-'United-States
Environmental Protection
Agency
Design Manual
Removal of Arsenic from
Drinking Water by
Adsorptive Media
As (III) As(V)
-------
EPA/600/R-03/019
March 2003
Design Manual:
Removal of Arsenic from
Drinking Water by Adsorptive Media
Prepared by
Frederick Rubel, Jr., P.E.
Rubel Engineering, Inc.
Tucson, Arizona 85712
Prepared for
Battelle
505 King Avenue
Columbus, Ohio 43281-2693
Under Contract with the U.S. EPA No. 68-C7-0008
Work Assignment Manager
Thomas J. Sorg
Water Supply and Water Resources Division
National Risk Management Research Laboratory
Cincinnati, Ohio 45268
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 4526
-------
Disclaimer
The information in this document has been funded by the United States Environ-
mental Protection Agency (U.S. EPA) under Work Assignment (WA) No. 4-32 of Con-
tract No. 68-C7-0008 to Battelle. It has been subjected to the Agency's peer and
administrative reviews and has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute an endorsement
or recommendation for use.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting
the Nation's land, air, and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions leading to
a compatible balance between human activities and the ability of natural systems to
support and nurture life. To meet this mandate, EPA's research program is providing
data and technical support for solving environmental problems today and building a
science knowledge base necessary to manage our ecological resources wisely,
understand how pollutants affect our health, and prevent or reduce environmental
risks in the future.
The National Risk Management Research Laboratory is the Agency's center for
investigation of technological and management approaches for preventing and
reducing risks from pollution that threatens human health and the environment. The
focus of the Laboratory's research program is on methods and their cost-
effectiveness for prevention and control of pollution to air, land, water, and
subsurface resources; protection of water quality in public water systems;
remediation of contaminated sites, sediments and ground water; prevention and
control of indoor air pollution; and restoration of ecosystems. NRMRL collaborates
with both public and private sector partners to foster technologies that reduce the
cost of compliance and to anticipate emerging problems. NRMRL's research
provides solutions to environmental problems by: developing and promoting
technologies that protect and improve the environment; advancing scientific and
engineering information to support regulatory and policy decisions; and providing the
technical support and information transfer to ensure implementation of environmental
regulations and strategies at the national, state, and community levels.
This publication has been produced as part of the Laboratory's strategic long-term
research plan. It is published and made available by EPA's Office of Research and
Development to assist the user community and to link researchers with their clients.
Hugh W. McKinnon, Director
National Risk Management Research Laboratory
-------
Abstract
This design manual is an in-depth presentation of the steps required to design and
operate a water treatment plant for removal of excess arsenic from drinking water
using the adsorptive media process. This treatment process is very reliable, simple
and cost-effective. Several adsorptive media products are available in the market-
place that have successfully demonstrated their capability to remove arsenic from
drinking water to levels well below the revised MCL, 0.010 mg/L. Other new products
continue to be developed. The adsorptive media products are preferential for the
removal of arsenic over other competing ions. Therefore, unless a water system
requires treatment capability for removal of other suspended or dissolved contami-
nants, the adsorptive media treatment method merits evaluation.
The adsorptive media process is implemented with operational options which vary
with the product selected. For water systems that are primarily concerned with finan-
cial feasibility, capital and operating costs, each operational option along with each
available adsorptive media product should be evaluated. This design manual pro-
vides the methods for competently performing each evaluation. The arsenic removal
capacity of some adsorptive media products, such as activated alumina, are very
sensitive to the pH of the water passing thru treatment. Others, such as iron-based
products, are not. Treatment processes incorporating pH adjustment capability
require careful handling and storage of corrosive chemicals (acid and caustic). Some
adsorptive media products, such as activated alumina, are capable of being chem-
ically regenerated for repetition of treatment cycles using the same corrosive chemi-
cals as those used for pH adjustment in the treatment process. Regeneration is not
recommended for other adsorptive media products. Whether or not pH of water being
treated is adjusted, the adsorptive media can be replaced in place of regeneration
upon exhaustion of arsenic capacity. This design manual presents the information
necessary to design and operate treatment systems for any combination of opera-
tional options and for any adsorptive media. It also discusses the capital and operat-
ing costs including the many variables which can raise or lower costs for identical
treatment systems.
IV
-------
Contents
Abstract iv
Figures vi i i
Tables ix
Acronyms and Abbreviations x
1.0 Introduction 1
1.1 Purpose and Scope 1
1.2 Background 1
1.3 Arsenic in Water Supplies 2
1.4 Arsenic Speciation 2
1.5 Removal of Arsenic 3
2.0 Arsenic Removal by Adsorptive Media Treatment Methods 5
2.1 Introduction 5
2.2 Granular Adsorptive Media 5
2.2.1 pH Adjustment System 5
2.2.2 Non-pH Adjustment System 6
2.3 Treatment With or Without pH Adjustment 6
2.4 Treatment Media Regeneration vs. Treatment Media Disposal 7
2.5 Manual vs. Automatic Operation 7
3.0 Design of Central Treatment System 9
3.1 Assemble Design Input Data and Information 9
3.2 Conceptual Design 11
3.2.1 Manual Operation 12
3.2.2 Automatic Operation 12
3.3 Preliminary Design 18
3.3.1 Treatment Equipment Preliminary Design 18
3.3.1.1 Treatment Bed and Vessel Design 22
3.3.1.2 Pipe Design 22
3.3.1.3 Instrumentation Design 23
3.3.1.4 Acid Storage and Feed Subsystem 23
3.3.1.5 Caustic Soda Storage and Feed Subsystem 24
3.3.2 Preliminary Treatment Equipment Arrangement 24
3.3.3 Preliminary Cost Estimate 24
3.3.4 Preliminary Design Revisions 25
3.4 Final Design 25
3.4.1 Treatment Equipment Final Design 26
3.4.1.1 Treatment Bed and Vessel Design 26
3.4.1.2 Pipe Design 27
3.4.1.3 Instrument Design 28
3.4.1.4 Acid Storage and Feed Subsystem 28
3.4.1.5 Caustic Soda Storage and Feed System 29
3.4.1.6 Regeneration Wastewater Surge Tank 29
3.4.2 Final Drawings 29
3.4.3 Final Capital Cost Estimate 30
3.4.4 Final Design Revisions 31
4.0 Central Treatment System Capital Cost 33
4.1 Introduction 33
-------
4.2 Discussion of Cost Variables 33
4.2.1 Water Chemistry 35
4.2.2 Climate 35
4.2.3 Seismic Zone 36
4.2.4 Soil Conditions 36
4.2.5 100-Year Flood Plain 36
4.2.6 Existing and Planned (Future) Water System Parameters 36
4.2.6.1 Number and Location of Wells 37
4.2.6.2 Potable Water Storage Facilities 37
4.2.6.3 Distribution and Consumption 37
4.2.7 pH Adjustment of Process Water Included vs. Not Included 37
4.2.8 Regeneration or Replacement of Spent Adsorptive Media 38
4.2.9 Backwash and Regeneration Disposal Concept 38
4.2.10 Chemical Supply Logistics 38
4.2.11 Manual Versus Automatic Operation 38
4.2.12 Financial Considerations 39
4.3 Relative Capital Cost of Arsenic Removal Central Water Treatment
Plants Based on Flowrate 39
5.0 Treatment Plant Operation 43
5.1 Introduction 43
5.2 Adsorptive Media Initial Startup 44
5.3 Treatment Process with Spent Treatment Media Regeneration 46
5.3.1 Treatment Mode 46
5.3.2 Backwash Mode 49
5.3.3 Regeneration Mode 49
5.3.4 Neutralization Mode 50
5.4 Treatment Process with Spent Treatment Media Replacement 51
5.4.1 Treatment Mode 51
5.4.2 Media Replacement Mode 51
5.5 Operator Requirements 51
5.6 Laboratory Requirements 51
5.7 Operating Records 52
5.7.1 Plant Log 52
5.7.2 Operation Log 52
5.7.3 Water Analysis Reports 52
5.7.4 Plant Operating Cost Records 52
5.7.5 Correspondence Files 52
5.7.6 Regulatory Agency Reports 52
5.7.7 Miscellaneous Forms 52
5.8 Treatment Plant Maintenance 52
5.9 Equipment Maintenance 52
5.10 Treatment Media Maintenance 54
5.11 Treatment Chemicals Supply 54
5.12 Housekeeping 54
6.0 Central Treatment Plant Operating Cost 55
6.1 Introduction 55
6.2 Discussion of Operating Costs 55
6.2.1 Treatment Chemical Costs 56
6.2.1.1 Acid Cost 56
6.2.1.2 Caustic Cost 58
6.2.2 Operating Labor Costs 59
6.2.3 Utility Cost 60
6.2.4 Replacement Treatment Media Cost 61
6.2.5 Replacement Parts and Miscellaneous Material Costs 62
VI
-------
6.3 Operating Cost Summary 62
7.0 References 63
Appendix A: Summary of Subsystems Including Components 65
Appendix B: Treatment System Design Example 69
Appendix C: Discussion of Acid Consumption Requirements for pH
Adjustment of Raw Water 73
Appendix D: Tabulations of Estimated Capital Cost Breakdowns for Arsenic
Removal Water Treatment Plants by Means of the Activated
Alumina Process at Typical and Ideal Locations 77
Appendix E: Alternative Methods for Removing Media from Very Small
System Tanks 83
Appendix F: English to Metric Conversion Table 85
VII
-------
Figures
Figure 3-1. Arsenic Removal Water Treatment Plant Water Analysis Report 10
Figure 3-2. Flow Diagram for Dual Vessel Series Downflow Treatment
System Without pH Adjustment, With Replacement of Spent
Media 13
Figure 3-3. Flow Diagram for Dual Vessel Series Downflow Treatment
System With pH Adjustment, With Replacement of Spent Media 14
Figure 3-4. Flow Diagram for Dual Vessel Series Downflow Treatment
System With pH Adjustment, With Regeneration of Spent Media 15
Figure 3-5. Treatment Bed and Vessel Design Calculations 16
Figure 3-6. Treatment System Plan for Adsorptive Media Without Process
Water pH Adjustment and With Spent Media Replacement 19
Figure 3-7. Treatment System Plan for Adsorptive Media With Process Water
pH Adjustment and Spent Media Replacement 20
Figure 3-8. Treatment System Plan for Adsorptive Media With Process Water
pH Adjustment and Spent Media Regeneration 21
Figure 3-9. Treatment Vessels Piping Isometric Adsorptive Media With or
Without Process Water pH Adjustment and With Spent Media
Replacement 30
Figure 3-10. Treatment Vessels Piping Isometric Adsorptive Media With
Process Water pH Adjustment and Spent Media Regeneration 30
Figure 4-1. Capital Cost vs. Flowrate at Typical Locations for Arsenic
Removal Water Treatment Plants by Means of the Activated
Alumina Process 34
Figure 4-2. Capital Cost vs. Flowrate at Ideal Locations for Arsenic Removal
Water Treatment Plants by Means of the Activated Alumina
Process 35
Figure 4-3. Code Pressure Vessel Fabricator Quotation for Adsorptive Media
Treatment Vessels 40
Figure 5-1. Valve Number Diagram 44
Figure 5-2. Basic Operating Mode Flow Schematics 47
Figure 5-3. Typical Breakthrough Curve for Arsenic 48
Figure 5-4. Arsenic Removal Water Treatment Plant Operation Log 53
Figure 5-5. 5,000-gal Chemical Storage Tank- Liquid Volume 54
VIII
-------
Tables
Table 1-1. Adsorptive Media List in NSF/ANSI STD 61 (May 2002) 2
Table 3-1. Preliminary Capital Cost Estimate Examples for Four Types of
Adsorptive Media Arsenic Removal Water Treatment Plants 22
Table 3-2. Final Capital Cost Estimate Examples for Typical Location for
Four Types of Adsorptive Media Arsenic Removal Water
Treatment Plants 25
Table 4-1. Final Capital Cost Estimate Example for Ideal Location for Four
Types of Adsorptive Media Arsenic Removal Water Treatment
Plants 36
Table 4-2. Process Pipe, Fittings, Valves, and Static Mixers - Itemized Cost
Estimate 41
Table 4-3. Chemical Feed Pumps, and Static Mixers - Itemized Cost
Estimate 42
Table 5-1. Valve Operation Chart for Treatment Vessels in Spent Adsorptive
Media Regeneration Operational Modes 45
Table 5-2. Calculated Activated Alumina (-28, +48 Mesh) Downflow
Pressure Drop Data 46
Table 5-3. Typical Process Conditions for Regeneration of an Activated
Alumina Treatment System 50
Table 6-1. Price for Typical -28, +48 Mesh Activated Alumina 61
Table 6-2. Operating Cost Tabulation for an Activated Alumina Plant 62
IX
-------
Acronyms and Abbreviations
ANSI American National Standards Institute
APHA American Public Health Association
ASME American Society of Mechanical Engineers
AWWA American Water Works Association
CPVC chlorinated polyvinyl chloride
EBCT empty bed contact time
EPDM ethylene propylene diene monomer
ETV Environmental Technology Verification
FRP fiberglass reinforced polyester
GFAA graphite furnace atomic adsorption
GHAA gaseous hydroxide atomic adsorption
gpd gallons per day
gpm gallons per minute
ICP-MS inductively coupled plasma-mass spectrometry
MCL maximum contaminant level
N/A not applicable
NPT National Pipe Thread
NSF NSF International
NTNC nontransient, noncommunity
OSHA Occupational Safety and Health Administration
PLC programmable logic controller
psig pounds per square inch gage
PVC polyvinyl chloride
SDWA Safe Drinking Water Act (of 1974)
STP stabilized temperature platform
TCLP Toxicity Characteristic Leaching Procedure
TDS total dissolved solids
U.S. EPA United States Environmental Protection Agency
WEF Water Environment Federation
-------
Acknowledgments
This manual was written by Frederick Rubel Jr., Rubel Engineering, Inc. with input
from Thomas J. Sorg, U.S. EPA, Lili Wang, Battelle Memorial Institute, and Bernie
Lucey, State of New Hampshire. The manual was reviewed by the following people
and their suggestions and comments were of valuable assistance in preparation of
the final document:
Dr. Abraham Chen, Battelle Memorial Institute, Columbus, OH
Dr. Dennis Clifford, University of Houston, Houston, TX
Mr. Jeff Kempic, U.S. EPA, Washington, DC
Mr. Rajiv Khera, U.S. EPA, Washington, DC
Dr. Jerry Lowry, Lowry Environmental Engineering, Blue, ME
Mr. Bernie Lucey, State of New Hampshire, Concord, NH
Mr. Michael McMullin, ADI, Fredericton, NB, Canada
Mr. Edward Robakowski, Kinetico, Inc. Newberry, OH
Mr. Thomas Sorg, U.S. EPA, Cincinnati, OH
Ms. Lili Wang, Battelle Memorial Institute, Columbus, OH
XII
-------
1.0 Introduction
1.1 Purpose and Scope
This manual has been prepared to present up-to-date
information on the design of central treatment plants for
the removal of arsenic from water supplies using the
adsorptive media process. Although the information pro-
vided in this manual is presented to serve the water treat-
ment industry for small central treatment plants ranging in
capacity from 30,000 to 1,000,000 gpd, the treatment
information, for the most part, can be adapted to both
larger and smaller systems. For the very small systems
having capacities of less than 30,000 gpd (20 gpm),
some of the equipment may be different and less expen-
sive (for example, fiberglass reinforced polyester [FRP]
tanks and automatic control valves likely would be used).
The detailed design information presented in this manual
applies to granular activated alumina and other granular
adsorptive media technology for selective removal of
arsenic from water supplies.
When arsenic is present above its maximum contami-
nant level (MCL) in a water supply in combination with
quantities of other organic and/or inorganic contami-
nants, the adsorptive media process may not be the opti-
mal method of arsenic removal. Those water supplies
should be evaluated on a case-by-case basis for selec-
tion of the appropriate treatment method, or combination
of methods.
1.2 Background
The Safe Drinking Water Act (SDWA) of 1974 mandated
that the United States Environmental Protection Agency
(U.S. EPA) identify and regulate drinking water contami-
nants that may have an adverse human health effect and
that are known or anticipated to occur in public water
supply systems. In 1975, under the SDWA, U.S. EPA
established a MCL for arsenic at 0.05 mg/L. During the
1980s and early 1990s, U.S. EPA considered changes
to the MCL, but did not make any. In 1996, Congress
amended the SDWA and these amendments required
that the U.S. EPA develop an arsenic research strategy,
publish a proposal to revise the arsenic MCL by January
2000, and publish a final rule by January 2001.
