figure 1: Diagram of the Membralox® Silverback™ Technology 3
Figure 2: Alkaline Cleaner Recycling at Gates Rubber Company 6
Figure 3: Data Collection Form 13
LIST OF TABLES
Table 1: Test Objectives and Related Test Measurements for Evaluation of the
Membralox® Silverback™ Model 900 10
Table 2: Sampling Frequency and Parameters to be Measured 12
Table 3: Summary of Analytical Tests and Requirements 16
Table 4: QA Objectives 21
APPENDIX A:
APPENDIX B:
APPENDIX C:
APPENDIX D:
APPENDIX E:
APPENDIX F:
vi
-------
Revision 0 - 4/4/00
1.0 INTRODUCTION
The purpose of this test plan is to document the objectives, procedures, equipment, and
other aspects of testing that will be utilized during verification testing of the Membralox®
Silverback™ Model 900 alkaline cleaner recycling system. This test plan has been
prepared in conjunction with the U.S. Environmental Protection Agency's (EPA's)
Environmental Technology Verification Program for Metal Finishing Pollution Prevention
Technologies (ETV-MF). The objective of this program is to identify promising and
innovative pollution prevention technologies through EPA supported performance
verifications. The results of the verification test will be documented in a verification report
that will provide objective performance data to metal finishers, environmental permitting
agencies, and industry consultants. A verification statement, which is an executive
summary of the verification report, will be prepared and signed by the EPA National Risk
Management Research Laboratory Director.
Alkaline cleaning baths are used by nearly all metal finishing companies to remove oil and
other soils prior to electroplating, anodizing, painting, or other finishing processes. These
processes are very sensitive to the presence of surface contaminants. Usually, alkaline
cleaning baths are used until they become moderately contaminated with oil and suspended
solids, and then they are discarded and replaced with fresh cleaning chemicals, despite the
fact that a significant portion of the bath's constituents are still usable. If used beyond a
certain point, the contamination will prevent good surface cleaning, which may lead to
defects in a part's finish. Also, contaminants in the cleaning baths can carry over to
subsequent tanks and cause deterioration of those solutions. The practice of discarding and
replacing contaminated cleaning baths is expensive due to the cost of cleaning chemistry
(most cleaners are proprietary formulations) and the cost of disposal.
This project will evaluate the ability of the Silverback™ unit to separate oil and suspended
solids from the alkaline cleaning bath under different soil loading rates. The Silverback™
unit employs a ceramic microfiltration membrane to perform the separation of
contaminants and cleaning solution.
Testing of the Silverback™ unit will be conducted at Gates Rubber Company located in
Versailles, MO. The Gates Rubber Company manufactures a wide range of hydraulic
couplings and hose assemblies. Alkaline cleaning is performed on metal parts at strategic
times during the manufacturing process, including prior to zinc electroplating. Used
alkaline cleaning solution is piped to a storage tank that feeds the Silverback™ unit.
Recovered alkaline cleaner from the unit is pumped to a recycled cleaner tank and reused.
Testing will be conducted over a two week time period, which will be divided into two 1-
week test runs. During the first test run, the unit will be operated at the normal soil-loading
rate. During the second test run, the soil-loading rate will be doubled. During both test
runs, the process will be monitored to measure the soil-loading rate, the recovery efficiency
of the unit, and certain process measurements.
1
-------
Revision 0 - 4/4/00
The test plan described in this document has been structured based on a format developed
for ETV-MF projects. This document describes the intended approach and explains testing
plans with respect to areas such as test methodology, procedures, parameters, and
instrumentation. Also included is Quality Assurance/Quality Control requirements of this
task that will ensure the accuracy of data, data interpretation procedures, and worker health
and safety considerations.
2.0 TECHNOLOGY DESCRIPTION
2.1. Theory of Operation
Alkaline cleaning solutions are a mixture of chemicals, including surfactants, alkali
salts, caustic soda, phosphates, and complexing agents. These baths build up
concentrations of oil and solids during use. Free oils can be removed by simple
skimming, and most solids can be removed by settling and/or cartridge filtration.
However, emulsified oils and colloidal solids are not affected by these devices. At
some point, the cleaning efficiency of the bath is impaired and the solution is
discarded, despite the fact that most of the bath's constituents are still usable. In
many cases, heavy-duty cleaners must be replaced once per week. The
microfiltration technology separates the emulsified oils and colloidal solids from the
aqueous cleaning solution, thereby extending the life of the bath. This technology is
also applicable to the recovery of cleaning solution drag-out from rinse waters.
Many commercial microfiltration systems used for this application, including the
Silverback™ technology, employ ceramic filter membranes in a crossflow filtration
configuration. These membranes are a relatively new development that permits
application of microfiltration to solutions and emulsions that are both heated and
corrosive. The ceramic membranes are produced in a range of pore sizes that
selectively permit a large percentage of the surfactants to pass through the
membrane. A typical pore size is 0.8 microns (|i) and most microfiltration units
have pore sizes greater than 0.2 |i.' Crossflow filtration, as opposed to barrier or
"dead-end" filtration, permits the application of this technology to high solids-feed
streams. With dead-end filtration, all of the feed solution is forced through filtration
media by an applied pressure. With a high solids-feed stream, the pores of a dead-
end filtration device plug. With crossflow filtration, the fluid to be filtered is
pumped across the membrane, parallel to its surface. By maintaining a high velocity
across the membrane, the retained material is swept off the membrane surface. [Ref.
1,2]
2.2 Description ofMembralox® Silverback™ Unit
The Membralox® Silverback™ Model 900 is an alkaline cleaner recycling system.
This unit contains a Membralox® TP 19-40 module made up of seven alpha alumina
elements, each with 19 lumens (channels) that are 4.0 mm in diameter. The inside of
each channel is lined with an alpha alumina or zicronia membrane having a total
surface area of 1.68 m2 (18.1 ft2) and a pore size of 0.2 |i. The manufacturer
1 1 millimeter = 1,000 microns
2
-------
Revision 0 - 4/4/00
indicates that the membrane is impervious to nearly all chemicals except for
phosphoric and hydrofluoric acids and can tolerate temperatures up to 200°F.
A diagram of the Silverback™ Model 900 is shown in Figure 1. In operation, the
contaminated cleaner enters a two-compartment 304 stainless steel tank through a
filter (polypropylene sock and stainless steel basket) that removes large particulate
material from the feed stream. The level in the tank is maintained by a level switch,
which controls the tank inlet valve and also acts as a low-level cutoff for the system
pump. The oils accumulate in the initial compartment (referred to as the settling
tank) and can be removed on a periodic basis through a drain port located on the
upper part of the tank.2 The liquid then moves to a second tank compartment
through a sub-surface passage; thereby leaving the floating oils in the first
compartment. The liquid in the second compartment (referred to as the recirculation
tank) is pumped through the ceramic filter located in the microfiltration module. A
portion of the water and cleaner chemicals are forced through the ceramic
membrane, while the oil and other soils are retained and recycled back to the tank.
A range of system capacities is available for feed stream flow rates of 150 gpd to
3,900 gpd (model 900 has a design capacity of 900 gpd). According to the
manufacturer, the flux rate of the membrane, which is dependent on the composition
and viscosity of the alkaline cleaner feed, is approximately 50 to 100 gallons per
square foot per day (GFD).
Figure 1. Diagram of the Membralox® Silverback™ Technology
2 This drain is not used at the Gates Rubber. No floating oils are removed.
3
-------
Revision 0 - 4/4/00
The Membralox® Silverback™ Model 900 is equipped with a backpulsing device
that is designed to clean the filtration modules during operation by periodically
pushing solution under pressure against the normal filtration direction. The
frequency and duration of the backpulse cycle is timer activated and controlled.
During the backpulse, a volume of approximately 700 ml of permeate is pushed back
through the ceramic membranes in about one tenth of a second. The unit then
returns to its normal flow pattern.
2.3 Commercial Status
The Membralox® Silverback™ Technology is fully commercialized.
Approximately 500 units have been sold.
2.4 Pollution Prevention Classification
The Membralox® Silverback™ Model 900 is a bath maintenance technology. Bath
maintenance refers to a range of pollution prevention practices and technologies that
preserve or restore the operating integrity of metal finishing process solutions,
thereby extending their useful lives. Due to rising costs for chemicals, energy, and
treatment/disposal and increasingly more stringent environmental requirements, bath
maintenance has become a greater priority to metal finishing companies, and the
methods and technologies they employ have increased in sophistication. Today,
firms are willing to expend significant amounts of capital and operating funds for
equipment and methods that primarily reduce the disposal frequency of their baths.