On January 22, 2001, U.S. EPA published a final
Arsenic Rule in the Federal Register that revised the
MCL for arsenic at 0.01 mg/L (10 ug/L). Two months
later, in March 2001, the effective date of the rule was
extended to provide time for the National Academy of
Science to review new studies on the health effects of
arsenic and for the National Drinking Water Advisory
Council to review the economic issues associated with
the standard. After considering the reports by the two
review groups, the U.S. EPA finalized the arsenic MCL
at 0.01 mg/L (10 ug/L) in January 2002. The final rule
requires all community and nontransient, noncommunity
(NTNC) water systems to achieve compliance with the
rule by February 2006. Adsorptive media processes are
capable of achieving that level.
Granular activated alumina was the first adsorptive
medium to be successfully applied for the removal of
arsenic from water supplies. With pH adjustment to 5.5,
the activated alumina process preferentially removes
arsenic in place of competing ions, removes arsenic
below the MCL, and provides a maximum removal
capacity for arsenic. It also has been the author's experi-
ence that both As(lll) and As(V) can be removed from
raw water with activated alumina when the pH is adjusted
down to 5.5.
The optimum granular adsorptive media mesh size for
activated alumina is -28, +48. Larger mesh sizes can be
used, but their arsenic capacities are lower. Finer mesh
material has not been used for this application other than
in laboratory bench-scale work. Mesh sizes for other
products are listed in Table 1-1.
Recently, other adsorptive media have been developed
and marketed for arsenic removal. These new materials
are either iron or aluminum (modified activated alumina)-
based. A listing of the activated alumina and the more
recently developed media that have obtained NSF Inter-
national (NSF) listing under NSF/ANSI STD 61 are
-------
Table 1-1. Adsorptive Media Listed in NSF/ANSI STD 61 (November 2002)
Base
Material
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Aluminum
Iron
Iron
Iron
Iron
Zeolite
Zirconium
Company
Alcan
Alcan
Alcan
Alcan
Alcoa
Alcoa
Apyron
Engelhard
Engelhard
Engelhard
ADI International
SMI
U.S. Filter/General Filter Products
Bayer AG
Water Remediation Technology
Magnesium Elekton
Product Name
AA-400G
AA-400G
AAFS-50
AAFS-50
DD-2
CRN
Aqua-Bind Arsenic
ATS Sorbent
ATC Sorbent
ARM
G2
SMI III
GFH
Bayoxide E 33
Z-33
Isolux
Material
Activated Alumina
Activated Alumina
Modified Activated Alumina
Modified Activated Alumina
Activated Alumina
Activated Alumina
Activated Alumina
Activated Alumina
Activated Alumina
Activated Alumina
Iron Modification
Iron/Sulfur
Iron Hydroxide
Iron Oxide
Modified Zeolite
Zirconium Hydroxide
Mesh Size
or as Noted
14 x28
28x48
14x28
28 x48
28x48
28x48
NA
NA
NA
•80
0.08-1 .25 mm
NA
0.32-2 mm
0.5-2 mm
8x40
NA
Regeneration
of Media
Yes
Yes
Yes
Yes
Yes
Yes
NA
Yes
Yes
Yes
Yes
NA
No
No
No
NA
Note: Mention of trade names or commercial products does not constitute endorsement or recommendation by U.S. EPA.
NA = not available.
shown in Table 1-1. Other media currently are being
researched by various companies and new products
likely will appear on the market in the future.
The arsenic removal capacity for some newly developed
adsorptive media also is enhanced by pH adjustment.
Furthermore, some newly developed adsorptive media
are able to be regenerated by means of chemical pH
adjustment upon exhaustion of arsenic removal capacity.
This manual is intended to apply to all presently avail-
able and future adsorptive media for removal of arsenic
from water supplies. This manual provides a design
methodology for the use of adsorptive media for arsenic
removal with or without pH adjustment, and with spent
adsorptive media regeneration or spent adsorptive media
replacement.
1.3 Arsenic in Water Supplies
Arsenic occurs in combination with other ions as arsenic
compounds. Unless contaminated by arsenic-bearing
wastes, the arsenic concentrations in surface water sup-
plies are normally less than the MCL. Ground water sup-
plies have higher arsenic concentrations which may
exceed the MCL due to the exposure of the water to
arsenic-bearing materials. Because the revision of the
MCL, a large number of systems which had previously
been within compliance will require treatment for the
removal of arsenic.
1.4 Arsenic Speciation
Arsenic is a common, naturally occurring drinking water
contaminant that originates from arsenic-containing rocks
and soil and is transported to natural waters through
erosion and dissolution. Arsenic occurs in natural waters
in both organic and inorganic forms. However, inorganic
arsenic is predominant in natural waters and is the most
likely form of arsenic to exist at concentrations that cause
regulatory concern.
The valence and species of inorganic arsenic are
dependent on the oxidation-reduction conditions and the
pH of the water. As a general rule of thumb, arsenite, the
reduced, trivalent form [As(lll)], normally is found in
ground water (assuming anaerobic conditions); and
arsenate, the oxidized pentavalent form [As(V)], is found
in surface water (assuming aerobic conditions). This
rule, however, does not always hold true for ground
water. Some ground waters have been found to have
only As(lll), others with only As(V), and still others with
the combination of both As(lll) and As(V). Arsenate
exists in four forms in aqueous solution, depending on
pH: H3As04, H/.SCV, HAsO42-, and AsO43-. Similarly,
arsenite exits in five forms: H4AsO3+, H3AsO3, HjAsO^,
HAs032- and AsO3^.
Until recently, studies on the preservation of the arsenic
species concluded that no effective methods existed for
the preserving of As(lll) and As(V) in water samples.
Because of the lack of a good preservation method, field
separation methods developed by Ficklin (1982), Clifford
et al., (1983) and Edwards et al. (1998) have been used
that employ an anion exchange column as the separa-
tion procedure. All the methods have been found to be
effective and their use is recommended to determine the
oxidation state of the arsenic in the source water to be
treated.
-------
1.5 Removal of Arsenic
In water supplies where the arsenic level exceeds the
MCL, steps should be taken to reduce that level to below
the MCL. This design manual focuses on the removal of
excess arsenic by using activated alumina and other
adsorptive media methods. However, other treatment
methods exist, such as ion exchange, membrane sepa-
ration, and chemical coagulation/filtration. Also, other
options, including alternate sources of supply, may offer
lower cost solutions. The first option is to locate an
existing water supply within the service area with known
quality that complies with the arsenic MCL in addition to
all other MCLs (both organic and inorganic). If another
source complies with the arsenic MCL but exceeds
another MCL (or MCLs), it may still be feasible to blend
the two sources and achieve a water quality that com-
plies with all MCLs. Other features associated with this
option may present liabilities, including, but not limited
to, high temperature, or undesirable quantities of non-
toxic contaminants such as turbidity, color, odor, hard-
ness, iron manganese, chloride, sulfate, and/or sodium.
A second option is to pump good quality water to the
service area from another service area. Similar to the
alternate source within the service area, this imported
source can be blended. However, the costs of installing
the delivery system and delivering the water become
increasingly unfavorable as the distance increases, the
rise in elevation increases, and/or the existence of
physical barriers occurs. The reliability, the cost and the
assurance that the consumers will only use that source
are factors to be considered. Another option (which
includes an element of risk) is to drill a new well (or
wells) within the service area. This approach should be
attempted only when there is sound reason to believe
that sufficient quantity of acceptable quality water can be
located. The cost (both capital and operating) of a new
well should not exceed the cost of treating the existing
source. Other options such as "point-of-use" treatment
systems are viable alternatives. However, the treatment
reliability of such units cannot be assured unless there
are stringent controls governing their operation and
maintenance. Also, the problem of assuming that all
users consume only water that has been treated where
untreated water also is available should be addressed.
-------
-------
2.0 Arsenic Removal by Adsorptive Media Treatment Methods
2.1 Introduction
This chapter provides an overview of the design consid-
erations that are applicable to adsorptive media treat-
ment systems; applicable details are covered in later
chapters. The design choices are as follows:
1. Selection of adsorptive media
2. Treatment with or without pH adjustment
3. Treatment media regeneration vs. treatment media
disposal
4. Manual vs. automatic operation (or semiautomatic
operation).
2.2 Granular Adsorptive Media
This design manual focuses on the implementation of
the granular adsorptive media method for the selective
removal of arsenic from water supplies with or without
pH adjustment and with or without spent media regener-
ation. The treatment method example presented employs
activated alumina media, which utilizes a single treatment
train and consists of two downflow pressure vessels in
series. This method is applicable to the use of any other
adsorptive media, and, therefore, one adsorptive media
can be replaced with another without replacing or mak-
ing major modifications to an installed treatment system.
Activated alumina has a long history of use as an
adsorptive treatment technology for arsenic removal.
The media is a byproduct of aluminum production. It is
primarily an aluminum oxide that has been activated by
exposure to high temperature and caustic soda. The
material is extremely porous and has a high average
surface area per unit weight (350 nf/g). The capacity for
arsenic removal by activated alumina is pH-dependent,
with the maximum removal capacity achieved at pH 5.5.
Adjusting the pH of the source water, therefore, provides
removal capacity advantages. As the pH deviates from
the 5.0-6.0 range, the adsorption capacity for arsenic de-
creases at an increasing rate. Process demonstrations
have shown that arsenic removal capacity has been
reduced by more than 15% at pH 6.0 compared to that
ofpH5.5(Rubel, 1984).
Fluoride, selenium, and other inorganic ions and organic
molecules also are removed by the same pH adjustment
activated alumina process. The process, however, is
preferential for arsenic at the optimum pH level of 5.5.
Other ions that compete with arsenic for the same
adsorptive sites at other pH levels are not adsorbed in
the pH range of 5.0-6.0. Included are silica and hardness
ions that are adsorbed in the pH range of 7-10.
Activated alumina either can be regenerated or can be
replaced with new media when the selected break-
through point is reached. At the optimum pH for arsenic
removal, fluoride, selenium, some organic molecules,
and some trace heavy metal ions are adsorbed; how-
ever, these are also completely regenerated along with
arsenic. Because these ions compete for the same
adsorptive sites with arsenic, their presence might deplete
the alumina capacity for arsenic. When excess fluoride
and arsenic are present in the water supply, a special
treatment technique is required (Rubel and Williams,
1980).
Newly developed adsorptive media for arsenic removal
consist primarily of iron-based materials or iron-modified
activated alumina products (see Table 1-1). Some of
these materials are not capable of regeneration and,
thus, are used solely on a replacement basis (throw-
away). Some of these media, mainly the iron-based
products, have demonstrated arsenic removal capacities
that exceed that of activated alumina particularly at pHs
above the optimum pH 5.5 level for alumina treatment.
The adsorptive capacity of these new materials also is
affected by pH; however, their pH sensitivity does not
resemble that of activated alumina. The benefit of pH
adjustment may come more from the elimination of com-
petition for adsorptive site by ions such as silica and
phosphate. Consequently, these materials can be
employed economically on a spent media replacement
basis without the incorporation of pH adjustment chemi-
cals and equipment. As new adsorptive media products
and technology evolve, more efficient and economical
arsenic removal treatment systems will become available.
2.2.1 pH Adjustment System
The adsorptive capacity of many adsorptive media, par-
ticularly activated alumina, is pH sensitive; removal
capacity increases with decreasing pH. Employing pH
adjustment, therefore, generally provides cost advan-
tages regardless of whether the media is regenerated or
replaced. Because the pH adjustment chemicals are
usually the same chemicals that are use for regenera-
-------
tion, it is generally advantageous to couple regeneration
with pH adjustment systems when the media can be
regenerated.
The advantages of using an adsorptive system with pH
adjustment and regeneration or replacement of spent
media are as follows:
1. System is low-cost and simple to operate.
2. System requires minimal operator attention (part
time) during treatment runs.
3. System can employ manual operation and is
adaptable to automatic operation.
4. Activated alumina media system has longer
treatment runs (greatest removal capacity). Other
media may have the same advantage.
5. Activated alumina system removes As(lll) and As(V)
at pH 5.5 (author's experience).
The disadvantages of using the pH adjustment method
are as follows:
1. System requires chemical feed equipment and the
storage and handling of corrosive chemicals (acid
and caustic) for pH adjustment of raw water and re-
adjustment of treated water.
2. pH adjustment chemicals increase inorganic ions
and total dissolved solids (TDS) in the treated water.
Secondary MCLs must be considered.
3. System with regeneration of spent media requires
disposal of wastewater.
2.2.2 Non-pH Adjustment System
Some adsorptive media do not provide significant gains
in removal capacity by lowering pH as does activated
alumina. These materials, as well as activated alumina,
are used without pH adjustment with good results par-
ticularly by very small systems that do not want to
handle pH adjustment chemicals. In the case where pH
adjustment is not used, regeneration is not advanta-
geous or practical. Consequently, a non-pH adjustment
system usually is coupled with replacement of spent
media only.
The advantages of utilizing an adsorptive system without
pH adjustment or regeneration of spent media are as
follows:
1. System is inexpensive to install and, depending on
the arsenic concentration and water quality (com-
petitive ions, etc.), operational cost may be low.
2. System does not require chemical feed and storage
equipment. The handling of corrosive chemical is
not required.
3. System requires minimal operator attention (part
time) during treatment runs.
4. System can employ manual operation, and
automatic operation may not be necessary.
5. If arsenic breakthrough occurs, the arsenic
concentration in the treated water will not exceed
that of the raw water.
6. Disposal of spent arsenic-bearing activated alumina
and iron based media products can be accomplished
as a nonhazardous waste (i.e., media passes Tox-
icity Characteristic Leaching Procedure [TCLP] test).
The disadvantages of utilizing the non-pH adjustment
method without regeneration of spent media are as
follows:
1. System has lower adsorptive removal capacity,
particularly an activated alumina system, resulting in
much shorter treatment runs.
2. Other ions (e.g., silica, phosphate, etc) generally
compete for adsorption sites with arsenic. The
extent of competition depends on the pH of the
source water.
3. System requires more frequent media replacement.
Expensive materials could result in costly operation.
2.3 Treatment With or Without
pH Adjustment
Prior to start of design, the best arsenic removal treat-
ment method for a given application should be selected.
Not all adsorptive media may be as pH-sensitive as acti-
vated alumina. The manufacturers of these materials
advise that, even though pH adjustment does enhance
arsenic removal performance, it is not required to achieve
cost-effective results. The selection of adsorptive media
will rely on either the manufacturer's media performance
claims, or the development of independent technical per-
formance data through field pilot testing or other means.
Though costly, it is highly recommended that technical
data be collected for a given application.
The decision to adjust treatment pH is determined in the
conceptual design phase of the project. If the decision is
-------
not to incorporate pH adjustment, then the capital cost
for the treatment system is reduced and regeneration of
adsorptive media is eliminated. If the decision is to incor-
porate pH adjustment for the treatment process, then the
capability to regenerate the adsorptive media in place of
media replacement is available (but optional).
2.4 Treatment Media Regeneration vs.
Treatment Media Disposal
The decision to regenerate or replace spent treatment
media for each system should be made based upon
economic, technical, and/or aesthetic operating require-
ments. A major factor to be evaluated is the disposal of
the regeneration wastewater.
Activated alumina and some other adsorptive media can
be regenerated chemically for reuse rather than being
disposed of after arsenic removal capacity has been
exhausted. For regenerable treatment media, an eco-
nomic/technical evaluation should be performed to deter-
mine whether to provide regeneration capability for the
treatment system. If the treatment plant is capable of
adjusting the raw and treated water pH, then the require-
ment to handle, store, and feed corrosive chemicals
(acid and caustic) is already included. However, for a
media replacement system that does not require major
chemical storage equipment, the procurement of more
costly packaging of pH adjustment chemicals will be
required.
Regeneration involves removing the arsenic from the
treatment media, precipitating the dissolved arsenic in
the regeneration wastewater, dewatering the arsenic-
bearing precipitated solids, and finally disposing of waste
solids and liquids in a method acceptable to the presid-
ing regulatory agency.
Due to increased capacity for arsenic removal resulting
from pH adjustment, the implementation of a pH adjust-
ment treatment system may be justified with or without
regeneration of the spent adsorptive media.
Treatment media regeneration is more likely to be eco-
nomically justified for systems with high flowrates and
high raw water arsenic concentrations due to the result-
ing rapid consumption of arsenic capacity. The higher
the arsenic concentration in the raw water, the higher the
probability that regeneration of treatment media will be
economically desirable. Each evaluation should include
the variables that affect the cost of spent media regener-
ation vs. replacement.
Some adsorptive media are not capable of regeneration
and, upon exhaustion of arsenic capacity, must be
removed for disposal. For those materials, regeneration
is not a consideration. For systems that are not large
enough to economically justify the processing of the
regeneration wastewater, regeneration generally is not a
consideration. However, very small systems with capa-
bility to economically dispose of regeneration waste-
water should evaluate this option.
Adsorptive media with very high arsenic removal capaci-
ties can economically justify media replacement rather
than regeneration, even though the media can be regen-
erated.
Chemical regeneration may not be economical without
implementation of the same chemicals for treatment pH
adjustment. Therefore, the regeneration option should
be discarded if water utilities prefer not to handle corro-
sive chemicals, or advocate that addition of treatment
chemicals might degrade the quality of the potable
water, or for other economical, technical, or aesthetic
concerns.