In addition to extending bath lives, solution maintenance often improves the average
operating efficiency and effectiveness of a process solution and therefore has a
positive impact on production rates and finish quality.
2.5 Environmental Significance
The Membralox® Silverback™ Model 900 is employed to reduce the disposal
frequency of alkaline cleaning baths. Alkaline cleaning baths are used by nearly all
metal finishing companies to remove oil and other soils prior to electroplating,
anodizing, painting, or other finishing processes. These processes are very sensitive
to presence of surface contaminants. Usually, alkaline cleaning baths are used until
they become moderately contaminated with oil and suspended solids, and then they
are discarded and replaced with fresh cleaning chemicals. If used beyond a certain
point, the contamination will prevent good surface cleaning, which may lead to
defects in a part's finish. Also, contaminants in the cleaning baths can carry over to
subsequent tanks and cause deterioration of those solutions. Spent cleaning baths
can be hauled to a treatment, storage, and disposal (TSD) facility or treated on-site.
The off-site treatment/disposal option is typically expensive, and on-site treatment
can present difficulties due to the presence of chelating compounds found in most
cleaners. These compounds tend to prevent complete precipitation of heavy metals
and therefore may interfere with a company's ability to meet effluent guidelines.
Also, treatment of cleaning baths generates significant quantities of sludge.
The disposal frequency of cleaning baths varies from company to company and
depends on several factors, including type of cleaner, type of soil, soil loading rate,
and cleanliness requirements. Prior to implementing a bath maintenance technology,
4
-------
Revision 0 - 4/4/00
Gates Rubber Company disposed of their cleaning baths (approximately 2,500 gal.)
approximately every two weeks. Subsequent to implementation of the technology,
the disposal frequency has been reduced to every six months. Additionally, the
contents of the Silverback™ recirculation tank are discarded every week (100 gal.).
Records from Gates Rubber Company will be reviewed during testing and
summarized in the verification report as background information. These records are
extensive and include a time period prior to installation of alkaline cleaner recycling
equipment. The verification report will include a statement clearly indicating that
these data are present for the purpose of providing background information and have
not been verified.
2.6 Local Installation
The Gates Rubber Company located in Versailles, MO, was selected by USFilter as
the test site for the Membralox® Silverback™ Model 900. The unit that will be
evaluated during this ETV-MF project was installed in September 1999. However,
Gates Rubber has previous experience with the Membralox® Silverback™, dating
back to 1997. The Gates Rubber Company manufactures a wide range of hydraulic
couplings and hose assemblies. They have an 115,000 ft2 manufacturing facility.
Alkaline cleaning is performed at various points in the plant. There are 12 in-
process cleaning tanks present in areas such as machining. Eleven of these units
hold 40 gal. of alkaline cleaner and one holds 75 gal. The largest cleaning operation
is located on the barrel plating (zinc) line, where there is a 1,800 gal. soak cleaning
tank and a 1,800 gal. electrocleaning tank. The 12 in-process cleaning tanks and the
soak cleaning tank are plumbed into the cleaner recycling system that will be tested
during this project. The electroclean tank is serviced by a separate recycling system
and will not be a part of this effort.
The predominant oil found on the parts processed through the cleaning systems is a
cutting oil (Tru Cut 2582). The MSDS for this product indicates it is a petroleum
hydrocarbon with additives. Its hazardous components are hydrotreated light
napthenic oils and hydrotreated heavy napthenic oils. Laboratory analyses show that
it is a mixed base mineral oil that contains an ester-based additive. Other additives
include a phenolic anti-oxidant. Traces of cellulose or a derivative thereof are also
present.
A diagram of the closed-loop cleaner recycling system involved in this ETV-MF
project is shown in Figure 2. The solution that is processed by the Silverback™ unit
comes from three sources, including the soak clean tank and drag-out rinse tank on
the zinc plating line and from in-process washers located in various parts of the
manufacturing facility. Each of these three sources is connected by piping
(represented by lines with arrows in Figure 2) to the storage tank.
5
-------
Revision 0 - 4/4/00
Figure 2. Alkaline Cleaner Recycling at Gates Rubber Company
Deionized water (spray) ^
Soak
Drag-Out
Clean
Rinse
Tank
Tank
In-Process
Washers
Return product to soak clean
tank and in-process washers
1J = Feed sample port
\S 1j = Product sample port
= Totalizer measuring volume of product produced
V ^J = Pressure gage (pump discharge pressure)
© = Pressure gage (downstream pressure)
Sllverback IM Unit
2-compartment tank
Drain
6
-------
Revision 0 - 4/4/00
The solution conveyed from in-process cleaning and the soak clean tank is used
alkaline cleaner. The solution conveyed from the drag-out rinse tank to the storage
tank is rinse water that contains alkaline cleaner. De-ionized (DI) water is added to
the drag-out rinse tank from spray bars located above the drag-out rinse tank. The
addition of this water causes the solution in this tank to overflow a weir and the
overflow is conveyed to the storage tank. The DI water sprays are automatically
activated when a barrel is removed from the drag-out rinse tank.
The solution that is in the storage tank is pumped to the Silverback™ unit and is
processed at a rate of approximately 1.1 gpm. The recovered product from the
Silverback™ unit is pumped to a recovered product tank, from where it is drawn-off
and reused in the soak clean tank and in-process washers.
Barrels exiting the drag-out rinse tank are further processed in an electro-cleaning
tank and subsequently zinc plated. The electro-cleaning tank is not connected to the
recovery system being tested during this project.
The alkaline cleaner used at this facility is CLEAN-R-120GR, which is formulated
and sold by PAVCO, a company located in Cleveland, OH. The MSDS for this
product can be found in Appendix A. The concentration of the cleaner is controlled,
based on alkalinity concentration. The product is purchased as a single component
concentrated liquid. When measurements indicate that the alkalinity concentration is
below a recommended operating level, additional liquid concentrate is added to the
cleaning system.
PAVCO indicated that the cleaner is composed primarily of the following chemicals
or classes of chemicals: sodium hydroxide, phosphate, glycol ether, octylphenol, and
amine.
3.0 EXPERIMENTAL DESIGN
3.1 Test Goals and Objectives
The overall goal of this ETV-MF project is to evaluate the ability of the
Silverback™ unit to recover alkaline cleaning chemistry that is used under actual
production conditions and to evaluate and characterize the operation of the unit
through measurement of various process parameters.
The following is a summary of primary project objectives. Under normal system
operating setpoints for the installation at Gates Rubber Company and varying
contaminant-loading rates:
• Prepare a material balance for alkaline cleaner constituents and soils in order to:
1. Evaluate the ability of the Silverback™ unit to recover alkaline cleaner.
2. Evaluate the ability of the Silverback™ unit to remove contaminants from
used cleaning solution.
• Determine the cost of operating the alkaline cleaning recovery system for the
specific conditions encountered during testing.
1. Determining labor requirements needed to operate and maintain the
Silverback™ unit.
7
-------
Revision 0 - 4/4/00
2. Determining the quantity of energy consumed by the Silverback™ unit
during operation.
• Quantify the environmental benefit by determining the reduction in bath
disposal frequency.
3.2 Critical and Non-Critical Measurements
Measurements that will be taken during testing are classified below as either critical
or non-critical. Critical measurements are those that are necessary to achieve the
primary project objectives. Non-critical measurements are those related to process
control or general background readings.
Critical Measurements:
• solution processing rate (gallons per minute)
• chemical characteristics of feed and product solutions (alkalinity, phosphate,
glycol, total phenol, total kjeldahl nitrogen (TKN), ammonia, oil and grease
(oil), total solids, and total suspended solids (TSS))
• concentrated cleaning solution additions (gallons) and related costs
• chemical characteristics of concentrated cleaning solution (alkalinity,
phosphate, glycol, total phenol, total kjeldahl nitrogen (TKN), ammonia, oil and
grease (oil), total solids, and total suspended solids (TSS))
• waste volumes (gallons) and chemical characteristics (alkalinity, phosphate,
glycol ether, octylphenol, oil, total solids, and TSS ) of those wastes
• chemical characteristics of feed and product solutions (alkaline cleaner
components and contaminants)
• cleaning chemical additions (gallons) and related costs
• O&M labor requirements (hours) and costs
• production throughput measured in terms of square footage of metal parts
cleaned
• energy use for pumps (kWh) and reheating of recycled solution (BTUs) and
costs
Non-Critical Measurements:
• solution temperature (degrees Fahrenheit) in second compartment (recirculation
tank) of Silverback™ unit
• differential pressure (psig) across the membrane (pump discharge
pressure/downstream pressure)
• backpulse frequency (minutes) and duration (seconds) settings
3.3 Test Matrix
Testing will be conducted in two distinct test periods, with each test period having a
duration of approximately one week. The soil loading of the unit will be varied from
the first test period to the second by a factor of approximately two.
During the first test period, the entire cleaning line and alkaline recycling system
will be operated using normal operating conditions found at Gates Rubber Company
(see section 2.6).
8
-------
Revision 0 - 4/4/00
Therefore, during the first test period, the soil concentration in the Silverback™ unit
feed solution will be at a "normal" level. This "normal" level will be quantitatively
determined during the first test period by collecting and analyzing samples of the
feed solution.
At the completion of the first test period, the contents of the two-compartment tank
of the Silverback™ unit, which holds the soil removed from the alkaline cleaning
solution, will be removed from the Silverback™ unit.
During the second test period, the entire cleaning line and alkaline recycling system
will be operated using the normal operating procedures (see section 2.6), with one
exception. To evaluate the operation of the Silverback™ unit under a high soil
loading condition than is normally found at Gates Rubber Company, the soil that
was removed during the first test period from the Silverback™ unit will be bled into
the feed solution (bled at a uniform rate into the storage tank) during the entire
second test period. These procedures will approximately double the soil-loading rate
during the second test period. The actual soil-loading rate during the second test
period will be quantitatively determined by collecting and analyzing samples of the
feed solution.
The higher soil loading during the second test period may cause operational
problems, such as plugging the membrane. This could occur, for example, due to a
buildup of solids in the recirculation tank. Efforts will be made to conduct the
second test for a minimum of five days. Any maintenance requirements (e.g.,
cleaning of filter) or operational changes needed to keep the unit operating will be
conducted and recorded. However, if it is necessary to discharge the solution in the
recirculation tank (e.g., due to solids buildup) prior to operating for five days, then
the test will be considered complete. Any data collected during a shortened test
period will be considered usable in the verification report subject to data quality
requirements (see section. 4). The cause of the shortened test period will be noted
and discussed in the report.
Test objectives and measurements are summarized in Table 1.
3.4 Testing And Operating Procedures
3.4.1 Set-Up and System Initialization Procedures
Prior to startup, the recirculation tank will be drained and cleaned and the membrane
will be cleaned (see Appendix B, Silverback™ O&M manual for procedural details).
The recirculation tank will then be filled with cleaner solution from the feed tank and
the Silverback™ unit will be started according to instructions in the O&M manual
(see Appendix B). Sampling will proceed once the unit has been operating normally
for a period of at least one hour and the unit is determined to be operating properly
with a product (permeate) flow rate of 1.1 gpm. This flow rate is the target-
operating rate used by Gates Rubber Company.
9
-------
Revision 0 - 4/4/00
Table 1. Test Objectives and Related Test Measurements for Evaluation of the Membralox® Silverback™ Model 900
Test
Test Objective
Test Measurement
1. Baseline:
Normal soil
loading rate.
Prepare a material balance for alkaline
cleaner constituents and soils.
Processing rate of solution through unit.
Chemical characteristics of feed solution.
Chemical characteristics of recovered product.
Volume and chemical characteristics of wastes removed from two-
compartment tank.
Quantity of concentrated cleaning solution added during testing.
Chemical characteristics of concentrated cleaning solution.
Evaluate the ability of the Silverback™
unit to process used alkaline cleaner
solution and separate usable cleaner
solution chemistry from bath contaminants.
Chemical characteristics of feed solution.
Chemical characteristics of recovered product.
Determine the alkaline cleaner recovery
rate of the system, normalized based on
production throughput and soil loading.
Volume of product produced.
Production throughput for all associated cleaning baths.
Chemical characteristics of feed solution.
Chemical characteristics of recovered product.
Determine labor requirements needed to
operate and maintain the Silverback™ unit.
O&M labor required during test period.
Determine the quantity of energy
consumed by the Silverback™ unit during
operation.
Quantity of energy used by pumps and filtration module.
Determine the cost of operating the
alkaline cleaning recovery system for the
specific conditions encountered during
testing.
Costs of O&M labor, materials, and energy required during test period.
Quantity and price of fresh cleaning chemicals added during testing.
Quantify/identify the environmental
benefit.
Review historical waste disposal records and compare to current practices.
2. High Soil
Loading Rate:
Approx. two
times the normal
soil loading rate.
Same as baseline
Same as baseline
10
-------
Revision 0 - 4/4/00
3.4.2 System Operation
The Silverback™ unit will be operated according to procedures contained in the
USFilter O&M manual (see Appendix B). Sample collection and operating
parameter measurements will be performed as indicated in Section 3.5 of this test
plan. The unit will operate 24 hours/day for five days during each test run. The
product flow rate will be maintained at approximately 1.1 gpm.3 To maintain this
rate, periodic adjustments will be made according to the instructions found in
Appendix B.
The recirculation tank will not be discharged during a test run unless it is necessary
to continue the operation of the unit. For example, this could occur if the processing
rate could not be maintained at a minimum of 0.75 gpm due to solids buildup in the
membrane.
At the completion of the first test cycle, the unit will be drained and cleaned and
restarted following the same procedures as those used for set-up and system
initialization.
The solution removed from the recirculation tank after the first test period will be
stored in a tank located near the alkaline storage tank. To simulate a higher soil-
loading rate during the second test period, one-tenth of the volume of the stored
recirculation solution (about 10 gal.) will be transferred to the alkaline storage tank
twice each day (at approximately 9 am and 5 pm). Whenever solution is transferred
to the storage tank, the event will be recorded on the data collection form (i.e., time
and volume transferred).
3.4.3 Sampling and Process Measurements
Sampling and process measurements will be taken according to the schedule
presented in Table 2. Sampling events and process measurements will be recorded
on the form shown in Figure 3. Each laboratory sample bottle will be labeled with
the date, time, sample ID number, and test parameters required. Sample preparation
methods are described in each individual analytical method.
Samples to be analyzed at an off-site laboratory will be accompanied by a chain of
custody form. The samples will be transported in appropriate sample transport
containers (e.g., coolers with packing and blue ice) by common carrier. The transport
containers will be secured with tape to ensure sample integrity during the delivery
process to the analytical laboratory. The Project Manager or designee will perform
sampling, labeling, and ensure that samples are properly secured and transported to
AMTest, Inc. in Redmond, WA, for analysis.
3 The Model 900 is rated by the manufacturer to operate at a processing rate of 900 gpd, or approximately 0.75 gpm. The
operating rate of 1.1 gpm was selected by Gates Rubber Company. The manufacturer of the unit indicated that it is not
uncommon for this model to process solution at a higher rate than the design rate. Achievable processing rates depend on several
factors including the chemistry and concentration of the cleaner and the characteristics and loading of the soil.
11
-------
Revision 0 - 4/4/00
Table 2. Sampling Frequency and Parameters to be Measured
Sample Location
Frequency/Type
Parameters
Feed Sample Port
1/day (24-hr. composites)
alkalinity, phosphate, glycol, total phenol,
total kjeldahl nitrogen (TKN), ammonia, oil
and grease (OIL ), total solids, and total
suspended solids (TSS)
Silverback™ Unit
(right side of two-
compartment tank,
where waste product
accumulates)
1/week (grab at completion of test
run)
alkalinity, phosphate, glycol, total phenol,
total kjeldahl nitrogen (TKN), ammonia, oil
and grease (OIL ), total solids, and total
suspended solids (TSS)
Silverback™ Unit
(right side of two-
compartment tank)
l/lst shift
temperature
Membrane Module
Pressure Gauges
3/lst shift
l/2nd shift
membrane pressures (pump
discharge/permeate)
Product Sample Port
1/day (24-hr. composites)
(alkalinity, phosphate, glycol, total phenol,
total kjeldahl nitrogen (TKN), ammonia, oil
and grease (OIL ), total solids, and total
suspended solids (TSS))
Product Line
3/lst shift
l/2nd shift
processing rate (instantaneous product flow
and total product flow)
Concentrated cleaning
solution
As needed.
volume added (gallons)
Concentrated cleaning
solution
1 time during test
alkalinity, phosphate, glycol, total phenol,
total kjeldahl nitrogen (TKN), ammonia, oil
and grease (OIL ), total solids, and total
suspended solids (TSS)
Alkaline Cleaning
Tank
l/lst shift
temperature
Product Storage Tank
l/lst shift
temperature
Sample ports have been installed to collect feed and product samples from the
Silverback™ unit. To collect the composites (24-hr. composites), polyethylene
tubes will be connected to the two sample ports and directed into 2.5 or five-gallon
HDPE containers that are maintained at a cool temperature by placing the glass
sample collection container in ice.