2.5 Manual vs. Automatic Operation
The water utility owner should be informed of the advan-
tages and disadvantages of operational options prior to
finalizing the decision on mode of operation. The system
can be operated manually, automatically, or semiautomat-
ically. Automatic operation reduces the operator effort,
but increases the cost of instrumentation and control
equipment as well as the skill level required of the oper-
ator who must be able to maintain more sophisticated
equipment.
Treatment systems utilizing adsorptive media are suitable
for manual operation. That operational mode requires the
treatment plant operator to accomplish the following:
1. Start/stop operation. Adjust flowrate.
2. Start/stop and adjust rate of chemical feed to control
pH. Monitor pH (for systems with treatment process
pH adjustment only).
3. Monitor and adjust system operating pressure.
4. Start/stop/control each backwash and regeneration
step (for systems with spent media regeneration
only).
5. Monitor and adjust water levels in reservoirs and
other containment facilities.
6. Monitor arsenic concentrations for raw water, treated
water, and intermediate sample points.
-------
A fully automatic instrumentation and control system
includes a programmable logic controller (PLC), an oper-
ator interface (screen with graphics), software, automatic
instrumentation (sensors, transmitters, controllers, alarms,
electrical conductors, pneumatic tubing, etc.), and auto-
matically controlled equipment (valves, pumps, chemical
feed pumps, air compressor, etc.). The instruments can
monitor pH, flow, level, pressure, and temperature. Arse-
nic concentration analyses require manual laboratory
procedures.
Semiautomatic operation entails automating any part of
the instrumentation and control functions, and the remain-
der is accomplished manually. Not included are the PLC,
operator interface, and required software. This opera-
tional mode reflects choices made by the owner with the
advice of the designer. The choices require analysis of
risk and treatment process efficiency vs. investment in
equipment and labor. This design manual presents infor-
mation regarding instrumentation and control functions,
all of which can be accomplished automatically or manu-
ally. The only exception is the laboratory analysis require-
ment for determination of arsenic concentration in raw
water, treated water, wastewater, and at intermediate
sample points.
Automatic operation is only practical for systems employ-
ing treatment process pH adjustment and spent media
regeneration. Semiautomatic operation is applicable to
systems that employ treatment process pH adjustment
with either spent media regeneration or replacement. For
systems without treatment process pH adjustment, auto-
matic operation is not practical. For those systems with-
out treatment process pH adjustment, semiautomatic
features for monitoring flow, pressure, and storage liquid
levels may be desirable.
-------
3.0 Design of Central Treatment System
The design of a central treatment system for the selec-
tive removal of arsenic from drinking water supplies is
a straightforward process. For simplicity, unless differ-
entiation of media is required, the term "adsorptive
media" represents all adsorptive media. Arsenic
removal treatment can be applied to existing water
systems that have high arsenic, and to new water
systems with high arsenic that must be reduced. The
design philosophy presented in this manual provides
information that can be applied to any arsenic removal
adsorptive media that is capable of removing As(lll)
and As(V). If an adsorptive medium is not capable of
removing As(lll), preoxidation of As(lll) to As(V) will be
required.
As(lll) can be easily convert to As(V) by several com-
monly used chemical oxidants. A laboratory study on
six chemical oxidants has recently been completed by
Ghurye and Clifford (2001). The results of this study
showed that chlorine, potassium permanganate, and
ozone were very effective oxidants, whereas chlorine
dioxide and monochloramine were not. The actual
amounts necessary to oxidize As(lll) must take into
account other oxidant demand substances in the
source water such as iron, manganese, and sulfide.
The study also showed that a solid oxidizing media
used for iron and manganese removal has the ability
to oxidize As(lll). Air oxidation that is effective for
oxidizing iron has been found to be ineffective for
As(lll) oxidation (Lowry and Lowry, 2002).
The information included presents flexibility to adapt to
any combination of the following options:
1. Selection of adsorptive media.
2. Treatment with or without raw and treated water
pH adjustment.
3. Spent adsorptive media regeneration or
replacement.
4. Manual, semiautomatic, or automatic operation.
If a treatment system employs chemicals for media
regeneration, it is prudent to use the same chemicals
to adjust the pH of the treatment process.
A four-step design process is employed in this manual.
The included steps are as follows:
1. Assemble design input data and information.
2. Conceptual Design.
3. Preliminary Design.
4. Final Design.
3.1 Assemble Design Input Data
and Information
The design input data and information should be
established prior to initiating the conceptual design. The
design input data and information include, but are not
limited to, the following:
1. Chemical analyses (see Figure 3-1) of representa-
tive raw water samples (includes all historical
analyses). Comprehensive raw water analyses of
all inorganic, organic, radionuclide, and
bacteriological contaminants also are required to
verify that this adsorptive media process is
applicable for the selective removal of arsenic.
2. Treated water quality compliance standards
issued by the regulatory agency within whose
jurisdiction the system resides.
3. Regulatory design standards.
4. Wastewater and waste solids disposal ordinances
issued by the responsible regulatory agency.
5. Data on system production and consumption
requirements (present and future).
6. Manual vs. automatic operation.
-------
CONTAINER
SAMPLE DATE
TAKEN BY:
Analysis *
Calcium
Magnesium
Sodium
Total Cations
Total Alkalinity (M)**
Phenolphthalein Alkalinity (P)**
Total Hardness**
Sulfate
Chloride
Nitrate
Phosphate (PO4)
Silica (SiO2)
Free Carbon Dioxide
Iron (Fe) Unfiltered
Iron (Fe) Filtered
Manganese
Turbidity (NTU)
Color (Units)
Fluoride
Total Arsenic
Soluble Arsenic
Particulate Arsenic
Arsenic (III)
Arsenic (V)
PH (Units)
Specific Conductance (micro-mhos)
Temperature (°F)
* All units reported in mg/L excepted as noted.
** as CaCO3.
Figure 3-1. Arsenic Removal Water Treatment Plant Water Analysis Report
10
-------
The treatment system is a subsystem within the larger
water utility system. Other subsystems are the well pump,
the storage reservoirs, the pressurization system, and
the distribution system. This design manual is applicable
when arsenic removal is the only treatment required.
Removal of other contaminants such as bacteria, sus-
pended solids, hardness, organics, or other contami-
nants also may be required. In those cases, alternative
treatment processes and/or additional treatment proces-
ses should be evaluated.
The sequence of other treatment steps should be com-
patible with the selected adsorptive media arsenic re-
moval method. Removal of suspended solids, organics,
and hardness should take place upstream of the adsorp-
tive media arsenic removal process. Disinfection with
chlorine should take place after arsenic removal using
activated alumina because it has been the author's
experience that chlorine will degrade the performance of
activated alumina. No known investigation has deter-
mined the amount of chlorine that can be tolerated by
the alumina; however, process degradation has been
eliminated on projects conducted by the author where
prechlorination was terminated. If chemical oxidation is
required for the conversion of As(lll) to As(V) for the
successful performance of another type of adsorptive
media, it is recommended that the preoxidation chemical
be prevented from coming in contact with the media,
unless advised otherwise by the media manufacturer.
Other treatment processes may be required upstream of
the arsenic removal process, but that decision will be
made on a case-by-case basis.
For ground water systems, the most practical concept is
to install the treatment plant in the immediate vicinity of
the well (space permitting). The well pump then will
deliver the water through treatment into distribution and/
or storage. If the existing well pump is oversized (pumps
at a much higher flowrate than the maximum daily flow-
rate requirement), it should be resized to deliver slightly
more (i.e., 125% minimum) than the peak requirement.
The flowrate dictates the treatment equipment size and
capital cost. The design rate should be minimized to the
extent possible to ensure that the capital cost of the
treatment system is minimized. Reducing flowrate for an
oversized pump can result in excessive equipment wear
and energy costs. The treatment media volume is a func-
tion of flowrate. The treatment vessels, pipe sizes, and
chemical feedrates all increase as the flowrate increases.
A well-matched pump likely can handle any additional
head loss associated with the treatment system without
significant drop in pump efficiency. If the additional head
loss cannot be met with the existing pump, several
options exist: increasing the size of the motor, increasing
the size of the impeller, or replacing the pump. Storage
should be provided to contain a minimum of one half the
maximum daily consumption requirement. This is based
on the premise that maximum consumption takes place
during 12 hrs of the day. Then, if treatment operates
during the entire 24 hrs, storage drawdown occurs
during 12 hrs and recovers during the remaining 12 hrs.
Construction materials must comply with OSHA stand-
ards, local building codes, and health department, require-
ments in addition to being suitable for the applicable pH
range and compatible with any pretreatment chemicals
used (e.g., chlorine, ozone, etc.). Both drinking water
treatment chemicals and system components should
comply with NSF/ANSI STD 61.
Treatment system equipment should be protected from
the elements. Although not mandatory in some locations,
it is prudent to house the system within a treatment
building.
Wastewater resulting from backwash and regeneration
of the treatment media can only be disposed of in a
manner permitted by state and/or local regulatory
authorities. Several options are available for disposal;
however, they are subject to climate, space and other
environmental limitations. Because each of the variables
can significantly affect both capital and operating costs,
careful evaluation of the available wastewater handling
options is required prior to making conceptual selections.
3.2 Conceptual Design
The second step in the design process is the conceptual
design, which provides a definition of the process. How-
ever, it does not provide equipment size, arrangement,
material selection, details, or specifications. Using the
design input data and information previously described
(Section 3.1), the following decisions should be made:
1. Selection of adsorptive media.
2. Decision on whether to implement treatment process
pH adjustment.
3. Decision regarding regeneration or replacement of
spent adsorptive media (applicable if treatment
process pH is used).
4. Decision regarding implementation of manual vs.
automatic operation (applicable if spent adsorptive
media is regenerated).
11
-------
There are four basic options from which a Conceptual
Design can be selected. Every combination of options
will perform the process and, under a selected set of
conditions, a certain combination may be preferred. The
options are as follows:
1. Gravity or pressure flow
2. Single or multiple treatment bed(s)
3. Upflow or downflow treatment flow direction
4. Series or parallel treatment vessel arrangement.
An efficient, cost-effective configuration is a pressure
system utilizing a dual vessel series downflow configura-
tion with bypass and reblending of raw water. Some
state regulations, however, may not allow the bypass of
untreated water to be blended with treated water. The
two-bed series configuration yields the highest arsenic
loading on the treatment media and the lowest treated
water arsenic level. The single treatment unit configura-
tion generally is less efficient unless there is an excep-
tionally large treated water storage capacity. A gravity
flow system does not provide the economics of a pres-
sure system; treatment flowrates are lower, repumping
of treated water is always required, and capital costs are
higher. Because free carbon dioxide (CO2) is released to
the atmosphere in gravity systems utilizing treatment
process pH adjustment, pH adjustment is easier to con-
trol in a pressure system. Downflow treatment has con-
sistently yielded higher arsenic removal efficiency than
upflow. Because the downflow concept utilizes a packed
bed, the flow distribution is superior. If the upflow beds
are restrained from expanding, they would in effect also
be packed. However, they would forfeit the necessary
capability to backwash. Once the bed configuration is
defined, a basic schematic flow diagram is prepared (see
Figures 3-2, 3-3, and 3-4). These diagrams present all of
the subsystems without pH adjustment and regeneration
(Figure 3-2), with pH adjustment without regeneration
(Figure 3-3), and with pH adjustment and regeneration
(Figure 3-4). An illustration of the treatment unit is pro-
vided as Figure 3-5. A summary of subsystem compo-
nents is presented in Appendix A.
For systems in which the raw water arsenic concentra-
tion is slightly above the arsenic MCL, bypassing and
reblending a fraction of the raw water with the remaining
fraction that is treated should be evaluated. This option
saves treatment chemicals, extends treatment media
cycle life, and reduces operating cost. If bypassing and
blending is found to be feasible, the treatment system
can be sized to treat less than 100% of the total flow.
Prior to proceeding with the Preliminary Design, financial
feasibility should be determined. Funding limits for the
project should be defined. A determination that funding
is available to proceed with the project should be made;
this requires a preliminary rough project estimate with an
accuracy of ±30%. If the preliminary rough estimate
exceeds the available funds, adjustments should be
made to increase funding or reduce project costs.
3.2.1 Manual Operation
In a manual operation, the treatment plant operator per-
sonally performs all of the operating functions and makes
all operating decisions. The treatment plant equipment
does not accomplish any function independent of the
operating personnel. The equipment is simple and per-
forms the basic functions that the operator implements.
The manual operation includes the following:
1. Motors (pumps, chemical pumps, etc.) with manual
start/stop controls. Some motors have manual speed
adjustment capability. Chemical pumps have man-
ual speed and stroke length adjustment capability.
2. Valves with manual handle, lever, handwheel, or
chainwheel operators.
3. Instrumentation sensors with indicators. Instrumen-
tation is installed in-line where operating data (flow-
rate, total flow, pressure, pH, and liquid levels) are
indicated. In-line pH sensors, magmeters, ultrasonic
level sensors are other instruments that require
electric service.
The adsorptive media treatment process can perform
manually with or without treatment process pH adjust-
ment and with spent media replacement or regeneration.
3.2.2 Automatic Operation
In an automatic operation, the treatment plant is oper-
ated by a PLC, which initially is programmed by the
operator, the computer supplier, or an outside specialist.
If programmed by someone other than the plant oper-
ator, the operator should be trained by that individual to
adjust program variables and, if necessary, modify the
program. The operator interface and printer are the equip-
ment items which the operator uses during the perform-
ance of treatment plant functions. In addition, the operator
should calibrate and check all of the components of the
automatic operating equipment system on a routine
periodic basis. Finally, the treatment plant operator or a
designated instrumentation and control specialist should
be capable of performing emergency maintenance and/
or repair of all components.
Every function included in an automatic system should
be capable of manual operation.
12
-------
FROM
SOURCE
<•
BYPASS (OPTIONAL)
BLENDED WATER
TO DISTRIBUTION
r/f) RAW (FEED)
X WATER
-»-o
-------
FROM
SOURCE
v-
•"1 1X3—IF
> BYPASS (OPTIONAL)
RAW (FEED) V
BLENDED WATER
TO DISTRIBUTION
.^—l^.
N0 on
MY MNK S
BACKWASH
TO WASTE
-*"9-
-£T- CHECK
© pH
© FLOW
@ FLOW
©
*/¥
0
S
IN LINE
FEED PUMP
Figure 3-3. Flow Diagram for Dual Vessel Series Downflow Treatment System With pH Adjustment,
With Replacement of Spent Media
14
-------
BYPASS (OPTIONAL)
WATER
TO OtSTRiaiJTION
EGEND
TO TANK
mt TWIK
CONTROL VALVE
CHECK VALVE
pH SENSOR/ANALYZER
FLOW INDICATOR
FLOW TOTALIZER
PRESSURE INDICATOR
PRESSURE RELIEF VALVE
AIR/VACUUM VALVE
SAMPLE
IN LINE STATIC
CHEMICAL EEEO PUMP
Figure 3-4. Flow Diagram for Dual Vessel Series Downflow Treatment System With pH Adjustment,
With Regeneration of Spent Media
15
-------
FREEBOARD
50% BED
EXPANSION
TREATMENT
MEDIA
6"
h/2
SS
I"1 Minimum clearance
required between
bottom of vessel and
concrete pad
d
h
V
M,
Mf
0
dH
H
SS
GIVEN
d >
y „,.
D =
y =
SYMBOLS
- TREATED WATER FLOW RATE (gpm)
- TREATMENT BED DIAMETER (ft,), d =
- TREATMENT BED DEPTH (ft.)
- TREATMENT BED VOLUME - J3t!ii (ft,1)
- DENSITY OF TREATMENT MEDIA (Ib./ft.5)
- WEIGHT OF MEDIA (Ibs.)
- OUTSIDE DIAMETER OF TREATMENT VESSEL (ft.)
- DEPTH OF DISHED PRESSURE HEAD (ft.)
- OVERALL HEIGHT OF SKID MOUNTED TREATMENT (ft)
- SIDE (ft)
h/2, 3'-0" < h < B'-O"
2 dH + h 4- h/2 + 6* + 1*
d + 1"
45 )b,/fts WITH IN
Md x V = 45V (Ib.)
Figure 3-5. Treatment Bed and Vessel Design Calculations
16
-------
The automatic equipment is more sophisticated and
costly than that used in a manual operation. When func-
tioning normally, an automatic operation can function
continuously with minimal operator attention. This is
recommended for treatment systems in remote areas or
areas that are difficult to access, and systems for which
operator availability is limited. The automatic operation
includes the following:
1. Motors (pumps, chemical pumps, air compressors,
etc.) with automatic start/stop and speed adjustment
controls. Chemical pumps may have manual stroke
length adjustment. Motors should also have a
manual on/off control.