The samples will collect in the HDPE containers at a uniform rate (approximately 5
to 8 ml/min) through the 24-hour period (controlled by valve on sample ports). At
the end of each 24 hr. sampling period, the HDPE will be labeled and properly
stored.
12
-------
Revision 0 - 4/4/00
Figure 3. Data Collection Form
Test Number: Page of
Start Date:
Start Time:
Date/
Time
Initials
Product
Flow,
gpm
Product
Flow,
(totalalizer)
gal
Membrane Pressures
Temperature in
Recirculation
Tank, °F
Temperature
in Recovered
Product Tank,
°F
Temperature
in Soak
Clean Tank,
°F
Sample
No.
Sample
Location
Notes and Observations
Pump
Discharge,
psi
Permeate,
psi
13
-------
Revision 0 - 4/4/00
The waste product will be sampled from the recirculation tank, just prior to draining
the unit. This sample will be collected while recirculation pump is running to assure
that the contents of the recirculation tank are completely mixed. Fresh alkaline
cleaning chemistry will be sampled from its storage container.
The temperature of the recirculation tank solution, soak clean tank solution, and
recovered product tank solution will be measured using a thermometer (NSF
certified). The membrane pressure readings will be taken from instrumentation
integrated on the Silverback™ unit. The instantaneous flow rate and total flow of
processed solution will also be taken from instrumentation integrated on the
Silverback™ unit. The readings from the flow totalizer will be used to determine the
quantity of solution processed by the unit. Due to the importance of this
measurement, the accuracy of the flow totalizer will be determined by comparing the
recorded flow rate to a manual "stopwatch and bucket" measurement. Based on
these results, recorded values will be corrected, if necessary.
Electricity use will be calculated by determining the power requirements and cycle
times of pumps and other powered devices. Gates Rubber Company will provide the
cost of labor, electricity, and other items needed for a cost analysis.
3.4.4 Production Measurements
Gates Rubber Company routinely monitors processing volumes of each type of part
(referred to as "group codes"). Also, they have available average surface area data
for each group code (in2) and type of metals processed. This information will be
used to calculate the square footage of parts cleaned each day.
Analytical Procedures
3.5.1 Alkaline Cleaner Constituents
Alkaline cleaners used for metal cleaning are formulated from three separate
chemical groups, some of which tend to overlap in function. These three groups are
builders, surfactants, and additives.
Builders are inorganic metal salts that provide alkalinity to the cleaning solution.
Typical builders include hydroxide, silicates (e.g., sodium metasilicate, sodium
sesquisilicate, and sodium orthosilicate), phosphates (e.g., tetrasodium
pyriphosphate, sodium tripolyphosphate, and sodium hexametaphosphate), sodium
carbonate (i.e., soda ash), and sodium tetraborate (i.e., borax). During cleaning, the
role of a builder/alkali is saponification of fatty oil, where insoluble fatty oil is
converted to soluble glycerin and soluble soap. The process of saponification
consumes alkali. [Ref. 1]
Surfactants (i.e., surface-active agents) are organic molecules that have two ends: a
hydrophilic (water loving) end and a hydrophobic (water hating) end. Surfactants
group together to form structures known as micelles, which are responsible for the
solubilization of oil. The size of the surfactant micelle grows as it solubilizes oil and
contaminants, creating an emulsion. During microfiltration, larger micelles (those
which have solubilized oil) are retained on the concentrate side of the filter, while
the smaller micelles pass through. [Ref. 1]
14
-------
Revision 0 - 4/4/00
Additives are more vague in function than builders and surfactants. Often additives
are used to promote the sequestration of metals in solution or aid in rust prevention.
[Ref. 1]
The constituents of the PAVCO cleaner are (1) builders: sodium hydroxide and
phosphate (2) surfactants: glycol ether and octylphenol, and (3) additive: amine.
To monitor for changes in builders, tests will be performed to measure alkalinity
(EPA 310.1) and phosphates (EPA 200.7). To monitor for changes in surfactants,
tests will be performed for glycol (GC-FID) and total phenol (EPA 420.2). Amine
will be monitored by testing for total kjeldahl nitrogen (TKN) (EPA 351.3) and
ammonia (EPA 350.1).
Regarding the method for analysis of glycol, the laboratory is working with samples
to refine a matrix specific extraction and preparation procedure. Exact method
specifics and details will be summarized by the laboratory and presented in the
verification test report.
A summary of analytical tests is presented in Table 3.
3.5.2 Oil
Oil is contributed to the cleaner bath when parts are processed. The oil is a
combination of machining and cutting oils and coolants that are used in
metalworking. These fluids may contain mineral oils, natural oils, fats and
derivatives, or synthetic lubricants. Oil loading and the efficiency of oil separation
will be measured by performing oil measurements on both the feed and product
streams of the Silverback™ unit.
The analytical method for measuring oil that will be used is a modified organics and
hydrocarbon gas chromatographic method, EPA Method 8015 (modified), the
Northwest Total Petroleum Hydrocarbon - Extended Diesel (NWTPH-DX) method
(See Appendix F). A gravimetric method for measuring both oil and grease in
aqueous and sludge samples was not chosen for analytical testing. Concerns about
possible interference's and "false positives" for oil concentration from surfactants
and/or proprietary chemicals within the system under evaluation led to consider
alternative analytical methods, or at least to demonstrate the efficacy of standard
methods to these materials and sample matrices.
15
-------
Revision 0 - 4/4/00
Table 3. Summary of Analytical Tests and Requirements
Parameter
Test Method
Sample Bottle
Sample
Volume
Required
Preservation
/ Handling
Hold
Time
Tests applicable to tracking contaminants entering/removed from the alkaline cleaner:
Oil
EPA 8015
modified
(NWTPH-DX)
2.5 or 5-gal HDPE
150 ml
Acidify with 1
mL conc. HC1
at lab, cool
storage (<4°C)
28 days
Total
Solids
EPA Method
160.3
2.5 or 5-gal HDPE
150 ml
cool storage
(<4°C)
7 days
TSS
EPA Method
160.2
2 .5 or 5-gal HDPE
same as total
solids
cool storage
(<4°C)
7 days
Tests applicable to tracking constituents of the alkaline cleaner:
Alkalinity
EPA 310.1
2.5 or 5-gal HDPE
500 ml
cool storage
(<4°C)
14 days
Phosphate
EPA 200.7
same bottle as oil
125 ml
Adjust to pH
<2 at lab
with conc.
h2so4
28 days
Glycol
GC-FID
2.5 or 5-gal HDPE
500 ml
cool storage
(<4°C)
28 days
Total
Phenol
EPA 420.2
same bottle as oil
125 ml
cool storage
(<4°C),
adjust to pH
<2 at lab
with conc.
h2so4
28 days
Amine:
TKN
EPA 351.3
same bottle as oil
125 ml
cool storage
(<4°C),
adjust to pH
<2 at lab
with conc.
h2so4
28 days
Amine:
NH3
EPA 350.1
same bottle as oil
125 ml
cool storage
(<4°C),
adjust to pH
<2 at lab
with conc.
h2so4
28 days
"Neat" samples of the exact formulated oil lubrication products, which are used on
the parts themselves, will be characterized and used as calibration standards in these
quantitative analyses (see appendix for chromatograms). Therefore, by using known
dilutions of "neat" standards, calibration curves and reference solutions can be
16
-------
Revision 0 - 4/4/00
drawn (reference solutions were used for quantification purposes). EPA Method
8015 can then be used to quantify the oil in the cleaner separate from the PAVCO
cleaner compounds. A copy of the method is attached for reference, (see Appendix
F)
An initial characterization and evaluation of these "neat" formulated products using
the modified Method 8015 was performed by the analytical laboratory (AMTest, Inc.
of Redmond, WA). Modifications to the Method 8015 involved slight changes in the
ramp time within the gas chromatographic program, which were within the
proscribed acceptable method modifications. Each type of oil evaluated yielded a
characteristic chromatographic signature. Based on the information received, no one
particular oil product is known to dominate over the others. Using the aliquots from
the neat solutions of the different formulated products, a mixed reference standard
was created and a range of calibration concentration standards derived. Results are
reported in milligrams/liter (mg/L) oil/grease. Therefore, by developing calibration
curves for the oils used at Gates Rubber, EPA Method 8015 can quantify the oil in
the cleaner separate from the cleaner compounds used at Gates Rubber. Preliminary
samples were collected from Gates Rubber and analyzed. The analysis results
verified the Method 8015 (modified) could successfully quantify the oils.