2. Valves with either pneumatic or electric operators.
Flow or pressure control valves with electronic posi-
tioners for valves with automatic operators. Valves
require manual overrides for operation during start-
up, power failure or compressed air failure. Valves
should have opening and closing speed controls to
prevent water hammer during automatic operation.
Valve electric position indicators are optional.
3. Automatic instrumentation may be electronic,
pneumatic, or a combination. The instruments and
controls should always be capable of transmitting
and receiving electronic information to and from the
PLC. In a fully automatic system all of the control,
monitoring, and alarm functions are monitored and
controlled by the PLC. Backup manual instruments
(e.g., flowrate indicators, pressure indicators, pH
indicators, and liquid level indicators) are recom-
mended to provide verification of automatic instru-
mentation if treatment plant budget is available.
Comprehensive automatic alarms that notify
operators and/or shut down increments or the entire
treatment system relating to every type of system
malfunction at the moment such events occur is a
necessary function that should be incorporated in all
applicable instrumentation components.
The adsorptive media treatment process with automatic
operation can perform with or without treatment process
pH adjustment and with spent media replacement or
spent media regeneration. Automatic operation is most
applicable to systems with treatment process pH adjust-
ment and treatment media regeneration. Systems
employing adsorptive media without treatment process
pH adjustment and media regeneration do not benefit
greatly from computer-controlled operations.
A semiautomatic operation that employs individual con-
trollers to automatically start/stop or adjust some, but not
all, of the operational items in the system can contribute
significantly to the treatment system operation without
computer control of the entire operation. These semi-
automatic functions should include alarms that will notify
operators of process functions exceeding limits estab-
lished for effective and/or safe operation. Alarm events
can be staged at single (e.g., high) or dual (e.g., high-
high) levels. In a dual-level alarm, the first level notifies
the operator that the performance is out of tolerance,
and the second level shuts down either a single process
function (e.g., a pump) or the entire process. Examples
of semiautomatic operational functions include, but are
not limited to, the following:
1. Flow control loop includes an electronic flow sensor
with totalizer (e.g., magnetic flowmeter) that sends
an electronic signal to an electronic flow controller
(with high and low flowrate alarms), which in turn
sends an electronic signal to a flow control valve
(butterfly valve or ball valve) with an actuator and
electronic positioner. The plant operator designates
the required flowrate at the flow controller. The
controller receives the flowrate measurement from
the flow sensor and transmits signals to the flow
control valve positioner in order to adjust the valve
position until the flowrate matches that required by
the process. If the flowrate deviates from the limits
established for the process, then a high flowrate or
low flowrate alarm will be issued.
2. Pressure control loop includes an electronic pres-
sure transmitter that sends an electronic signal to an
electronic pressure controller (with high and low
pressure alarms), which in turn sends an electronic
signal to a pressure control valve with an actuator
and electronic positioner. The plant operator
designates the required pressure at the pressure
controller. The controller receives the pressure
measurement from the pressure transmitter and
transmits signals to the pressure control valve posi-
tioner to adjust the valve position until the pressure
matches that required by the process. If the pres-
sure deviates from the limits established for the
process, then a high pressure or low pressure alarm
should be issued.
3. pH control loop includes an electronic pH sensor
which transmits a pH signal to a pH analyzer (with
high and low level alarms) which in turn sends an
electronic signal to a converter which transmits a
pulse signal to a chemical feed pump (acid or
caustic) to adjust the feed pump stroke speed. The
plant operator designates the required pH at the pH
analyzer. The pH analyzer receives the pH mea-
surement from the pH sensor and transmits signals
to the chemical feed pump (via the converter) to
adjust the pump stroke speed until the pH matches
that required by the process. If the pH deviates from
17
-------
the limits established for the process, then a high pH
or low pH alarm should be issued.
4. Liquid level control loop includes an electronic liquid
level sensor (e.g., ultrasonic level sensor) which
transmits an electronic liquid level signal to a level
controller which indicates the liquid level and trans-
mits an electronic signal to one or more motors
(pump, mixer, etc.) to start or stop. At the level con-
troller, the plant operator designates the required
liquid levels at which motors are to start and stop.
The level controller receives the liquid level mea-
surement from the liquid level sensor and transmits
signals to the motor(s) to start or stop. If the liquid
level deviates from the limits established for the
process, then a high or low liquid level alarm should
be issued.
Many other process functions are performed automat-
ically by means of relays and other electrical devices. An
example is the electrical interlock of chemical feed
pumps with raw water pumps, which prevents chemical
feed into the process without the flow of process water.
Another example is the use of a flow switch in a pres-
sure relief valve discharge pipe, which, upon detection of
water flow, issues an alarm and stops the process feed
pump. The list of individual failsafe automatic functions
can be extensive. All applicable codes, standards, and
OSHA requirements should be reviewed to determine
which requirements are applicable to the project. Then
based upon sound judgment, available budget, treat-
ment plant operator capability, and availability, a deci-
sion should be made as to whether a given function
should be automatic or manual.
3.3 Preliminary Design
After completion and approval of the Conceptual Design
by the client, the regulatory agency(s), and any other
affected party, the Preliminary Design commences. This
stage includes sizing of the equipment, selecting materi-
als for construction, determining an equipment layout,
and upgrading the preliminary capital cost estimate to a
±20% accuracy. The deliverable items are:
1. Schematic flow diagrams (see Figures 3-2, 3-3,
and 3-4)
2. Preliminary process equipment arrangement
drawings (see Figures 3-6, 3-7, and 3-8 for
examples)
3. Outline specifications
4. Preliminary capital cost estimate (see Table 3-1).
3.3.1 Treatment Equipment
Preliminary Design
This section provides the basic methodology for sizing
equipment items and selecting materials of construction
for arsenic removal treatment systems using granular
adsorptive media with pH adjustment and regeneration
of exhausted treatment media. An example illustrating
this method is provided in Appendix B. The example is
based on use of dual vessel series downflow granular
adsorptive media with pH adjustment, exhausted media
regeneration, and manual operation. The empty bed
contact time (EBCT) used for this application is 5 min
per vessel. For systems using different process param-
eters (EBCT, without pH adjustment, with disposal of
exhausted adsorptive media) the design information pre-
sented in this document is easily adjusted. For automatic
or semiautomatic operation the system basic design does
not change; however, equipment material and installa-
tion costs will vary.
3.3.1.1 Treatment Bed and Vessel Design
In accordance with the discussion presented in Section
2.2, the recommended treatment concept is based on
the use of two treatment pressure vessels piped in ser-
ies using the downflow treatment mode. Treatment ves-
sel piping also should be configured to provide for media
backwashing (upflow). The treatment vessel materials of
construction employed in the design example presented
in Appendix B are carbon steel (grade selection based
on cost-effective availability) fabrication, assembly, and
testing that complies with American Society of Mechan-
ical Engineers (ASME) Code Section VIM, Division 1.
The interior should be lined with abrasion-resistant vinyl
ester or epoxy coating. Interior lining material should be
NSF-certified for potable water application, and suitable
for pH range 2.0-13.5. Vessel pressure rating should be
50 psig or the minimum necessary to satisfy system
requirements. Other vessel materials of construction
(e.g., fiberglass), internal lining materials (e.g., abrasion
resistant epoxy, rubber, etc.), and stainless steel without
lining, may also be employed.
Prior experience with activated alumina indicates that
the volume of treatment media (V) in each treatment
vessel is • ft3 per gpm of process water flowrate to pro-
vide an EBCT time of 10 min in the dual vessel downflow
process (e.g., 5 min EBCT in each vessel). Actual
residence time is approximately half the EBCT, because
the space between the grains of media is approximately
50% of the total bed volume. (Note: When raw water is
bypassed and blended back with treated water, only the
treated water is included in sizing the treatment media
volume.) In order to prevent "wall effects", bed diameter
(d) should be equal to or greater than one-half the bed
18
-------
&&S&Z3&.
A/Wv^/M
(OPTIONAL)
BYPASS
BACKWASH
*• WATER
TO
SAMPLE
PANEL
Figure 3-6. Treatment System Plan for Adsorptive Media Without Process Water pH Adjustment and
With Spent Media Replacement
depth (h). Good practice indicates that bed depth should
be a minimum of 3 ft and a maximum of 6 ft. At less than
minimum depth, distribution problems may develop; and,
at greater than maximum depth, fine material removal
and pressure loss becomes a problem. For very small
systems using tanks of 1-2 ft in diameter, the bed depths
could be as low as 2 ft. The treatment bed and vessel
design is illustrated in Figure 3-5. A typical example for
determining treatment bed and treatment vessel dimen-
sions is presented in Appendix B.
Five minutes is recommended as a minimum limit for the
EBCT for activated alumina. For EBCTs of other adsorp-
tive media, the designer should rely either upon the
manufacturer's instructions, or develop the technical
data independently by means of field pilot studies. As
the EBCT decreases below the recommended value,
two undesirable features occur. First, the treatment is
less efficient (% arsenic removal is reduced), resulting in
treated water arsenic concentration not reaching a low
enough concentration; and second, regeneration fre-
quency or spent media replacement frequency increases,
requiring more operating cost, operator attention, and
proportionately more downtime. Conversely, raising the
EBCT above the recommended level increases the size
of the treatment beds and their vessels, thereby increas-
ing capital cost and space requirements.
19
-------
(OPTIONAL)
ACID
TANK
o
(OPTIONAL)
CAUSTIC
TANK
(OPTIONAL)
BACKWASH
TO WASTE
PANEL
Figure 3-7. Treatment System Plan for Adsorptive Media With Process Water pH Adjustment and
Spent Media Replacement
20
-------
ACID
STORAGE TANK
CAUSTIC
STORAGE TANK
CAUSTIC DAY TANK
FOR TREATED
Ph ADJUSTMENT AND
SPENT MEDIA
REGENERATION
(OPTIONAL)
BYPASS
REGENERATION
WASTEWATER
TO SURGE TANK
(NOT
SAMPLE
PANEL
Figure 3-8. Treatment System Plan for Adsorptive Media With Process Water pH Adjustment and
Spent Media Regeneration
21
-------
Table 3-1. Preliminary Capital Cost Estimate Examples for Four Types of Adsorptive Media
Arsenic Removal Water Treatment Plants
Location:
Flowrate: 570 gpm
Cost ($1,000)
Date: Manual Operation Manual Operation
w/Media w/Media
Replacement Replacement
w/o pH Adjustment w/pH Adjustment
Manual Operation
w/Media
Regeneration
w/pH Adjustment
Automatic
Operation w/Media
Regeneration w/pH
Adjustment
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, Valves, and Accessories
Instruments and Controls
Chemical Storage Tanks
Chemical Pumps and Accessories
Subtotal
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
78
33
27
8
N/A
N/A
146
Process
30
12
13
55
78
33
34
13
45
5
208
Equipment Installation
42
22
15
79
78
33
50
19
45
10
235
44
22
15
81
78
33
68
70
45
10
304
48
38
15
101
Miscellaneous Installed Items
Regeneration Wastewater Surge Tank
Building and Concrete
Site Work, Fence, and Miscellaneous
Subtotal
Contingency 20%
Total(a)
N/A
45
15
60
53
314
N/A
70
17
87
75
449
130
70
24
224
108
648
130
70
24
224
126
755
(a) Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes not
included.
N/A = not applicable.
Pressure vessel fabrication is standardized by diameter
in multiples of 6-inch outside diameter increments. Tool-
ing for manufacture of pressure vessel dished heads is
set up for that standard. Design dimensions differentiate
between pressure vessel and treatment bed diameters.
The vessel outside diameter (D) is approximately 1 inch
greater than the bed (or vessel inside) diameter, which
conservatively provides for both vessel walls with lining
as well as fabrication tolerances. If the pressure is high
(100 psig or greater), the 1 inch will increase to reflect
the increased vessel wall thickness.
Although many methods are available for distributing the
water flow through a treatment bed, the following method
has been successfully used in adsorptive media water
treatment plants that are presently in operation. The
water is piped downward into the vessel through an inlet
diffuser. This diverts the flow into a horizontal pattern.
From there it radiates in a horizontal plane prior to
starting its downward flow through the adsorptive media
bed. The bed, in turn, is supported by a false flat bottom,
which is supported by the bottom head of the pressure
vessel by means of concentric rings. The false flat
bottom also supports the horizontal header and plastic
fabric sleeved perforated lateral collection system. Treat-
ment media are placed in the vessel through circular
manway(s) with hinged cover(s) in the top head of the
vessel.
3.3.1.2 Pipe Design
For systems with treatment process pH adjustment and
spent media regeneration, material should be suitable
for ambient temperature, pH range of 2.0-13.5, system
pressure, and potable water service. At a low pH, carbon
steel is not acceptable unless interior lining is included.
Stainless steel is acceptable; however, it may be too
costly. Plastic materials such as polyvinyl chloride (PVC)
are satisfactory. PVC is usually the best selection based
on its availability, NSF certification for potable water
service, low cost, and ease of fabrication and assembly.
The drawbacks to the PVC materials are their loss of
strength at elevated temperatures (above 100°F); their
coefficients of thermal expansion; their external support
requirements; their deterioration from exposure to sun-
light; and their vulnerabilities to damage from impact.
Nevertheless, these liabilities are outweighed by the low
cost and suitability for the service. The piping can easily
be protected from all of the above concerns, except ele-
vated ambient and/or water temperatures. If elevated
temperature exists, the use of FRP pipe is recom-
mended. This material provides the strength and support
that is lacking in the pure plastic materials.
22
-------
For systems without treatment process pH adjustment
metallic pipe (e.g., carbon steel, copper, etc.) may be
used in place of plastic. However, care must be exer-
cised to prevent occurrence of corrosive conditions
including, but not limited to, process water pH, free CO2,
chloride concentration, and sulfate concentration, as well
as galvanic and pit corrosion.
The piping system should be economically sized to allow
for delivery of design flow without excessive pressure
losses. If water velocities present conditions for water
hammer (due to fast closing valves, etc.), shock-prevent-
ing devices will be provided.
Isolation and process control valves should be wafer
style butterfly type, except in low flowrate systems where
small pipe size dictates the use of true union ball valves.
The use of inexpensive, easily maintained valves that
operate manually provides minimum capital cost. The
valves are automated by the inclusion of pneumatic or
electric operators.
Pressure regulator and rate of flow control valves are
recommended for safe operation of manually controlled
treatment systems.
See Appendix B for pipe size design using the example
employed for vessel and treatment media design.
3.3.1.3 Instrumentation Design
System functional requirements that are adapted to com-
mercially available instruments should be specified.
Included are:
Range
Varies'3'
Varies'3'
0-14
Varies'3'
30-120°F
Accuracy
±2%
±1%
±0.1
±1%
±1%
Instrument
1. Flow sensor
(indicator/totalizer)
2. Pressure indicator
3. pH sensor/analyzer/alarm
4. Level sensor/indicator
5. Temperature indicator
(optional)
(a) Range to be compatible with application, maximum measurement
not to exceed 90% of range.
3.3.1.4 Acid Storage and Feed Subsystem
Acid feed and storage subsystems are included with
treatment systems that include pH adjustment of process
water only (with or without regeneration of exhausted
adsorptive media.) The acid storage tank should be
sized to contain tank truck bulk delivery quantities of
concentrated sulfuric acid. For water systems that are
not permitted to increase the sulfate concentration of the
water, hydrochloric acid can be substituted. However,
this acid is more costly, more difficult to handle, and
results in highly corrosive treated water; therefore hydro-
chloric acid is not recommended. Bulk delivery provides
the lowest unit price for the chemical. In small plants,
acid consumption may not be enough to justify large vol-
ume purchase of chemicals. In the smaller plants, drums
or even carboys may be more practical; therefore, for
that type operation, the requirement for a storage tank is
eliminated. A 48,000-lb tank truck delivers 3,100 gal of
66°B« H2SO4 (15.5 Ib/gal). A 5,000-gal tank provides a
50% cushion. The example in Appendix B illustrates the
method of designing the components of this system.
The sulfuric acid carbon steel storage tank does not
require an interior lining; however, the interior should be
sandblasted and vacuum-cleaned prior to filling with acid.
The storage tank should be protected from the elements
and include a containment basin located outside of the
treatment building. Typically, the containment basins are
sized for 110% of the capacity of the storage tank. The
66°B» H2SO4 freezes at -20°F. Therefore, unless the treat-
ment plant is located in an extremely cold climate, no
freeze protection is required. All piping is to be 2-inch
carbon steel with threaded cast iron fittings and plug
valves. Elastomer seals, seats and gaskets should be
Vitone.
The acid pumps are standard diaphragm models with
materials of construction suitable for 66°B» H2SO4 service.