Another reference step evaluated the efficacy of the modified Method 8015 test
method for these samples and matrices. Aliquots were analyzed using the modified
Method 8015, the conventional freon extraction-gravimetric method, as well as the
recently approved EPA Method 1664 (hexane extract). The modified Method 8015
and freon methods yielded comparable results. Although the freon method yielded
acceptable results, freon has been phased out as an acceptable material under the
Montreal Accord, and hence will not be in use within analytical methods in the very
near future. The hexane extraction method did not yield successful extraction
results. The aqueous matrix turned milky (akin to liquid gelatin), requiring several
cleanup steps and resulting poor sample recovery (un-reproducibility of results).
Due to the arduous sample preparation that would have been required, these test
samples were not carried through to analyses. The cost for such analyses would be
prohibitively expensive, so the hexane extraction method was discounted from
further consideration in this study.
3.5.3 Total Solids and Suspended Solids
Solid material is contributed to the cleaning bath when parts are processed. These
particles are known to impair the efficiency of the cleaner. The Silverback™ unit
removes larger particles in the filter basket (25 |i) when solution first enters the unit,
and it retains smaller particles in the recirculation tank during microfiltration (0.2 |i
pore size).
To determine the effectiveness of the Silverback™ unit with regard to removal of
particulates, tests for total filterable residue (EPA 160.2) will be performed. The
referenced method produces values commonly referred to as total suspended solids
(TSS). The total residue test (EPA 160.3) will be used in material balance
17
-------
Revision 0 - 4/4/00
calculations. The associated calculations are presented in Section 4.2 Data
Reduction, Validation, and Reporting.
QUALITY ASSURANCE/QUALITY CONTROL REQUIREMENTS
Quality Assurance/Quality Control activities will be performed according to the applicable
section of the Environmental Technology Verification Program Metal Finishing
Technologies Quality Management Plan (ETV-MF QMP) [Ref. 3],
4.1 Quality Assurance Objectives
The first QA objective is to ensure that the process operating conditions and test
methods are maintained and documented throughout each test and laboratory
analysis of samples. The second QA objective is to use standard test methods
(where possible) for laboratory analyses. The test methods to be used are listed in
Table 3.
4.2 Data Reduction, Validation, and Reporting
4.2.1 Internal Quality Control Checks
Raw Data Handling. Raw data is generated and collected by laboratory analysts at
the bench and/or sampling site. These include original observations, printouts, and
readouts from equipment for sample, standard, and reference QC analyses. Data is
collected both manually and electronically. At a minimum, the date, time, sample
ID, instrument ID, analyst ID, raw signal or processed signal, and/or qualitative
observations will be recorded. Comments to document unusual or non-standard
observations also will be included on the forms, as necessary. The form presented in
Figure 3 will be used for recording data on-site. Additionally, process logs
generated by Gates Rubber, an ISO 9001 certified company, will be copied to
supplement these forms. These logs show, among other items, chemical additions
and production throughput. These forms will be reviewed and copied on a daily
basis.
The on-site Project Team member will generate Chain of custody forms and these
forms will accompany samples when they are shipped off-site.
Raw data will be processed manually by the analyst, automatically by an electronic
program, or electronically after being entered into a computer. The analyst will be
responsible for scrutinizing the data according to laboratory precision, accuracy, and
completeness policies. Raw data bench sheets and calculation or data summary
sheets will be kept together for each sample batch. From the standard operating
procedure and the raw data bench files, the steps leading to a final result may be
traced. The ETV-MF Program Manager will maintain process-operating data for use
in verification report preparation.
Data Package Validation. The generating analyst will assemble a preliminary data
package, which shall be initialed and dated. This package shall contain all QC and
raw data results, calculations, electronic printouts, conclusions, and laboratory
sample tracking information.
18
-------
Revision 0 - 4/4/00
A second analyst will review the entire package and check sample and storage logs,
standard logs, calibration logs, and other files, as necessary, to ensure that all
tracking, sample treatments, and calculations are correct. After the package is
reviewed in this manner, a preliminary data report will be prepared, initialed, and
dated. The entire package and final report will be submitted to the Laboratory
Manager.
The Laboratory Manager shall be ultimately responsible for all final data released
from the laboratory. The Laboratory Manager or designee will review the final
results for adequacy to task QA objectives. If the manager or designee suspects an
anomaly or non-concurrence with expected or historical performance values, or with
task objectives for test specimen performance, the raw data will be reviewed, and the
generating and reviewing analysts queried. If suspicion about data validity still
exists after internal review of laboratory records, the manager will authorize a re-
test. If sufficient sample is not available for re-testing, a re-sampling shall occur. If
the sampling window has passed, or re-sampling is not possible, the manager will
flag the data as suspect. The Laboratory Manager signs and dates the final data
package.
Data Reporting. A report signed and dated by the Laboratory Manager will be
submitted to the ETV-MF Project Manager. The ETV-MF Project Manager will
decide the appropriateness of the data for the particular application. The final report
contains the laboratory sample ID, date reported, date analyzed, the analyst, the SOP
used for each parameter, the process or sampling point identification, the final result,
and the units. The ETV-MF Program Manager shall retain the data packages as
required by the ETV-MF QMP [Ref. 3],
4.2.2 Calculation of Data Quality Indicators
Analytical performance requirements are expressed in terms of precision, accuracy,
representativeness, comparability, completeness, and sensitivity (PARCCS).
Summarized below are definitions and QA objectives for each PARCCS parameter.
Duplicates and spike duplicates will be performed on one out of every ten samples.
Sample splitting will occur in the analytical laboratory.
Matrix spike/spike duplicates will be performed at a frequency of 10% (one out of
every ten samples), as specified in the modified EPA 8015 method protocol.
Additional spikes and duplicates will be performed where applicable, with respect to
the analyte of interest. Please refer to the methods for reference.
With respect to programmatic duplicates, field splits (duplicates) of pre-treatment
(and post-treatment) samples will be performed at a frequency of one per field
sample batch processed and/or ten field samples of a given type (e.g., at least one per
ten days of field samples), within the prescribed holding times and method
requirements.
4.2.2.1 Precision
Precision is a measure of the agreement or repeatability of a set of replicate
results obtained from duplicate analyses made under identical conditions.
Precision is estimated from analytical data and cannot be measured directly.
19
-------
Revision 0 - 4/4/00
The precision of a duplicate determination can be expressed as the relative
percent difference (RPD), and calculated as:
RPD = {(|Xi - X2|)/(Xi + X2)/2} x 100 =
|X:"X2
(X, +X2)
x 100
where, Xi = larger of the two observed values and X2 = smaller of the two
observed values.
Multiple determinations will be performed for each test on the same test
specimen. The replicate analyses must agree within the relative percent
deviation limits provided in Table 4.
4.2.2.2 Accuracy
Accuracy is a measure of the agreement between an experimental
determination and the true value of the parameter being measured. Accuracy is
estimated through the use of known reference materials or matrix spikes. It is
calculated from analytical data and is not measured directly. Spiking of
reference materials into a sample matrix is the preferred technique because it
provides a measure of the matrix effects on analytical accuracy. Accuracy,
defined as percent recovery (P), is calculated as:
P =
(SSR-SR)
SA
x 100
where: SSR = spiked sample result
SR = sample result (native)
SA = the concentration added to the spiked sample
Analyses will be performed with periodic calibration checks with traceable
standards to verify instrumental accuracy. These checks will be performed
according to established procedures in the contracted laboratory(s) that have
been acquired for the U.S. Filter verification test. Analysis with spiked samples
will be performed to determine percent recoveries as a means of checking
method accuracy. QA objectives will be satisfied if the average recovery is
within the goals described in Table 4.
20
-------
Revision 0 - 4/4/00
Table 4. QA Objectives
Critical
Measurements
Matrix
Method
Reporting
Units
Method of
Determination
MDL
Precision
(RPD)
Accuracy
(%
Recovery)
Completeness
Oil
Water
EPA Method
8015
modified
mg/1
GC-FID
0.2
<30
50 - 150
95
Total Solids
Water
EPA Method
160.3
mg/1
Gravimetric
1.0
<30
80 - 120
95
TSS
Water
EPA Method
160.2
mg/1
Gravimetric
1.0
<30
80 - 120
95
Alkalinity
Water
EPA 310.1
mg/1
Titration
1.0
<30
8-120
95
Phosphate
Water
EPA 200.7
mg/1
0.005
<30
8-120
95
Glycol
Water
#1
mg/1
GC-FID
<30
8-120
95
Total Phenol
Water
EPA 420.2
mg/1
0.005
<30
8-120
95
TKN
Water
EPA 351.3
mg/1
0.25
<30
8-120
95
NH3
Water
EPA 350.1
mg/1
0.005
<30
8-120
95
Temperature
Water
Thermometer
°C (°F)
-
-
-
1
100
Chemical
additions
Water
gallons
-
-
-
"
Flow
Water
Flow meter
(totalizer)
gallons
-
-
-
#2
"
1. Glycol will be determined using a matrix specific GC-FID analysis (no applicable method number). Exact method specifics and
details will be presented in the verification testing report.