Standard sulfuric acid service pumps should be speci-
fied. In the preliminary design, the sizing is determined
by field test or theoretical calculation (see Appendices B
and C.) Acid feedrate varies with the total alkalinity and
the free CO2 content of the raw water. The feedrate is
accurately determined experimentally by adjusting a raw
water sample pH to 5.5 by acid titration. In a manual
treatment plant operation, the operator should check the
pH periodically and maintain it at 5.5. The pump stroke
speed and length should be adjustable to accommodate
these variations. An in-line static mixer should be
installed immediately downstream of each acid injection
point. This provides thorough mixing of the acid which
results in an accurate pH measurement by a pH sensor
located at the discharge end of the mixer. The pH
probes that are used to control pH should be calibrated
against standard buffers at least once per week. For
treatment systems that regenerate adsorptive media an
acid feed is required to lower pH of water for neutraliza-
tion of a treatment bed prior to placing that bed back into
treatment service after regeneration. Neutralization pH
feedrate is initially set at 2.5 and increases in steps until
the treatment pH of 5.5 is achieved. Finally wastewater
from the regeneration of an adsorptive media bed is
collected in a surge tank where the pH is adjusted to 6.5,
a level at which arsenic coprecipitates with aluminum
hydroxide or ferric hydroxide. An additional acid feed
pump is required to feed acid to the wastewater.
23
-------
3.3.1.5 Caustic Soda Storage and
Feed Subsystem
The caustic soda storage tank also is sized to contain
tank truck bulk delivery quantities of 50% or 25% sodium
hydroxide. A 48,000-lb tank truck delivers 3,850 gal of
50% NaOH which provides a 25% cushion in a 5,000-gal
storage tank. 50% NaOH freezes at 55°F; 25% NaOH
freezes at 0°F. Therefore, 50% NaOH, which is prefer-
able because of price, requires an immersion heater to
prevent freezing. The caustic is used for treatment bed
regeneration and neutralization of treated water. Regen-
eration frequency is a function of raw water arsenic
concentration, flowrate and treatment media arsenic
capacity. The amount of caustic required to neutralize
the treated water, that is to raise the pH from 5.5 to the
pH required for corrosion protection for the water sys-
tem, is a function of the water chemistry at each installa-
tion. The actual caustic feedrate is easily determined
experimentally by readjusting the treated water pH by
titrating a sample with caustic until the desired pH is
achieved. If a fraction of the raw water bypasses treat-
ment and is blended with treated water, then the chem-
ical required for pH adjustment is reduced. In raw water
with high alkalinity the lowering of pH produces high
levels of dissolved CO2. In those waters, removal of the
CO2 by aeration raises the pH (prior to blending), pro-
viding a less expensive treatment due to reduction of
caustic required to raise the pH of the treated water. In
low alkalinity water, the chemical addition is less expen-
sive. The carbon steel caustic storage tank is covered in
Appendix B. This vessel should be heat-treated to stress
relieve welds. The carbon steel does not require an
interior lining; however, it does require sandblasting and
vacuum-cleaning prior to filling. All piping is to be 2-inch
carbon steel with threaded cast iron fittings and plug
valves. Elastomer seals, slots and gaskets should be
ethylene propylene diene monomer (EPDM).
Because 50% NaOH freezes at 55°F, it should maintain
a minimum temperature of 70°F. This is handled by
a temperature-controlled electrical immersion heater.
Twenty-five percent sodium hydroxide freezes at 0°F;
therefore, unless it is located in an extremely cold cli-
mate, freeze protection is not required. The storage tank
should be placed in a containment basin inside of an
enclosure outside of the treatment building.
A pump is required to feed caustic into the effluent main
through an in-line static mixer where the treated water is
neutralized. For regeneration, a larger caustic feed pump
is required for pumping the caustic through a static mixer
in the regeneration feed pipe. There the caustic is diluted
to the 5% (by weight) concentration required to regener-
ate the adsorptive treatment media.
3.3.2 Preliminary Treatment Equipment
Arrangement
Once all of the major equipment size and configuration
information is available, a layout (arrangement drawing)
is prepared. The layout provides sufficient space for
proper installation, operation and maintenance for the
treatment system as well as each individual equipment
item. OSHA standards should be applied to these deci-
sions during the equipment arrangement design stage.
These requirements may be supplemented or super-
seded by state or local health and safety regulations, or,
in some cases, insurance regulations. A compact
arrangement to minimize space and resulting cost
requirements is recommended. Figures 3-6, 3-7, and 3-8
illustrate typical preliminary arrangement plans. These
arrangements provide no frills, but do include ample
space for ease of operation and maintenance. Easy
access to all valves and instruments reduces plant oper-
ator effort.
The type of building used to protect the treatment
system (and operator) from the elements depends on
the climate. Standard pre-engineered steel buildings are
low-cost, modular units. Concrete block or other material
also may be used. Standard building dimensions that
satisfy the installation, operation, and maintenance space
requirements for the treatment system should be
selected. The building should provide access doors, light-
ing, ventilation, emergency shower and eye wash, and a
laboratory bench with sink. All other features are optional.
When the arrangement is completed, the preliminary
cost estimate is prepared.
Manual operation is the method employed in the design
example in Appendix B. The basic process requirements
should be reviewed at each stage of design to assure
that every item required to operate the process is
included. Although detail design occurs during the final
design phase, provision for operator access for every
equipment item should be provided. Automatic operation
does not require total accessibility; access for mainte-
nance functions for which ladder or scaffold access will
suffice. The extra equipment items required solely for
automatic operation (including but not limited to PLC,
and operator, interface) occupy minimal space and are
located in positions that are most accessible to the
operator.
3.3.3 Preliminary Cost Estimate
At completion of the Preliminary Design, the preliminary
cost estimate is prepared based upon the equipment
24
-------
that has been selected, the equipment arrangement and
the building selection. This estimate should be based on
the material equipment quantities, unit prices to labor
and material, and finally summarized in a format that is
preferred by the owner (see Table 3-1 for example). This
estimate should have an accuracy of ±20%. To assure
sufficient budget for the project, it is prudent to estimate
on the high side at this stage of design. This may be
accomplished by means of a contingency to cover
unforeseen costs, and an inflation escalation factor.
3.3.4 Preliminary Design Revisions
The Preliminary Design package (described above) then
is submitted for approval prior to proceeding with the
Final Design. This package may require the approval of
regulatory authorities, as well as the owner. Requested
acceptable changes should be incorporated and resub-
mit for approval. Once all requested changes are imple-
mented and Preliminary Design approval is received, the
Final Design can proceed.
3.4 Final Design
After completion and approval of the Preliminary Design
by the client, the Final Design proceeds. This includes
detail design of all of the process equipment and piping,
complete process system analysis, complete detail
design of the building including site work, and a final
capital cost estimate accurate to within 10%. The deliv-
erable items are:
1. Complete set of construction plans and
specifications
2. Final capital cost estimate (See Table 3-2).
The Final Design starts with the treatment system equip-
ment (if applicable, including the wastewater surge tank);
continues with the building (including concrete slabs and
foundations, earthwork excavation/backfill/compaction,
heating, cooling, painting, lighting, utilities, laboratory,
personnel facilities, etc.); and finishes with the site work
(including utilities, drainage, paving and landscaping).
The latter items apply to every type of treatment plant;
although they are integral with the treatment system,
they are not addressed in this manual. The only portions
of the Final Design that should be addressed are the
pertinent aspects of the treatment equipment which were
not covered in the Preliminary Design (Section 3.3). Dur-
ing the Conceptual Design and Preliminary Design, the
basic equipment that accomplishes the required func-
tions were selected, sized, and arranged in a compact,
efficient layout. The decision was cost-conscious, using
Table 3-2. Final Capital Cost Estimate Examples for Typical Location for Four Types of
Adsorptive Media Arsenic Removal Water Treatment Plants
Location:
Flowrate: 570 gpm
Cost ($1,000)
Date:
Manual Operation
w/ Media
Replacement
w/o pH Adjustment
Manual Operation
w/Media
Replacement
w/pH Adjustment
Manual Operation
w/Media
Regeneration
w/pH Adjustment
Automatic
Operation w/Media
Regeneration w/pH
Adjustment
Treatment Vessels
Treatment Media
Process Piping, Valves, and Accessories
Instruments and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, and Accessories
Subtotal
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Regeneration Wastewater Surge Tank
Building and Concrete
Site Work, Fence, and Miscellaneous
Subtotal
Contingency 10%
Total"1
Process Equipment
73 73
31 31
32 36
7 11
N/A 40
N/A 6
143 197
Process Equipment Installation
31 43
10 17
10 13
51 73
Miscellaneous Installed Items
N/A N/A
40 62
14 15
54 77
25 35
273 382
73
31
49
16
40
12
221
46
17
13
76
120
62
23
205
51
553
73
31
64
66
40
13
287
51
41
13
105
120
62
23
205
60
657
(a) Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes not
included.
25
-------
minimum sizes (or standard sizes) and the least expen-
sive materials that satisfied the service and/or environ-
ment. However, in the Final Design, this effort can be
defeated by not heeding simple basic cost control princi-
ples. Some of these are:
1. Minimize detail (e.g., pipe supports-use one style,
one material, and components common to all sizes).
2. Minimize the number of bends in pipe runs (some
bends are necessary-those that are optional only
increase costs).
3. Minimize field labor; shop fabricate where possible
(e.g., access platforms and pipe supports can be
mounted on brackets that are shop fabricated on
vessel).
4. Skid-mount major equipment items (skids distribute
weight of vessels over small mat foundations in
place of piers and spread footings, thereby costly
foundation work is eliminated).
5. Use treatment vessels as a heat sink to provide
insulated building cooling or heating or both
(eliminates heating and/or cooling equipment in
addition to reducing energy cost). Consideration
must be given, however, to humid climates where
cold tanks will result in sweating problems.
6. Simplify everything.
All subsystems should be analyzed (refer to schematic
flow diagrams in Figure 3-2, 3-3, and 3-4) to account for
all components in both equipment specifications and
installation drawings. The drawings and specifications
should provide all information necessary to manufacture
and install the equipment. Extra effort to eliminate ambi-
guity in detail and/or specified requirements should be
exercised. All items should be satisfactory for service
conditions besides being able to perform required func-
tions. Each item should be easy to maintain; spare parts
necessary for continuous operation should be included
with the original equipment. All tools required for initial
startup as well as operation and maintenance should be
furnished during the construction phase of the project.
Once construction, equipment installation, and checkout
are complete, the treatment plant should proceed into
operation without disruption. After all components in
each of the subsystems have been selected, hydraulic
analysis calculations should be made to determine the
velocities and pressure drops through the system.
Calculations should be run for normal treatment flow and
backwash flow. The latter is more severe, but of short
duration. If pressure losses are excessive, the design
should be modified by decreasing or eliminating losses
(e.g., increase pipe size, eliminate bends or restrictions).
Upon completion of installation, functional checkout
requirements should be accomplished. All piping should
be cleaned and hydrostatically pressure tested prior to
startup. All leaks should be corrected and retested.
Recommended test pressure is 150% of design pres-
sure. Potable water piping and vessels should be disin-
fected prior to startup. Disinfection procedures should be
in compliance with regulatory agency requirements and
material manufacturer's disinfection requirements/limita-
tions. All electrical systems should satisfy a functional
checkout. All instruments should be calibrated; if accu-
racy does not meet requirements stated in Section
3.3.1.3, the instruments are to be replaced.
When the plant operation begins, a check on actual sys-
tem pressure drop is required. If there is a discrepancy
between design and actual pressure drop, the cause
should be determined (obstruction in line, faulty valve,
installation error, design error, etc.) and rectified. Pres-
sure relief valves should be tested; if not accurate, they
should be adjusted or replaced. Although this activity
takes place during treatment plant startup (covered in
Chapter 5.0), it should be incorporated on a construction
document requirement.
3.4.1 Treatment Equipment
Final Design
This section provides a discussion of the details that apply
specifically to arsenic removal water treatment plants.
3.4.1.1 Treatment Bed and Vessel Design
The treatment media volume was designed by determi-
nation of bed dimensions and resulting weight in the
Preliminary Design (see Section 3.3.1.1). It is recom-
mended that a minimum of 10% extra treatment media
be ordered. For lowest price and ease of handling, the
material should be ordered in fiber drums (approximately
5-8 ft3) on pallets. Several sources of granular adsorptive
media are available for service in the application.
Specification requirements should be NSF-certified for
potable water application, mesh size -28, +48 (or as
recommended by media supplier), and demonstrated
arsenic removal capacity.
The vessel design should be simple. The vessel should
have a support system to transfer its loaded weight to
the foundation and ultimately to the soil. The loaded
weight includes the media, the water, attached appurte-
nances (platform, pipe filled with liquid, etc.), the vessel,
and applicable seismic and/or wind loads. The support
legs should be as short as possible, reducing head room
requirements as well as cost. If the equipment is skid-
mounted, the vessel legs should be integral with the skid
to distribute the weight over an area greater than the
dimension of the vessel. This distribution eliminates point
26
-------
loads of vessel support legs, so costly piers, footings,
and excavation requirements are eliminated. The skid
should have provisions for anchorage to the foundation.
Exterior brackets (if uniform and simply detailed) are not
costly and provide supports that eliminate need for cum-
bersome costly field fabrications. Conversely, interior
brackets, though required to anchor (or support) vessel
internal distribution or collection systems, should be held
to a bare minimum because they are costly to line.
Epoxy (or rubber) linings with abrasion resistance qual-
ities are recommended. Vessel interior lining should
extend through vessel opening out to the outside edge of
flange faces. Alternatively, vessels may be constructed
of stainless steel (no lining required). Openings in the
vessels should be limited to the following:
1. Influent pipe - enters vertically at center of top head.
2. Effluent pipe - exits horizontally through vertical
straight side immediately above false flat bottom in
front of vessel, or vertically at the center of the
bottom head.
3. Air/vacuum valve (vent) - mounts vertically on top
head adjacent to influent pipe.
4. Media removal - exits horizontally through vertical
straight side immediately above false flat bottom at
orientation assigned to this function.
5. Manway - 16-inch diameter (minimum) mounted on
top head with center line located within 3 ft of center
of vessel and oriented toward work platform.
Manway cover to be hinged ordavited.
It is recommended that pad flanges be used for pipe
openings in place of nozzles. Pad flanges are flanges
that are integral with the tank wall. The exterior faces are
drilled and tapped for threaded studs. The pad flanges
save the cost of material and labor, and are much easier
to line; they also reduce the dimensional requirements of
the vessel. The vessel also requires lifting lugs suitable
for handling the weight of the empty vessel during installa-
tion. Once installed, the vessel should be shimmed and
leveled. All space between the bottom surface of the
skid structure and the foundation should be sealed with
an expansion-type grout; provisions should be included
to drain the area under the vessel.
The type of vessel internal distribution and collection pip-
ing used in operational arsenic removal plants is defined
in the Preliminary Design (see Section 3.3.1.1). Because
there are many acceptable vessel internal design con-
cepts, configuration details will be left to sound engineer-
ing judgment. The main points to consider in the design
are as follows:
1. Maintain uniform distribution
2. Provide minimum pressure drop through internal
piping (but sufficient to assure uniform distribution)
3. Prevent wall effects and channeling
4. Collect treated water within 2 inches of bottom of
treatment bed
5. Anchor internal piping components to vessel to
prevent any horizontal or vertical movement during
operation
6. Ensure that construction materials are suitable for
pH range of 2.0-13.5 (PVC, stainless steel are
acceptable).
Underdrain failures create significant problems; treat-
ment media loss, service disruption and labor to repair
problems are very costly. A service platform with access
ladder is required for use in loading treatment media into
the vessel. Handrail, toe plate, and other OSHA-required
features should be included.
3.4.1.2 Pipe Design
Each piping subsystem should be reviewed to select each
of the subsystem components (see Figures 3-2, 3-3, and
3-4). Exclusive of the chemical subsystem, five piping
subsystems and two optional subsystems are listed in
the Conceptual Design (see Section 3.2); they are:
1. Raw water influent main
2. Intervessel pipe manifold
3. Treated water effluent main
4. Raw water bypass main
5. Backwash regeneration feed main (optional)
6. Wastewater main
7. Sample panel (optional).
The detail design now proceeds for each of those sub-
systems. First, the equipment specification for each equip-
ment component in each subsystem should be defined.
This is followed by a detailed installation drawing, which
locates each component and provides access for opera-
tion and maintenance. As each subsystem nears comple-
tion, provisions for pipe system support and anchorage,
as well as for thermal expansion/contraction, should be
incorporated in the detail design.
The interface where the concentrated chemical and treat-
ment unit branch piping join is designated as a chemical
injector detail. The chemical injector detail should include
provisions to protect materials of construction from the
heat of dilution of concentrated corrosive chemicals. The
27
-------
key factor is to prevent flow of concentrated chemical
when raw water (dilution water) is not flowing. The dilu-
tion water should dissipate the heat. The actual injection
should take place in the center of the raw water pipe
through an injector that extends from the concentrated
chemical pipe. The injector material should be capable
of withstanding the high heat of dilution that develops
specifically with sulfuric acid and to a lesser degree with
caustic soda. Type 316 stainless steel and Teflon™ are
satisfactory. It also is very important that the concen-
trated chemical be injected upward from below; other-
wise concentrated chemicals with specific gravities
greater than that of water will seep by gravity into the
raw water when flow stops. As described previously, the
chemical pumps are to be de-energized when the well
pump (or other feed pump) is not running.
The treated water pH should be monitored carefully. A
pH sensor installed in the treated water main indicates
the pH at an analyzer. This analyzer should be equipped
with adjustable high and low level pH alarms. The alarms
should be interlocked with the well pump (or other feed
pump) control (magnetic starter), shutting it down when
out-of-tolerance pH excursions occur. A visual and/or
audio alarm should be initiated to notify the operator
regarding the event.
A chemical injector detail similar to that used for acid in
the treatment unit branch piping should be used in the
treated water main to inject caustic in order to raise pH
in the treated water. If aeration for removal of CO2 is
used in place of or in combination with caustic soda
injection for raising treated water pH, then system pres-
sure will be dissipated and the treated water will be
repressurized. If the water utility has ground level stor-
age tanks, the aeration-neutralization concept can be
accomplished without need for a clean/veil and repres-
surization. The aerator can be installed at an elevation
that will permit the neutralized treated water to flow to
storage via gravity.
Easy maintenance is an important feature in all piping
systems. Air bleed valves should be installed at all high
points; drain valves should be installed at all low points.
This assists the plant operator in both filling and draining
pipe systems. Air/vacuum valve and pressure relief valve
discharges are to be piped to drains. This feature satis-
fies both operator safety and housekeeping require-
ments. Bypass piping for flow control, pressure control,
flowmeter, and other in-line mechanical accessories is
optional. Individual equipment item bypass piping is
costly and requires extra space. However, if continuous
treatment plant operation is mandatory, bypass piping
should be included.
3.4.1.3 Instrument Design
Ease of maintenance is very important. Instruments
require periodic calibration and/or maintenance. Without
removal provisions, the task creates process control
problems. Temperature indicators (optional) require ther-
mal wells installed permanently in the pipe. Pressure
indicators require gauge cocks to shut off flow in the
branch to the instrument. pH sensing probes require iso-
lation valves and union type mounting connections
(avoids twisting of signal cables). Supply of pH standard
buffers (4.0, 7.0, and 10.0) should be specified for pH
instrument calibration. A laboratory bench should be
located adjacent to the sample panel. The sample panel
receives flow directly from sample points located in the
process piping. The sample panel consists of a manifold
of PVC or polyethylene tubing with shutoff valves, which
allows the plant operator to draw samples from any point
in the process at the laboratory bench. Laboratory
equipment should be specified to include wall cabinet,
base cabinet with chemical resistant counter top and
integral sink, 115V/14>/60Hz 20-amp duplex receptacle,
laboratory equipment/glassware/reagents for analysis of
pH, arsenic, and other ions. A deionized water capability
for cleaning glassware and dilution of samples should be
included.
3.4.1.4 Acid Storage and Feed Subsystem
Operator safety for work within close proximity of highly
corrosive chemicals takes priority over process func-
tional requirements. Emergency shower and eyewash
must be located within 20 ft of any work area at which
operator exposure to acid or caustic soda exists. Protec-
tive clothing should be specified. Neutralization materials
(e.g., sodium carbonate) should be provided to handle
spills. Potential spill areas must be physically contained.
Containment volumes should be sufficient to completely
retain maximum spillage.
Chemical bulk storage tanks are covered in the Prelimi-
nary Design.
To minimize corrosion of acid pipe material, acid flowrate
is recommended to be less than 0.1 ft/sec. Threaded
pipe and fittings are not recommended; tubing and
Swagelok fittings are recommended. CPVC or Teflon™
are satisfactory except for their vulnerability to damage
from external impact forces. Therefore protective clear
reinforced plastic tubing completely containing the plas-
tic chemical lines is recommended. Positive backflow
prevention should be incorporated in each chemical feed
line. Day tanks should be vented to the atmosphere,
have a valved drain, and have a fill line float valve for
failsafe backup control to prevent overflow. For treat-
ment systems that use HCI instead of H2SO4 for pH
adjustment, it is recommended that references on
28
-------
materials acceptable for the handling and storing of this
acid be consulted.
One acid feed pump is required for influent water pH
adjustment. Acid feed pumps are required to adjust pH
during neutralization following a regeneration, and to
neutralize regeneration wastewater in the wastewater
surge tank. Though preferable to use separate pumps
for each function, it is feasible to accomplish all three
functions with a single pump. The pump should be sized
for a minimum of 110% of the maximum flowrate that it
will provide; it should have a turndown limit no greater
than 50% of the minimum required flow. Acid pump
power should be interlocked with the well pump (or other
feed pump) so that the acid pump is de-energized when
that pump is not running. If the chemical feed pump is
mounted above the day tank, a foot valve is required in
the suction tube. Antisiphon provisions should be included
in the system. Because considerably more acid (approxi-
mately 1 gal/ft3 of activated alumina) is consumed during
the regeneration of an activated alumina bed than during
routine treatment operation, a day tank will need to be
refilled several times during the neutralization phase of
the regeneration. The day tank should be sized for a
minimum of 200% of the daily acid consumption for the
treatment process pH adjustment requirement. The day
tank should be translucent with gallon calibration on the
tank wall. The day tank should be set in an open-top,
acid-resistant containment basin. All relevant regulatory
authorities should be consulted to ensure compliance
with all safety regulations.
3.4.1.5 Caustic Soda Storage and Feed System
The safety requirements stated for acid (Section 3.4.1.4)
also apply to caustic soda. Vinegar should be provided
to neutralize caustic spills.
The day tank and pump design features recommended
for acid systems also apply to caustic. Two caustic
pumps and day tanks are required. The process pH
adjustment pump should be sized to pump 110% of the
maximum process required. The rule of thumb for sizing
the caustic soda regeneration feed pump requires provi-
sions of 2 gal of 50% NaOH/ft3 of activated alumina for
activated alumina systems per hour. Depending upon
the size of the system, a centrifugal pump or an air-
operated diaphragm pump are feed pump options. The
process pH adjustment day tank should be sized for
200% of the maximum daily consumption. The regener-
ation day tank should be the next standard tank size
greater than the requirement for one regeneration. Both
tanks can be set in one containment basin, sized for the
largest tank. The regeneration pump can be calibrated
by means of timing the flow and adjusting as necessary
to arrive at the design flowrate. Carbon steel threaded
pipe or PVC pipe is suitable for the service. All relevant
regulatory authorities should be consulted to ensure
compliance with all safety regulations.
3.4.1.6 Regeneration Waste water Surge Tank
Although treatment and disposal of regeneration waste-
water are not included in this design manual, a surge
tank to receive the wastewater is indicated. The waste-
water surge tank should receive the entire batch of
regeneration wastewater from the start of backwash to
the completion of treatment bed neutralization. To pro-
vide adequate capacity for containment of the entire
batch of regeneration wastewater, this tank should be
sized to contain 400 gal/ft3 for activated alumina sys-
tems. For other adsorptive media for which media regen-
eration is included, the media manufacturer should pro-
vide regeneration process parameters. This tank should
be a ground-level atmospheric carbon steel or PVC tank.
The tank should include a carbon steel floor and roof
and an interior epoxy lining. The tank should include a
reinforced concrete containment structure. The tank
should include fill, chemical feed, drain overflow vent,
multiple discharge, and multiple sample pipe connec-
tions. The tank should include one ground-level manway
and one roof manway (with safety ladder and handrails),
provisions for a liquid level indicator, for an ultrasonic
liquid sludge level sensor, liquid level controller, and a
side entry mixer.
3.4.2 Final Drawings
All of the information required for complete installation of
an arsenic removal water treatment plant should appear
in the final construction drawings and specification pack-
age.
Isometric drawings for clarification of piping subsystems
are recommended; these views clarify the assembly
for the installer (see Figures 3-9 and 3-10). Cross-
referencing drawings, notes, and specifications also is
recommended.
3.4.3 Final Capital Cost Estimate
Similar to the preparation of the preliminary cost esti-
mate, the final cost estimate is prepared based on a take
off of the installed system. The estimate is now based
upon exact detailed information rather than general infor-
mation which was used during the preliminary estimate.
The estimate is presented in the same format (see Table
3-2) and is to be accurate within ±10%. Because finan-
cial commitments are consummated at this stage, this
degree of accuracy is required.
29
-------
A/
\
FREATMENT
TYP,
Figure 3-9. Treatment Vessels Piping Isometric Adsorptive Media With or Without Process Water
pH Adjustment and With Spent Media Replacement
TREATMENT VE55?
TYP,
\
RECEN,
Figure 3-10. Treatment Vessels Piping Isometric Adsorptive Media With Process Water pH Adjustment
and Spent Media Regeneration
30
-------
3.4.4 Final Design Revisions and resubmitted for approval. If communication with the
approving parties has taken place during the course of
Upon their completion, the final construction drawings and the design, then time-consuming resubmittals should not
specifications are submitted for approval to the owner be necessary. Upon receipt of approval, the owner, with
and the regulatory authorities. If changes or additional assistance from the engineer, solicits bids for the con-
requirements are requested, they should be incorporated struction of the arsenic removal water treatment plant.
31
-------
-------
4.0 Central Treatment System Capital Cost
4.1 Introduction
The client should be provided with the least expensive
absorptive media central treatment system that can
remove the excess arsenic from a sufficient quantity of
water that will satisfy all water consumption require-
ments. The economic feasibility evaluation should include
the initial capital cost along with the operating and main-
tenance costs. This chapter covers the capital cost,
which is affected by many factors, including operating
costs.
The water treatment flowrate is the major factor affecting
capital costs, but it is not the only factor. Other factors
which can have varying impact upon the capital cost
include, but are not limited to, the following:
1. pH adjustment process water vs. raw water without
pH adjustment
2. Regeneration or replacement of spent adsorptive
media
3. Backwash and regeneration wastewater disposal
concept
4. Chemical supply logistics
5. Manual vs. automatic operation
6. Raw water arsenic concentration. Other chemical
and physical parameters including but not limited to
pH, alkalinity, iron, manganese, hardness, silica,
sulfate, sodium, and turbidity.
7. Adsorptive media selected for treatment system
8. Climate (temperature, precipitation, wind, etc.)
9. Seismic zone
10. Soil conditions
11. 100-year flood plain
12. Existing and planned (future) potable water system
parameters
(i) Number of wells, location, storage, distribution
(ii) Potable water
(iii) Water storage (amount, elevation, location
(iv) Distribution (location, peak flows, total flow,
pressure, etc.)
(v) Consumption (daily, annual)
13. Financial considerations (cost trends, capital
financing costs, cash flow, labor rates, utility rates,
chemical costs, etc.)
Once the capital cost impacts that each of the above
variables can create have been determined, it becomes
apparent that a cost curve (or capital cost tabulation)
based on flowrate alone is inadequate. Capital cost
curves are presented in Figure 4-1 for activated alumina
media with and without pH adjustment of process water
and with regeneration or replacement of spent media. A
tabulation of the breakdown of these capital costs for this
example is provided in Appendix D. If the impact of these
variables on the cost curves is considered, then a mean-
ingful preliminary project cost estimate (as described in
Section 3.2, Conceptual Design) can be produced.
A user-friendly cost estimating computer program (using
Microsoft Excel Visual Basic) recently has been devel-
oped by Battelle on the use of activated alumina and ion
exchange for arsenic removal (Battelle, 2002). This pro-
gram was funded by the U.S. EPA under Work Assign-
ment 3-20 of Contact No. 68-C7-0008. A copy of the
computer program and the associated document can be
obtained from U.S. EPA National Risk Management Re-
search Laboratory, Water Supply and Water Resources
Division, in Cincinnati, OH, 45268.
4.2 Discussion of Cost Variables
Each of the variables mentioned above has direct impact
upon the total installed cost for a central treatment
33
-------
7.0
6,0
o
3
4,0
2,0
1.0
MANUAL OPERATOR */
»€N? W/0 pH ADJUSTMEN
gpm
Figure 4-1. Capital Cost vs. Flowrate at Typical Locations for Arsenic Removal Water Treatment Plants
by Means of the Activated Alumina Process
system. Ideally, conditions could exist in which a mini-
mum cost system can be designed. Comparable capital
cost curves are provided in Figure 4-1 for treatment
systems in typical locations and in Figure 4-2 for treat-
ment systems in ideal locations. A hypothetical example
of an ideal situation would resemble the following:
1. Raw water quality presents no problem (moderate
temperature, low alkalinity, low concentrations of
competitive ions, etc.)
2. Warm moderate climate (no freezing, no high
temperature, minimal precipitation, no high wind)
3. No seismic requirements
4. Existing concrete pad located on well compacted,
high-bearing capacity soil
5. Single well pumping to subsurface storage reservoir
with capacity for peak consumption day
6. Existing wastewater disposal capability adjacent to
treatment site (e.g., a large tailings pond at an open
pit mine)
7. Acid and caustic stored in large quantities on the site
for other purposes
8. Manual operation by labor that is normally at the site
with sufficient spare time
9. Funding, space, etc. available.
This ideal situation never exists in reality. Occasionally
one or more of the ideal conditions occur, but the fre-
quency is low. If the final estimate for the example used
in Appendix B is revised to incorporate the above ideal
conditions, the cost estimate would be reduced from
$553,000 to $278,000 (see Table 4-1). Conversely, ad-
verse conditions could accumulate, resulting in a cost far
in excess of that for the typical treatment system for the
same treatment capability. The following subsections pro-
vide the basic insight needed to benefit from the above
variables.
34
-------
4.0
3,0
o
o
3
o
o
o
2,0
1.0
50
200
300 400
FLOtRATE, qpm
600
TOO
Figure 4-2. Capital Cost vs. Flowrate at Ideal Locations for Arsenic Removal Water Treatment Plants by
Means of the Activated Alumina Process
4.2.1 Water Chemistry
The water chemistry can affect capital as well as operat-
ing costs. With a clear picture of the raw water quality, its
possible variations, and its adverse characteristics, the
effect upon the capital cost can be determined readily.
High water temperature (greater than 100°F) requires
higher cost piping material and/or pipe support. Varying
water temperature requires inclusion of special provi-
sions for thermal expansion and contraction. Very high
arsenic may require larger treatment units to reduce the
frequency of regeneration. High alkalinity requires higher
acid consumption for pH adjustment resulting in larger
feed pumps, day tank, piping, etc. This might result in an
aeration step for post treatment pH adjustment in place
of caustic addition. High turbidity arsenic, iron, manga-
nese, suspended solids, and/or other contaminants can
require the addition of pretreatment steps to accomplish
removal prior to arsenic removal, or the implementation
of a different arsenic removal treatment method.
Each of the physical and chemical characteristics of the
raw water should be evaluated. The technical as well as
the economical feasibility for the entire project could
hinge on these factors.
4.2.2 Climate
Temperature extremes, precipitation, and high wind will
necessitate a building to house the treatment system
equipment. High temperature along with direct sunlight
adversely affects the strength of plastic piping materials.
Freezing is obviously damaging to piping and in some
extreme cases also to tanks. Temperature variation intro-
duces requirements for special thermal expansion/con-
traction provisions. A building with heating and/or cooling
and adequate insulation will eliminate these problems
and their costs, but will introduce the cost of the building.
The building cost should reflect wind loads as well as ther-
mal and seismic requirements. Operator comfort in place
of economic considerations may dictate building costs.
35
-------
Table 4-1. Final Capital Cost Estimate Example for Ideal Location for Four Types of Adsorptive Media Arsenic
Removal Water Treatment Plants
Location:
Flowrate: 570 gpm
Cost ($1,000)
Date:
Treatment Vessels
Treatment Media
Manual Operation Manual Operation Manual Operation
w/Media w/Media w/Media
Replacement Replacement Regeneration
w/o pH Adjustment w/pH Adjustment w/pH Adjustment
Process Piping, Valves, and Access.
Instruments and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, and
Mechanical
Electrical
Painting and Miscellaneous
Access.
Subtotal
Subtotal
Regeneration Wastewater Surge Tank
Building and Concrete
Site Work, Fence, and Miscellaneous
Contingency 10%
(a) Engineering, exterior utility
Subtotal
Total181
pipe and conduit,
73
31
32
7
N/A
N/A
143
29
5
8
42
N/A
5
0
5
19
209
Process Equipment
73
31
36
11
0
6
157
Process Equipment Installation
40
12
11
63
Miscellaneous Installed Items
N/A
5
0
5
23
248
wastewater and waste solids processing system,
73
31
49
16
0
12
181
43
12
11
66
0
5
0
5
26
278
finance charges,
Automatic Operation
w/Media Regeneration
w/pH Adjustment
73
31
64
66
0
13
247
48
36
11
95
0
5
0
5
35
382
real estate cost and taxes no
included.
The installation costs for the buildings and regeneration
wastewater surge tank along with their associated civil
work becomes a major portion of the overall capital cost.
Care in interpreting the climatological conditions and
their requirements is necessary.