2. Accuracy will be verified by stopwatch and bucket method.
21
-------
Revision 0 - 4/4/00
4.2.2.3 Completeness
Completeness is defined as the percentage of measurements judged to be valid
compared to the total number of measurements made for a specific sample matrix
and analysis. Completeness is calculated using the following formula:
Completeness = Valid Measurements x 100
Total Measurements
Experience on similar projects has shown that laboratories typically achieve about
90 percent completeness. QA objectives will be satisfied if the percent completeness
90 percent or greater as specified in Table 4.
4.2.2.4 Comparability
Comparability is another qualitative measure designed to express the confidence
with which one data set may be compared to another. Sample collection and
handling techniques, sample matrix type, and analytical method all affect
comparability. Comparability is limited by the other PARCCS parameters because
data sets can be compared with confidence only when precision and accuracy are
known. Comparability will be achieved in the U.S. Filter technology verification by
the use of consistent methods during sampling and analysis and by traceability of
standards to a reliable source.
4.2.2.5. Representativeness
Representativeness refers to the degree to which the data accurately and precisely
represents the conditions or characteristics of the parameter represented by the data. For
the purposes of this demonstration, representativeness will be achieved by presenting
identical analyte samples to the specified lab(s) and executing consistent sample
collection and mixing procedures.
4.2.2.6. Sensitivity
Sensitivity is the measure of the concentration at which an analytical method can
positively identify and report analytical results. The sensitivity of a given method is
commonly referred to as the detection limit. Although there is no single definition of
this term, the following terms and definition of detection will be used for this
program.
Instrument Detection Limit (IDL) is the minimum concentration that can be
measured from instrument background noise
4.2.2.7 Method Detection Limit
(MDL) is a statistically determined concentration. It is the minimum concentration
of an analyte that can be measured and reported with 99 percent confidence that the
analyte concentration is greater than zero as determined in the same or a similar
matrix [because of the lack of information on analytical precision at this level,
sample results greater than the MDL but less than the practical quantification limit
(PQL) will be laboratory qualified as "estimated"]
22
-------
Revision 0 - 4/4/00
MDL is defined as follows for all measurements:
MDL = t(n-l,l-
-------
Revision 0 - 4/4/00
4.2.2.10 Energy Use
Energy requirements for the Silverback™ unit will be calculated by summing the total
quantity of horsepower hours and dividing by 1.341 HP-hr/kWh to arrive at
electricity needs. Volumes collected in gallons will be mathematically converted to
liters prior to calculation.
Heat required to raise solution temperature
lb B TTJ
(BTUs/cycle) = Prodvoi x 2.2 — x AT x 1
L lb°F
Where: Prodvoi = product volume collected during the cycle (liters)
AT = average temperature difference between recovered product tank and
soak clean tank (F°)
4.2.2.11 Cost Analysis
This analysis will quantify the cost benefit of the technology. The costs for
operating the Silverback™ unit at Gates Rubber Company will be calculated to
operating costs for a time period prior to installation of the unit at Gates Rubber
Company. For the baseline conditions, the most recent applicable data available
from Gates Rubber Company will be used. The cost analysis will compare
operating costs, including costs for: alkaline cleaner solution, drag-out losses, other
materials, waste treatment/disposal, labor, and utilities.
4.2.2.12 Waste Generation Analysis
This analysis will quantify the environmental benefit of the technology. The waste
generation rates for operating the alkaline cleaner system with the Silverback™ unit
at Gates Rubber Company will be calculated and compared to waste generation
rates for a time period to installation of the unit at Gates Rubber Company. For the
baseline conditions, the most recent applicable data from Gates Rubber Company
will be used. The waste generation analysis will consider type/characteristics of
wastes generated and volume and frequency of waste generated.
4.3 Quality Audits
Technical System Audits. An audit will be performed during verification testing by
the CTC QA Manager according to Section 2.9.3 Technical Assessments of the
ETV-MF QMP [Ref. 3] to ensure testing and data collection are performed
according to the test plan requirements. In addition to the CTC Technical System
Audit, the EPA Quality Assurance Manager may also conduct an audit to assess the
quality of the verification test.
24
-------
Revision 0 - 4/4/00
Internal Audits. In addition to the internal laboratory quality control checks,
internal quality audits will be conducted to ensure compliance with written
procedures and standard protocols.
Corrective Action. Corrective Action for any deviations to established quality
assurance and quality control procedures during verification testing will be
performed according to section 2.10 Quality Improvement of the ETV-MF QMP
[Ref. 3],
Laboratory Corrective Action. Examples of non-conformances include invalid
calibration data, inadvertent failure to perform method specific QA, process control
data outside specified control limits, failed precision and/or accuracy indicators, etc.
Such non-conformances will be documented on a standard laboratory form.
Corrective action will involve taking all necessary steps to restore a measuring
system to proper working order and summarizing the corrective action and results
of subsequent system verifications on a standard laboratory form. Some non-
conformances are detected while analysis or sample processing is in progress and
can be rectified in real time at the bench level. Others may be detected only after a
processing trial and/or sample analyses are completed. Typically, these types of
non-conformances are detected by the Laboratory Manager. In all cases of non-
conformance, sample re-analysis will be considered as one source of corrective
action by the Laboratory Manager. If insufficient sample is available or the holding
time has been exceeded, complete re-processing may be ordered to generate new
samples if a determination is made by the Task Leader that the non-conformance
jeopardizes the integrity of the conclusions to be drawn from the data. In all cases,
a non-conformance will be rectified before sample processing and analysis
continues.
5.0 PROJECT MANAGEMENT
5.1 Organization/Personnel Responsibilities
The ETV-MF Project Team that is headed by CTC will conduct the evaluation of
the Membralox® Silverback™ Model 900. The ETV-MF Program Manager, Donn
Brown, will have ultimate responsibility for all aspects of the technology
evaluation. The ETV-MF Project Team Leader assigned to this evaluation is
George Cushnie. Mr. Cushnie and/or his staff member will be on-site throughout
the test period and will conduct or oversee all sampling and related measurements.4
USFilter will be responsible for training assigned personnel on operation of the
technology and will assist in startup of the system if necessary. They will also be
on-call during the test period for response in the event of equipment problems.
Gates Rubber Company personnel will be responsible for operation of the
Silverback™ equipment, related cleaning lines, and ancillary equipment such as
4 The CTC ETV-MF Program Manager, Donn Brown, will make a determination as to the qualifications of any staff member
assigned to the project. This will occur prior to testing.
25
-------
Revision 0 - 4/4/00
pumps and system instrumentation. With instruction from the ETV-MF Project
Team Leader or staff member, periodically Gates Rubber Company personnel will
collect samples or record data from instrumentation readings. Gates Rubber
Company personnel will also provide safety training as described in Section 9.0.
The ETV-MF Team Leader and the Gates Rubber Company have the authority to
stop work when unsafe or unacceptable quality conditions arise.
AMTest Laboratories is responsible for analyzing verification test samples. The
Laboratory Manager, Kathy Fugiel, will be point of contact. AMTest Laboratories
is accredited by the State of Washington, Department of Ecology for the analyses
identified in this test plan.
The ETV-MF Project Manager and Gates Rubber have the authority to stop work
when unsafe or unacceptable quality conditions arise. The CTC ETV-MF Program
Manager will provide periodic assessments of verification testing to the EPA ETV
Pilot Manager.
6.0 EQUIPMENT AND UTILITY REQUIREMENTS
All equipment and utilities required for this ETV-MF project are owned by and are
currently installed at Gates Rubber Company. These utility requirements include:
• Electrical supply to panel: 560VAC, 60 Hz, three-phase
• Instrument air: <1 scfm, 80 psi, dry, oil-free
• Steam: 6 lb./hr., or
Hot Water: 160°F
7.0 HEALTH AND SAFETY PLAN
This Health and Safety Plan provides guidelines for recognizing, evaluating, and
controlling health and physical hazards throughout the workplace. More specifically, the
Plan specifies for assigned personnel, the training, materials, and equipment necessary to
protect themselves from hazards created by acids, and any waste generated by the process.
7.1 Hazard Communication
All personnel assigned to the project will be provided with the potential hazards,
signs and symptoms of exposure, methods or materials to prevent exposures, and
procedures to follow, if there is contact with a particular substance. The Gates
Rubber Company Hazard Communication Program will be reviewed during training
and will be reinforced throughout the test period. All appropriate MSDS forms will
be available for chemical solutions used during testing.