4.2.3 Seismic Zone
Compliance with the seismic design requirements of the
local building codes can impact capital costs. Buildings
and tall slender equipment are vulnerable to seismic
loads. The magnitude of seismic design requirements
should be determined. In zones of extreme seismic
activity, low profile equipment and buildings are recom-
mended.
4.2.4 Soil Conditions
Unless soil-boring data are already available for the
treatment system site, at least one boring in the location
of the foundation for each heavy equipment item (treat-
ment vessels, chemical storage tanks, and regeneration
wastewater surge tank) is required. If the quality of the
soil is questionable (fill, or very poor load-bearing capac-
ity), additional soil borings should be obtained. Poor soil
may require costly excavation/backfill and foundations.
Combinations of poor soil with rock or large boulders can
make foundation work more complex and costly. Rock
and boulders in combination with extreme temperatures
can result in very high installation costs for subsurface
raw, treated, and wastewater pipe mains.
4.2.5 100-Year Flood Plain
For water treatment facilities located within a 100-year
flood plain, the entire site should be relocated to another
site outside of the 100-year flood plain, be elevated 3 ft
above the 100-year flood plain level, or be protected on
all sides by an armored berm that extends a minimum of
3 ft above the 100-year flood plain level.
4.2.6 Existing and Planned (Future)
Water System Parameters
Many existing and planned (future) facility configurations
can either significantly increase or decrease the capital
cost. The most important factors are discussed in this
section.
36
-------
4.2.6.1 Number and Location of Wells
When only one well requires treatment, the removal of
arsenic should be accomplished prior to entering the
distribution system. Theoretically, treatment can occur
before or after entering storage. Practically speaking,
treatment prior to entering storage is much easier to con-
trol because the treatment plant flowrate will be con-
stant. If treatment takes place after storage, or if there is
no storage, flowrate is intermittent and variable, and pH
control is only achievable for a sophisticated automatic
pH control/acid feed system.
When more than one well requires treatment, a decision
is required regarding whether a single treatment plant
treating water from all wells manifolded together or indi-
vidual treatment plants at each well present a more effi-
cient and cost-effective concept. Factors such as distance
between wells, distribution arrangement, system pres-
sure, and variation in water quality should be evaluated
in that decision. If all of the wells are in close proximity
and pump similar quantity and quality water, a single
treatment plant serving the entire system is preferable.
When wells are widely dispersed, manifolding costs
become prohibitively expensive, thus dictating imple-
mentation of individual treatment plants at each well.
Frequently, the distances may be such that the decision
is not clear cut; then other variables such as water
quality, system pressure, distribution configuration, land
availability should be evaluated.
Systems that require multiple treatment plant installa-
tions can achieve cost savings by employing an identical
system at each location. This results in an assembly line
approach to procurement, manufacture, assembly, instal-
lation, and operation. Material cost savings, labor reduc-
tion and engineering for a single configuration will
reduce the cost for the individual plant.
4.2.6.2 Potable Water Storage Facilities
Similar to the wells, the number, size, and location of
storage tanks can affect treatment plant size (flowrate)
and capital cost. If there is no storage capacity in the
system, the well pump should be capable of delivering a
flowrate equal to the system momentary peak consump-
tion; this could be many times the average flowrate for a
peak day. Therefore, if no storage capacity exists, a stor-
age tank should be added with the treatment system for
treatment water storage. Otherwise, automatic pH instru-
ments and controls will be required to pace pH adjust-
ment chemical feedrates to the varying process water
flowrate.
Most systems have existing storage capacity. The stor-
age may be underground reservoirs, ground level stor-
age tanks, or elevated storage tanks (located on high
ground or structurally supported standpipes). The first
two require repressurization; the latter does not. The
elevated storage tanks apply a backpressure on the
ground level treatment system requiring higher pressure
(more costly) construction of treatment vessels and pip-
ing systems. If aeration of treated effluent for pH adjust-
ment is selected with an elevated storage tank, the
treated water should be contained in a clean/veil and re-
pumped to storage. However, the treatment system ves-
sels and piping may be low-pressure construction. When
storage is at or below ground level storage, loss of sys-
tem pressure is not a factor.
The amount of storage capacity also affects treatment
system cost. The larger the storage capacity (within lim-
its), the lower the required treatment plant flowrate (and
resulting cost). A minimum storage capacity of one-half
of system peak day consumption is recommended.
4.2.6.3 Distribution and Consumption
The factors that determine the sizing of the treatment
system are the well (or feed) pump flowrate, the storage
capacity, and the system consumption characteristics.
Those features should be coordinated to provide a
capacity to deliver a peak treated water supply to satisfy
all possible conditions of peak consumption. If there is
adequate storage capacity, the momentary peaks are
dampened out. The peak day then defines the system
capacity. The well (or feed) pump then is sized to deliver
the peak daily requirement. The treatment system in turn
is sized to treat a minimum of what the well (or feed)
pump delivers.
The distribution system may anticipate future growth or
increased consumption. The well (or feed) pump then
either should pump a flow equal to or greater than the
maximum anticipated peak daily flows, or should be able
to adjust to future increased flowrate. The treatment plant
in turn should incorporate capacity to treat the ultimate
peak flowrate or include provisions to increase the treat-
ment capacity in the future.
4.2.7 pH Adjustment of Process Water
Included vs. Not Included
The decision should be made regarding whether or not
to include treatment process pH adjustment by means of
acid (to lower pH) and caustic (to raise pH). The purpose
of including pH adjustment for some adsorptive media
such as activated alumina is to significantly increase the
arsenic removal capacity. A one-time capital cost
increase is required for chemical feed and storage
subsystems as well as constant increased operating cost
for consumable acid and caustic, and addition of sulfate,
sodium, and TDS to the treated water. However, the pH
37
-------
adjustment method may be the most cost-effective
method of removing arsenic from water due to the sig-
nificant increase of treatment cycle life for the treatment
media. This is the key to use of adsorptive media with
treatment process pH adjustment, regardless of whether
the adsorptive media is replaced or regenerated. The
material manufacturer should be consulted for technical
information relating to the process improvement resulting
from the addition of pH adjustment to the treatment pro-
cess. The decision should relate to characteristics of the
adsorptive media and the water analysis for each indi-
vidual application. Pilot studies also can provide infor-
mation to aid in the decision.
4.2.8 Regeneration or Replacement
of Spent Adsorptive Media
Regeneration should not be included for spent adsorp-
tive media unless the treatment process also includes
pH adjustment of process water. Unless the treatment
process already includes chemical feed and storage
subsystems for the treatment process pH adjustment,
adding those subsystems only for media regeneration is
not economically feasible.
Chemical regeneration of adsorptive media that is satu-
rated with arsenic is economically sound for large sys-
tems with high arsenic concentrations. As the size of the
system is decreased, and/or the raw water arsenic con-
centration is decreased, the economic benefit compared
to the capital and operating cost is diminished. For very
small systems, the design may include the use of port-
able tanks that are removed and replaced with new
media. In this situation, the media is likely to be regen-
erated at the vendor's facility and reused.
4.2.9 Backwash and Regeneration
Disposal Concept
Regeneration wastewater and waste solids processing
and disposal is not included in the scope of this docu-
ment. Depending on wastewater discharge limits estab-
lished by the U.S. EPA, state and local regulatory agen-
cies, wastewater disposal is a significant cost item that
should be evaluated in the capital (and operating) cost
projection. Requirements can vary from zero discharge
to discharge into an available existing receiving facility.
Disposal and/or discharge can be accomplished by chem-
ical coprecipitation of arsenic with precipitated aluminum
or ferric hydroxide by adjustment of pH to 6.0-6.5 and
dewatering of precipitated suspended solids. The dewat-
ered solids should pass the U.S. EPA TCLP. The waste-
water, though containing low arsenic concentrations, will
contain elevated levels of TDS, sodium, and sulfate. If
regulatory agency permits disposal by conventional meth-
ods (surface discharge, percolation), the disposal costs
are not large. The total volume of wastewater regenera-
tion generally is 300-400 gal/ft3 of adsorptive media. With
pH adjustment, the activated alumina process can
achieve 10,000 (74,800 gal/ft3) to 25,000 (187,000 gal/ft3)
bed volumes of treated water depending on the arsenic
concentration in the raw water. Therefore, the ratio of
wastewater to treated water is insignificant («1 %).
In the event a zero discharge of wastewater is required,
the wastewater supernatant and filtrate (from solids
dewatering) should be fed back to the head of the treat-
ment plant and very slowly added to the raw water. The
dewatered solids containing nearly all of the arsenic are
then removed for disposal. Although this concept has not
been incorporated in a full-scale treatment plant, it has
been successfully accomplished on a pilot scale by the
author.
4.2. 10 Chemical Supply Logistics
Sulfuric acid (normally 66°B» HjSO^ and caustic soda
(normally 50% NaOH) are commercially available and
are usually the least expensive chemicals to use for pH
adjustment. Other chemicals such as hydrochloric acid
and caustic potash (KOH) are technically acceptable, but
almost always more costly, and therefore are not com-
monly used. The acid and caustic are much cheaper
when purchased in bulk quantities, usually 48,000-lb
tank trucks. In very small plants, the cost of storage
tanks for those volumes is not justified and therefore,
smaller volumes with higher unit prices are procured
(drums and carboys). In very large treatment plants, cost
can be lowered by procuring the chemicals via 200,000-
Ib railroad tank cars. However, this approach requires a
rail siding and rail unloading facility; nevertheless, it does
present an option of lowering the overall cost. A chemi-
cal unloading rail terminal presents another intriguing
option for facilities with multiple treatment plants. In this
approach, smaller site storage tanks are supplied via
"mini tank trucks" relaying chemicals to the treatment
site from the rail terminal. This brings down the size (and
cost) of chemical storage tanks at each site. However,
this could increase the truck traffic of corrosive chemi-
cals through populated areas, a risk that may not be
acceptable.
4.2.11 Manual Versus
Automatic Operation
Automatic operation is technically feasible. However, the
periodic presence of an operator is always required. The
capital cost of automation (computer hardware/software,
valve operators, controls, instrumentation, etc.) as well
as maintenance costs may exceed budget limits that the
client will accept. Therefore, either manual or semiauto-
38
-------
matic operation is normally furnished. The advantages
and disadvantages of manual, automatic and semiauto-
matic operation require careful evaluation prior to deter-
mination of the proper selection.
4.2.12 Financial Considerations
Many financial factors should be considered by the
designer and the client. The client can superimpose
financial restrictions (beyond any of the technical factors
mentioned above) which result in increased (or de-
creased) capital cost. These include, but are not limited
to, the following: inflationary trends, interest rates, financ-
ing costs, land costs (or availability), cash flow, labor
rates, electric utility rates, and chemical costs. All or
some of these factors could affect the capital investment
with reduced operating cost because interest rates are
low, inflation is anticipated, cash is available, and labor
and electric utility rates are high; or the opposite can be
true. The varying combinations of factors that could
develop are numerous; each one will affect the ultimate
capital cost.
4.3 Relative Capital Cost of Arsenic
Removal Central Water Treatment
Plants Based on Flowrate
The relative capital costs of activated alumina central
treatment plants based on the treated water flowrate are
presented in Figures 4-1 and 4-2. Both cost curves are
based on the same treatment system design criteria.
Tabulations of the breakdowns of the capital costs for
both curves are provided in Appendix D. The curve in
Figure 4-1 is based on the facility criteria employed in
the hypothetical design for the 570-gpm treatment arse-
nic system in Appendix B. The curves in Figure 4-2 are
based on the "ideal" facility requirements presented
earlier in this chapter for the same treatment system
(see Table 4-1). This information demonstrates the dra-
matic differences in capital cost that can occur for the
same treatment plant in different circumstances. The
costs related to the curve in Figure 4-1 are representa-
tive of average capital costs. Examples of some of the
equipment, material and labor cost proposal and esti-
mating items employed in Figures 4-1 and 4-2 are
included in Figure 4-3 and Tables 4-2 and 4-3.
39
-------
CODE PRESSURE VESSEL FABRICATOR QUOTATION
FOR ADSORPTIVE MEDIA TREATMENT VESSELS (two required)
Vessel Specification and Quotation #1280m 07/24/01
Customer
Attention
R.F.Q. Pricing for your Arsenic Removal Water
Treatment Project
Description Vertical Skid-Mounted Vessel
Size 120" O.D. x 8'0" S/S; Capy, 5,450 gal
Design Pressure and Temp 50 PSIG @ 175° Fahrenheit
Corrosion Allowance None requested or provided
Design Criteria A.S.M.E. Section VIM, Div. 1
Radiography Spot (RT-3)
Code Stamp Yes and National Board Registration
Constructed of Carbon steel
Supports (4) carbon steel legs with skid to provide 24"
to bottom seam
Nozzles and Appurtenances:
2 20" quick opening manway
1 4" CL150 FF single-tapped pad flange, hillside-type
1 4" CL150 FF single-tapped pad flange
2 8" CL150 FF single-tapped pad flanges
1 False bottom
8 Interior carbon steel lateral support clips
1 Interior carbon steel header support clip
2 sets Exterior pipe support brackets
2 Lifting lugs
1 Uncaged ladder from grade to top head
1 Skid
Valves, gauges, gaskets, or any items not listed above are excluded.
Surface Preparation and Coatings:
Interior surface prep: SSPC-SP-5 White metal sandblast
Interior surface coat: Plasite 4006 (35 MDFT)
Exterior surface prep: SSPC-SP-6 commercial sandblast
Exterior surface primer: Rust inhibitive primer
Exterior topcoat: None requested or provided
Note, interior coating is forced cured to meet NSF 61 requirements for potable water
Shipping: Weight, 9,500 Ib; Dims., 10' diameter x 12.5' OAL.
Price: FOB Madera CA, $ 27,500.00 each, not including taxes.
Price based on a quantity of 2, and is valid for 90 days.
Delivery Schedule: Based upon current schedule.
Drawings for approval: 2 weeks after order.
Fabricate and ship: 12 to 14 weeks after drawing approval.
Terms: Progress payment to be arranged.
Figure 4-3. Code Pressure Vessel Fabricator Quotation for Adsorptive Media Treatment Vessels
40
-------
Table 4-2. Process Pipe, Fittings, Valves, and Static Mixers - Itemized Cost Estimate"
Item
8" Schedule 80 PVC Pipe (P/E)
8" Schedule 80 PVC Coupling (s x s)
8" Schedule 80 PVC Tee (3x3x3)
8" Schedule 80 PVC 90° ELL (s x s)
8" Schedule 80 PVC Van Stone Flange(s)
8" Wafer Style PVC Butterfly Valve with EPDM Seals
8" PVC Wafer Style Check Valve with EPDM Seals
8" PVC In-Line Static Mixers
Totals
Quantity
400ft
8
30
18
66
25
3
4
Material Unit
Price""
($)
8.00/ft
50 ea.
170ea.
120ea.
55 ea
280 ea
650 ea.
1,700ea.
Total
Material
($)
3,200
400
5,100
2,160
3,630
7,000
1,950
6,800
30,240
Labor Unit
Price|c|
($)
5.00/ft
12.50ea.
15.00ea.
12.50ea.
12.50 ea.
50.00 ea.
1 00.00 ea.
1 00.00 ea.
Total
Labor
($)
2,000
100
450
225
825
1,250
300
400
5,550
Total
($)
5,200
500
5,550
2,385
4,455
8,250
2,250
7,200
35,790|d|
(a) Manually operated 570 gpm arsenic removal water treatment system with treatment process pH adjustment and spent media regeneration.
(b) Prices effective August 2001 (markup included).
(c) Labor rate @ $50/hr.
(d) Tools, installation equipment, pipe supports, accessories, bolts, nuts, gaskets, mobilization, material storage, etc. not included.
41
-------
Table 4-3. Chemical Feed Pumps, and Static Mixers - Itemized Cost Estimate"
Item
Material
Unit Price""
Quantity ($)
Total
Material
($)
Labor Unit
Price|c|
($)
Total
Labor
($)
Total
($)
IV)
Acid feed pumps for 66°B- H2SO4 for adjustment of raw water pH for potable water treatment.
Chemical metering pump will be positive displacement. A bleed valve will be provided for the manual
evacuation of entrapped vapors and safe relief of pressure in the discharge line. Flowrate 0-2.5 gph.
Turndown 1,000:1. Pressure: 50 psig (max). Suction lift: 6-0-(min.) for acid. Temperature 70°F-
90°F. Materials of construction: PVDF pump head, housing, suction tubing, discharge tubing and
bleed valve, Teflon8-faced Hypalon8 diaphragm, Teflon8 seats and o-rings, ceramic ball checks.
Includes: injector, foot valve, suction and discharge tubing. Connections: D-inch I.D. tubing.