7.2 Emergency Response Plan
Gates Rubber Company has a contingency plan to protect employees, assigned
project personnel, and visitors in the event of an emergency at the facility. This
plan will be used throughout the project. All assigned personnel will be provided
with information about the plan during training.
26
-------
Revision 0 - 4/4/00
7.3 Hazard Controls Including Personal Protective Equipment
All assigned project personnel will be provided with appropriate personal protective
equipment (PPE) and any training needed for its proper use, considering their
assigned tasks. The use of PPE will be covered during training as indicated in
Section 9.0.
The following PPE will be required and must be worn at all times while in the
Gates Rubber facility: Steel toed footwear and eyeglasses with side splashguards.
The alkaline recycling system is essentially a closed-loop process and fully
contained within the manufacturing building. There are no apparent hazards to the
surrounding community due to operation or testing of the system.
7.4 Lockout/Tagout Program
No new equipment will be installed; therefore, a lockout/tagout program will not be
necessary.
7.5 Material Storage
Any materials used during the project will be kept in proper containers and labeled
according to Federal and State law. Proper storage of the materials will be
maintained based on associated hazards. Spill trays or similar devices will be used
as needed to prevent material loss to the surrounding area.
7.6 Safe Handling Procedures
All chemicals and wastes or samples will be transported on-site in non-breakable
containers used to prevent spills. Spill kits will be strategically located in the
project area. These kits contain various sizes and types of sorbents for emergency
spill clean up. Emergency spill clean up will be performed according to the
Emergency Response Plan.
8.0 WASTE MANAGEMENT
The alkaline recycling equipment will be tested on processes already in-place and operating
at Gates Rubber Company. This equipment currently generates a waste as a result of
weekly or biweekly discharging of the recirculation tank. This waste material is presently
treated on-site and discharged to the local POTW, in accordance with local, state, and
Federal laws.
During testing, no additional wastes will be generated other than the normal discharge of
the recirculation tank. Gates Rubber Company using their normal practices will handle this
waste. Therefore, no special or additional provisions for waste management will be
necessary.
9.0 TRAINING
It is important that the verification activities performed by the ETV-MF Program be
conducted with high quality and with regard to the health and safety of the workers and the
environment. By identifying the quality requirements, worker safety and health, and
environmental issues associated with each verification test, the qualifications or training
27
-------
Revision 0 - 4/4/00
required for personnel involved can be identified. Training requirements will be identified
using the Job Training Analysis (JTA) Plan [Ref. 5],
The purpose of this JTA Plan is to outline the overall procedures for identifying the hazards
and quality issues and training needs for each verification test project. This JTA Plan
establishes guidelines for creating a work atmosphere that meets the quality,
environmental, and safety objectives of the ETV-MF Pilot. The JTA Plan describes the
method for studying ETV-MF project activity and identifying training needs. The ETV-
MF Operation Planning Checklist (,Appendix C) will be used as a guideline for identifying
potential hazards, and the Job Training Analysis Form (,Appendix D) will be used to
identify training requirements. After completion of the form, applicable training will be
performed. Training will be documented on the ETV-MF Project Training Attendance
Form (,Appendix E). Health and safety training will be coordinated with Jerry Capps, the
Gates Rubber Company Human Resources Director.
10.0 REFERENCES
1. Peterson, David, "Practical Guide to Industrial Metal Cleaning," Hanser Gardner
Publications, Cincinnati, OH, 1997.
2. Cushnie, George C., "Pollution Prevention and Control Technology for the
Electroplating Industry," NCMS, 1994.
3. Concurrent Technologies Corporation, "Environmental Technology Verification
Program Metal Finishing Technologies (ETV-MF) Quality Management Plan."
December 9, 1998.
4. US EPA Office of Research and Development,. "Preparation Aids for the
Development of Category IV Quality Assurance Project Plans, " EPA/600/8-91/006,
February, 1991.
5. Concurrent Technologies Corporation, "Environmental Technology Verification Program
Metal Finishing Technologies (ETV-MF) Pollution Prevention Technologies Pilot Job
Training Analysis Plan," May 10, 1999.
11.0 DISTRIBUTION
Alva Daniels, EPA (3)
Mike Chan, USFilter
Jim Boyd, Gates Rubber Company
George Cushnie, CAI Resources, Inc.
Donn Brown, CTC (3)
Clinton Twilley, CTC
Marion Rideout, CTC
Kathy Fugiel, AMTest
28
-------
APPENDIX A
PAVCO CLEAN-R-120GR Material Safety Data Sheet
29
-------
APPENDIX B
USFilter Membralox® Silverback™ Model 900 Operation and Maintenance
Manual (selected sections)
30
-------
Appendix C
ETV-MF Operation Planning Checklist
31
-------
ETV-MF Operation Planning Checklist
The ETV-MF Project Manager prior to initiation of verification testing must complete this form.
If a "yes " is checkedfor any items below, an action must be specified to resolve the concern on
the Job Training Analysis Form.
Project Name: Expected Start Date:
ETV-MF Project Manager:
Will the operation or activity involve the following: Yes No Initials & Date
Completed
Equipment requiring specific, multiple steps for controlled shutdown? (e.g.
in case of emergency, does equipment require more than simply pressing a
"Stop" button to shut off power?) Special Procedures for emergency shut-
down must be documented in Test Plan.
Equipment requiring special fire prevention precautions? (e.g. Class D fire
extinguishers)
Modifications to or impairment of building fire alarms, smoke detectors,
sprinklers or other fire protection or suppression systems?
Equipment lockout/tagout or potential for dangerous energy release?
Lockout/tagout requirements must be documented in Test Plan.
Working in or near confined spaces (e.g., tanks, floor pits) or in cramped
quarters?
Personal protection from heat, cold, chemical splashes, abrasions, etc.? Use
Personal Protective Equipment Program specified in Test Plan.
Airborne dusts, mists, vapors and/or fumes? Air monitoring, respiratory
protection, and /or medical surveillance may be needed.
Noise levels greater than 80 decibels? Noise surveys are required.
Hearing protection and associated medical surveillance may be necessary.
X-rays or radiation sources? Notification to the state and exposure
monitoring may be necessary.
Welding, arc/torch cutting, or other operations that generate flames and/or
sparks outside of designated weld areas? Follow Hot Work Permit
Procedures identified in Test Plan.
The use of hazardous chemicals? Follow Hazard Communication
Program, MSDS Review for Products Containing Hazardous Chemicals.
Special training on handling hazardous chemicals and spill clean-up may
be needed. Spill containment or local ventilation may be necessary.
Working at a height of six feet or greater?
C-l
-------
ETV-MF OPERATION PLANNING CHECKLIST
The ETV-MF Project Manager prior to initiation of verification testing must complete this form.
If a "yes " is checkedfor any items below, an action must be specified to resolve the concern on
the Job Training Analysis Form.
Project Name:
ETV-MF Project Manager:
Will the operation or activity involve the following: Yes No Initials & Date
Completed
Processing or recycling of hazardous wastes? Special permitting may be
required.
Generation or handling of waste?
Work to be conducted before 7:00 a.m., after 6:00 p.m. and/or on
weekends? Two people must always be in the work area together.
Contractors working in CTC facilities? Follow Hazard Communication
Program.
Potential discharge of wastewater pollutants?
EHS aspects/impacts and legal and other requirements identified?
Contaminants exhausted either to the environment or into buildings?
Special permitting or air pollution control devices may be necessary.
Any other hazards not identified above? (e.g. lasers, robots, syringes)
Please indicate with an attached list.
The undersigned responsible party certifies that all applicable concerns have been indicated in the "yes" column,
necessary procedures will be developed, and applicable personnel will receive required training. As each concern is
addressed, the ETV-MF Project Manager will initial and date the "initials & date completed" column above.
ETV-MF Project Manager:
(Name) (Signature) (Date)
C-2
-------
APPENDIX D
Job Training Analysis Form
32
-------
Job Training Analysis Form
ETV-MF Project Name:
Basic Job Step
POTENTIAL EHS
ISSUES
Potential Quality
Issues
Training
ETV-MF Project Manager:
Name Signature
Date
D-l
-------
Appendix E
ETV-MF Project Training Attendance Form
33
-------
ETV-MF Project Training Attendance Form
ETV-MF Pilot Project:
Date
Training
Completed
Employee Name
Last First
Training Topic
Test
Score
(If applic.)
ETV-MF Project Manager:
E-l
-------
APPENDIX F
Oil Analysis
EPA Method 8015, Modified
(Northwest Total Petroleum Hydrocarbon -
Extended Diesel) (NWTPH-DX)
34
-------
NWTPH-DX
Diesel Range Organics In Soil And Water
Summary
The NWTPH-D Method adapts EPA SW-846 Methods 3540 and 8000 and covers the
quantitative analysis of semi-volatile petroleum products in soils. The method involves
extracting the sample with methylene chloride and injecting of a portion of the extract into a gas
chromatograph equipped with a flame ionization detector. This method specifies criteria for the
identification and quantitation of semi-volatile petroleum products. When the type of fuel is
unknown #2 diesel will be used to quantitate the sample. The reporting limit is 25 mg/kg for soil
and 0.10 mg/1 for water samples eluting from the jet fuels range to the diesel #2 range. For
petroleum products eluting after diesel #2 the reporting limits are 100 mg/kg for soil and 0.20
mg/1 for water (assuming 100% total solids for soil).
Equipment and Reagents
Gas Chromatograph
Flame Ionization detector
Column: J & W DB-5 30M x . 32mm with ,25um film thickness
Capillary column
Maxima Data System
Analytical Balance accurate to at least 0.00lg
Horn Sonicator
Volumetric Flasks, 10ml Ground Glass Stoppered
150ml beakers
Sodium Sulfate
Methylene Chloride
K-D Equipment (refer to K-D section)
Nitrogen evaporator
Sulfuric Acid, concentrated
Silica gel cartridges
Various Petroleum products for standards
Collection Requirements
All samples should be collected in I-Chem containers and preserved at 4 degrees Celsius until
extracted. The holding time from the date of collection to extraction, is 14 days for soils and
preserved water. For unpreserved water, the holding time is 7 days. Preservation is
accomplished by adjusting pH to about 2 using 1:1 HC1.
Standards
Fuel Stock Standard:
Choose the appropriate fuel for comparison to the sample fingerprint. Weight approximately
0.10 g into a 10ml vol. flask and dilute to volume with DCM. Label and record the exact
concentration.
F-l
-------
Calculate the concentration as follows:
Stock Cone, ug/ml = weight diesel (g^) x 1.000.000 ug/g
10ml
Calibration Standard:
Prepare calibration standards from the stock diesel standard at concentrations of 25, 50, 200, 200,
300 ug/ml by adding appropriate volumes to a 10ml vol. flask and diluting to volume with
methylene chloride. For fuels heavier than diesel #2, prepare standards at concentrations of 50,
100, 150, 300, and 400 ug/ml.
To calculate volume (ul) of stock standard to add to 10ml vol. flask use the equation below:
Vol. Diesel Stock, ul = Cal. Std. Cone ug/ml x lOOOug x 10
Diesel stock ug/ml
Dilute the flask to lOmls with DCM.
Stock Surrogate Standard:
Make up a surrogate of bromofluorobenzene and 2-fluorobiphenyl, which contains
approximately 8000 ug/ml by weighing about .080 g of the surrogate compounds into a 10ml
volumetric flask and filling to volume with mehtylene chloride.
Working Surrogate Spike (800 ug/mO:
Add the appropriate volume of the stock standard to a 10ml volumetric flask which has been
filled with 5mls of methylene chloride taking care not to add the surrogate standard solution into
the solvent without contacting the neck of the flask. Fill the flask to volume, stopper and mix.
Store at 4 degrees Celsius.
Volume Surrogate = 800 ug/ml x 10ml x lOOOul
Stock, ul Surrogate stock conc. ug/ml
GC/FID PARAMETERS FOR FUEL SCANS
Instrument Parameters:
Column:
J & W DB-5 30M x ,32mm with .25 film
thickness, capillary column
Injection Sample Volume:
Injector Temperature:
Ion Block Temperature:
Initial Temperature:
Initial Time:
Initial Rate:
Final Temperature:
Final Time:
Purge Valve on time:
Purge Valve off time:
2 ul
290 C
300 C
35 C
300 C
5 minutes
1.5 minutes
36 minutes
5 minutes
10 degrees/min.
F-2
-------
Purge Valve on time:
Purge Valve off time:
1.5 minutes
36 minutes
Hydrogen Flow:
Air Flow:
Make-up Gas Flow:
Carrier Gas:
25 - 30 ml/min.
300 - 400 ml/min.
30 ml/min.
Helium
Helium Carrier gas Head Pressure = 12 psi
Sample Extraction Soil
Accurately weigh approximately 20 grams of soil (note that if the sample is hydrated, more than
20 grams is needed) 20 grams of anhydrous sodium sulfate into a 150ml beaker and mix
completely with a spatula. The mixture should have a grainy texture. If it forms a clump, add
more sodium sulfate, grind to a grainy texture and note this in the extraction log. Add 100 ul of
Working Surrogate spike and 50 ml of methylene chloride; sonicate this for 3 min. utilizing the
horn sonicator. (Refer to Horn sonicator instructions if unfamiliar with the operation of the
instrument at the end of this SOP).
Allow the mixture to stand. Collect the extract in a 250 ml Kuderna-Danish (KD) Flask to which
is connected a ten ml. concentrator tube and a sodium sulfate drying apparatus.
Repeat the extraction twice more using 50ml of Methylene Chloride and add the extract to the
same KD flask. Attach a 3 ball Snyder column and concentrate the extract to a final volume of
10 ml. If the extract is highly colored or forms a precipitate, a dilution may be necessary to stay
within the calibration range. If samples need to be cleaned up, refer to "Clean-up" procedure at
the end of this method.
Store the samples at 4 degrees Celsius in a glass vial until ready for analysis.
Water Extraction Procedure
Pour 500 mis of the sample into a 2 liter sepratory funnel. Adjust the pH to 2 if needed. Add
200 ul of surrogate working std. Extract the sample with 50 mis of DCM. Pour the extract
through sodium sulfate into a K-D set up. Extract the sample twice more with 50 mis. DCM
adding the extract to the K-D set up. Concentrate the sample to 5 - 10 mis on a steam bath.
Remove the ampule and continue to concentrate on a N-Vap to below 2 mis. Adjust the final
volume to 2 mis in a vol. flask. Clean-up the sample if needed using the procedure at the end of
the SOP.
Determine the Total Solids Percent of soil sample.
GC Run to include the following:
1. Five point calibration curve
2. 10% duplicates
3. Surrogate std (100 ug/L working surrogate spike to 10 ml)
4. Mid std check every ten samples analyzed
5. End std check at the end of each run.
F-3
-------
Data Validation:
1. Continuing calibration checks and end checks must fall +/- 15% of the known value of the
std.
2. Surrogate recoveries must be between 50% - 150%.
3. Standard curve must have a minimum correlation of 0.99.
Sample Calculations
The retention time range windows for integration must be adjusted to incorporate the majority of
the components of the petroleum product of interest. If an exact match cannot be made, a
standard is chosen which closely represents the sample. In all cases, the selected retention time
window used for quantitation must, at a minimum, include any unresolved envelope of
compounds as well as all discrete components peaks with an area greater than or equal to 10% of
the largest peak. These must be integrated to the baseline as a group.
Be sure to subtract the area of the surrogates if the surrogate falls within the retention time
window.
Adjustments of retention time windows may be made if interferences are present, i.e, overlap of
oil into diesel area.
Sample Cone, mg/kg = Sample conc. ug/ml x v x DF
Sample weight x TS
V = Final Volume of extract
DF = Dilution Factor
TS = Decimal percent solids of sample
Horn Sonicator Settings
Sonicator Type: Ultrasonic, Inc. Model W-385 (475 watt) with No. 207 3/4" Tapped Disruptor
Horn
Settings: 3 minutes
Output Control Knob: 10
Mode: Pulse
Percent Duty Knob: 50%
Sample Clean-up Procedure:
When samples contain a significant amount of naturally occurring non-petroleum organics, e.g.
leaf litter, bark, etc. which may contribute to biogenic interference, the following clean-up
technique may be employed to assist in their reduction or elimination.
1. Transfer 2ml of the sample extract to a 4 ml vial.
2. Add .3 to .5 ml concentrated sulfuric acid to the vial and shake for 1 minute.
3. Allow the phases to separate and transfer the upper layer to another 4 ml vial.
4. Add about ,4g of silica to the vial and shake.
5. Repeat the procedure a second time. Transfer the cleaned extract to an auto sampler vial for
analysis.
6. If the clean-up affects the analyte of interest, clean the standards in the same way as the
samples.
F-4
-------
I
—2"
Jj
¦T
3 . 6 O -
3 , 40 -
^2
fi
*—
S
p5S>
II
si
If
> 3.ZO
-4
I
o
-4
X
3 - OO -U
2 . WO
-'T —
O , 50
1
1 . OO
—I—
1.30
x 10J
1
a.oo
mi nvtss
F-5
—j—
2 . SO
1
3 . OO
-------
------- |