Acid feed pumps for 66°B- H2SO4 for raw water pH adjustment for neutralization of regenerated
treatment media for pH adjustment of regeneration wastewater. Chemical feed pump to be air-
operated diaphragm type. Size: %-inch self-priming. Pump to include compressed air supply
filter/regulator. Flowrate 1-4 gpm. Suction lift: 6-0-(min.) for sulfuric acid. Discharge pressure:
50 psig (max.) Temperature 70°F-90°F. Air pressure: 100 psi (max.) Materials of construction:
Kyner body. Teflon™ diaphragms and check valves. Connections: Sulfuric acid - %-inch NPT,
Compressed air-%-inch NPT. Self-lubricating.
Caustic soda feed pumps for 50% NaOH for adjustment of treated water pH. Chemical metering
pump will be positive displacement diaphragm type pump. A bleed valve will be provided for the
manual evacuation of entrapped air or vapors and safe relief of pressure in the discharge line. Pump
control will be manual. The electronic circuitry will be EMI-resistant and will employ a metal oxide
varistor for lightning protection. Flowrate 0-5 gph. Turndown 1,000:1. Pressure: 100 psig (max).
Suction lift: 6-0- (min.) for caustic soda. Temperature 70°F-90°F. Materials of construction: Glass-
filled polypropylene pump head, housing, and bleed valve, Teflon8-faced Hypalon8 diaphragm,
Teflon8 seats and o-rings, ceramic ball checks. Includes: injector, foot valve, suction and discharge
tubing. Connections: --inch I.D. tubing.
Caustic soda feed pump for 50% NaOH for raising feedwater pH for regeneration of treatment media.
Chemical feed pump to be air-operated diaphragm type. Size: %-inch self priming. Pump to include
compressed air supply filter/regulator. Flowrate 4-7 gpm. Suction lift: 6-0- (min.) for caustic soda.
Discharge pressure: 50 psig (max.) Temperature 70°F-90°F. Air pressure: 100 psi (max.) Materials
of construction: Polypropylene body. Teflon8 diaphragms and check valves. Connections: Caustic-
%-inch NPT, Compressed air- %-inch NPT. Self-lubricating.
Totals
1
900 ea.
1,100
400 ea.
400
1,500
650 ea.
1,300
400 ea.
800 2,100
1,100ea.
1,100
400 ea.
400
1,500
750 ea.
750
3,250
400 ea.
400
1,150
2,000 5,250
(a) Manually operated 570 gpm arsenic removal water treatment system with treatment process pH adjustment and spent media regeneration.
(b) Prices effective August 2001 (markup included).
(c) Labor rate @ $50/hr.
-------
5.0 Treatment Plant Operation
5.1 Introduction
Upon completion and approval of the final design package
(plans and specifications), the owner (client) proceeds to
advertise for bids for construction of the treatment plant.
The construction contract normally is awarded to the firm
submitting the lowest bid. Occasionally, circumstances
arise that disqualify the low bidder, in which case the
lowest qualified bidder is awarded the contract. Upon
award of the construction contract, the engineer may be
requested to supervise the work of the construction con-
tractor. This responsibility may be limited to periodic vis-
its to the site to assure the client that the general intent of
the design is being fulfilled; or it may include day-to-day
inspection and approval of the work as it is being per-
formed. The engineer should review and approve all
shop drawings and other information submitted by the
contractor and/or subcontractors and material suppliers.
All acceptable substitutions should be approved in writ-
ing by the engineer. Upon completion of the construction
phase of the project, the engineer normally is requested
to perform a final inspection. This entails a formal
approval indicating to the owner that all installed items
are in compliance with the requirements of the design.
Any corrective work required at that time is covered by a
punch list and/or warranty. The warranty period (normally
one year) commences upon final acceptance of the proj-
ect by the owner from the contractor. Final acceptance
usually takes place upon completion of all major punch
list items.
Preparation for treatment plant startup, startup and oper-
ator training may or may not be included in the construc-
tion contract. Although this area of contract responsibility
is not germane to this manual, the activities and events
that lead up to routine operation are. This chapter dis-
cusses those steps in the sense that the operator is per-
forming them. The operator could be the contractor, the
owner's representative, or an independent third party.
System operating supplies, including treatment chemi-
cals, laboratory supplies, and recommended spare parts
should be procured, and stored on site. The treatment
plant operating and maintenance instructions (O&M
Manual) should be available at the project site. Included
in the O&M Manual are the valve number diagram which
corresponds to brass tags on the valves (see Figure 5-
1), a valve directory furnished by the contractor, and a
valve operation chart (see Table 5-1).
The filter vessel and piping should be disinfected in
accordance with American Water Works Association
(AWWA) standard procedures. The treatment bed
material then is placed in the treatment vessels and the
plant is ready to start operation.
For systems that regenerate spent adsorptive media,
there are four basic modes of operation: treatment,
backwash, regeneration, and neutralization. Operating
details for each of these modes are covered in this chap-
ter. It is important to note that each of the above modes
uses raw water during each operation.
For systems that replace spent adsorptive media, there
are two basic modes of operation: treatment, and
replacement of spent media. The latter mode consists of
removal of spent media, and placement and conditioning
of new media (per initial startup as described in Section
5.2). The removal of spent adsorptive media can be
accomplished by various methods. Because the spent
adsorptive media is already wet, the simplest method is
accomplished by flushing the adsorptive media in a water
slurry out of the treatment vessel, through a valved
media removal nozzle located in the side of the vessel
immediately above the false flat bottom, and into a con-
tainment vessel. The containment vessel should be port-
able for transport to a disposal site. The containment
vessel also should incorporate screened drains to permit
transfer water to drain from the spent media in the
containment vessel. The containment vessel should be
capable of holding 150% of the volume of the spent
adsorptive media. Spent media removal from the treat-
ment vessel also can be accomplished by manual means,
vacuum equipment, and other pneumatic transfer sys-
tems. Examples of several other removal methods are
given in Appendix E.
43
-------
TO DISTRIBUTION
LEGEND
f€^|
X HI 4 10 *
•
II ™
TO
Figure 5-1. Valve Number Diagram
MY TANK
CONTROL VALVf
CHECK VALVE
pH SENSOR/ANALYZER
FLOW INDICATOR
FLOW TOTALIZER
RELIEF VALVE
AIR/VACUUM VALVE
SAMPLE
IN LINE STATIC
FEED PUMP
44
-------
Table 5-1. Valve Operation Chart for Treatment Vessels in Spent Adsorptive Media Regeneration
Operational Modes'3'
Mode
Treatment - lead position
1
•
2
X
3
•
Valve No.
4 5
X
X
Regeneration
Chemicals
6
X
7
X
8
X
Caustic
X
Acid
X
Regeneration
Drain
Backwash
Drain
Upflow regeneration
Upflow rinse
Drain
Downflow regeneration
Downflow rinse
Downflow neutralization pH 2.5
Downflow neutralization pH 4.0
Downflow neutralization pH 5.5
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
.
X
•
•
X
X
X
X
X
X
X
.
X
•
•
X
X
X
X
X
X
X
X
X
X
X
X
.
X
.
X
X
X
X
•
X
X
•
X
X
X
X
X
X
X
X
X
X
X
X
.
•
•
Treatment
Treatment - lag position
Treatment regeneration other vessel
Treatment - lead position
X
•
•
•
X
X
X
X
•
•
•
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
(a) Refer to Figure 5-1 for valve location.
Legend: x = valve closed; • = valve open.
Vessel A
Valve No.
A1 Feedwater
A2 Feed from Vessel B
A3 Treated to Vessel B
A4 Treated water (to distribution)
A5 Regeneration upflow feed
A6 Regeneration upflow waste
A7 Regeneration downflow feed
A8 Regeneration downflow waste
Vessel B
Valve No.
B1 Feedwater
B2 Feed from Vessel A
B3 Treated to Vessel A
B4 Treated water (to distribution)
B5 Regeneration upflow feed
B6 Regeneration upflow waste
B7 Regeneration downflow feed
B8 Regeneration downflow waste
5.2 Adsorptive Media Initial Startup
The operator should thoroughly review the O&M Manual,
become familiarized with every component of the plant,
and resolve any questions that arise.
The placement of the adsorption media in the treatment
vessel, which takes place immediately prior to initial
startup or during replacement of spent media, is a critical
step in the future system performance. The dry material
usually is delivered in drums or sacks. The volume of the
media is determined on a dry weight basis. The actual
density varies with the degree of packing of the bed.
Unless instructed otherwise by the manufacturer, 45 Ib/ft3
is a suggested media density for use in weight calcula-
tions for activated alumina. For media density of other
adsorptive media, consult the manufacturer. The virgin
granular activated alumina material is "coated" with
caustic. A small amount of fines can become airborne
and are irritating to the personnel who are handling
them. Eye, skin, and inhalation protection are recom-
mended during vessel loading activity.
The vessel should be half-filled with water prior to plac-
ing the alumina through a manway in the top head of the
vessel. As activated alumina is carefully distributed into
the vessel from the top, heat is generated by the wetting
of the caustic "coating" on the alumina grains. The water
in the tank dissipates this heat, thereby preventing
cementing of the bed. The water also separates the fines
from the granular materials, protects the underdrain
assembly from impact, and initiates stratification of the
bed. It is recommended that the bed be placed in two or
three lifts. In the two-bed treatment system, alternate
placing of media and backwashing steps can be worked
together between the two treatment units. Thereby,
media placement can be a continuous operation. The
bed should be thoroughly backwashed with raw water
after each lift. The backwash rate should be adjusted to
provide 50% bed expansion. For activated alumina, this
is typically 7 gpm/ft2 except for extremely warm or cold
water for which flowrates may have to be adjusted up or
down respectively. During bed placement, each back-
wash step should be a minimum of 30 min and, depend-
ing on the quantity of fines in the media, could extend to
2 hr. The purpose of this stringent effort is to remove all
of the fines from the bed. If the fines remain in the bed,
45
-------
potential problems can develop such as channeling,
excessive pressure drop, or even cementing. The extra
backwashing effort during bed placement permits fines
at the bottom of the bed to work their way up and out to
waste. Because the lower portions of the bed (which
contain the largest particles) do not expand during
backwash, fines not backwashed out of the bed at that
stage may be permanently locked into the bed. The
backwash water should be directed to waste.
5.3 Treatment Process with Spent
Treatment Media Regeneration
Upon completion of backwashing of a virgin bed, the bed
should be drained and the vessel opened. Approximately
•- to %-inch of fine bed material should be skimmed from
the top of the bed. The finest grain material tends to
blind the bed, causing channeling and/or excessive pres-
sure drop. Once that material is removed, the vessel can
be closed and refilled with water.
At this point the plant should be cleaned up. Airborne
fines that form a dust-like coating on piping and equip-
ment should be removed. Good housekeeping should
begin immediately and be continued on a permanent
basis.
The pressure loss checkout mentioned in Section 3.4,
Final Design, should be accomplished at this point, just
prior to startup. See Table 5-2 for calculated pressure
drop through activated alumina treatment media. If there
is a pressure loss problem, it should be corrected prior to
treatment startup. For other adsorptive media, consult
the manufacturer for information on pressure drop.
Table 5-2. Calculated Activated Alumina (-28, +48
Mesh) Downflow Pressure Drop Data
Water Flowrate
(gpm/ft2)
2.0
3.0
4.0
5.0
6.0
7.0
Pressure Drop
in psi per Foot of
Bed Depth
0.009
0.018
0.028
0.040
0.053
0.068
Modified Reynolds
Number
2375
3555
4735
5900
7111
8291
5.3.1 Treatment Mode
Prior to start of operation, the pH instrumentation should
be calibrated. The most critical requirement for efficient
low-cost operation is the control of the raw water
adjusted pH. For activated alumina, the optimum condi-
tion for maximum arsenic removal exists when the
treatment pH is in the range of 5.0-6.0. The best results
have occurred when the pH is held rigidly at 5.5 (Rubel,
1984, 1981). Because acid feedrates are a function of
raw water alkalinity, they vary from one water to another.
As raw water pH moves above 6.0 or below 5.0, arsenic
removal capacity deteriorates at an increasing rate.
However, when the alkalinity of the raw water is
extremely high and/or the cost of acid is very high, it can
be more cost-effective to operate in a pH range of 6.0-
6.5 in order to reduce the acid consumption (even
though arsenic removal efficiency is also reduced). For
other adsorptive media, consult the manufacturer for
information regarding treatment process pH adjustment
requirements.
The downflow treatment for the first (virgin) run can now
begin. See the valve operation chart (Table 5-1) for
valve positions for this function. It is recommended that
one vessel be placed in operation at a time. This allows
the operator to concentrate on initial raw water pH
adjustment on one treatment unit until it is in stable oper-
ation; the operator then can devote full concentration to
the second treatment unit.
The basic flow schematic for the treatment mode
illustrated in Figure 5-2.
is
With activated alumina, the initial effluent pH is high with
no arsenic removal (similar to the neutralization mode
explained later). After a short period, both pH and arse-
nic in the treated water drop to anticipated levels. At that
time, the treated water can be directed to storage and/or
distribution. The first treatment unit will be returned to
operation in the lead position after the pH of the second
treatment unit has also been stabilized at pH 5.5. De-
pending on the requirements of the state or local regu-
latory agency, samples may have to be analyzed at a
certified testing laboratory prior to approval of distribution
of treated water.
In the series process utilizing two treatment vessels, the
entire arsenic removal process takes place in a treat-
ment band that initially is contained in the lead vessel.
The arsenic ions are completely removed within the
treatment band. After an extended treatment period, the
adsorptive media at the top of the treatment band
becomes saturated. The treatment band then begins to
migrate downward slowly through the treatment bed until
arsenic starts to break through. Breakthrough is defined
as the first detectable amount of arsenic appearing in the
effluent from the lead column. Although the detectable
level will vary depending on the analytical method used
to measure the arsenic, it would likely be near 3 ug/L. An
example of a breakthrough curve of the lead column is
shown in Figure 5-3. As breakthrough occurs, there is a
long period of slowly increasing arsenic concentration
the treated water. The treatment band then enters the
treatment media in the lag column where treatment
46
-------
TREATMENT FLCW - UNIT "A" IN LEAD POSITION
TREMNENr
¥E5SB,
B
IlKTMEiT
sissa
TO W«IF**TER SU80E TANK a^iwi
UNIT "A" BACKWASH. UPFLOW REGENERATION & UPFLOW RINSE
SOURCE
ffy W* {<=Ea» X^
• X WATER '
^W!KW f 1 ="~°
10 SURGE 9W< Wlr ""<
UNIT "A" DOWNFLOW & NEUTRALIZATION
- UNIT "B" IN
Figure 5-2. Basic Operating Mode Flow Schematics
47
-------
—X— After tank B1
- After tank A1
-Effluent
•Influent
3.1 3.3
Bed Volumes of Water Treated—x1000 BV
3.6 3.8 4.1 4.3 4.6 4.9 5.1 5.4
5.6 5.9
120.0
100.0 -
< 20.0 -
0.0
Activated alumina replaced and tanks
repositioned on May 25, 1999
~ ""lO^pgTTATseTTld V^v. v^^v^ — "
-*-
09/30/98 11/11/98 01/06/99 02/17/99 03/31/99 05/12/99 06/23/99 08/04/99
Sampling Date
Figure 5-3. Typical Breakthrough Curve for Arsenic. The vertical dash line represents the date of the
media replacement. The horizontal dash line represents 10-ug/L arsenic level. Tank B1
was used as a roughing filter and tank A1 as a polishing filter before the media replacement
on May 25, 1999; after that, tank A1 was used as a roughing filter and the recharged tank
B1 as a polishing filter (Source: Wang et al., 2002).
removes the remaining arsenic. As treatment progres-
ses, the treatment band progresses downward through
the lead column until the media in the column is com-
pletely saturated. At that point, the arsenic concentration
in the raw water entering and the treated water leaving
the lead column are the same. The treatment band is
entirely contained in the lag column. The lead column
then can be removed from the treatment train to provide
regeneration of the treatment media. For systems that
do not regenerate the treatment media, the spent treat-
ment media in the lead column should be replaced with
new (virgin) treatment media. Whether the media is
regenerated or replaced, the arsenic removal treatment
capacity is restored for a follow on treatment cycle in the
treatment vessel. The treatment vessel with fresh adsorp-
tive media is returned for treatment service in the lag
position. The treatment vessel that was formally in the
lag position is placed in the lead position.
Concurrently, in the vessel that has completed the regen-
eration process, the treated water pH gradually drops to
the adjusted raw water pH level where it remains through
the duration of the run. Because the pH of the treated
water is lower than the normally accepted minimum pH
of 6.5, it should be raised either by chemical addition,
aeration, and/or blending with raw water. Regardless of
the method of pH adjustment, it should take place and
be stabilized at the desired level prior to delivering the
treated water into distribution.
High pH in the treated water is also a concern. Normally
the maximum allowable pH is 8.5; however, there are
exceptions where pH 9.0 may be permitted. Most sys-
tems desire pH in the 7.5-8.0 range. When the treated
water is approved and the pH stabilized for distribution, it
flows out of the plant past a failsafe pH sensor with high
and low level alarms. If there is a pH excursion exceed-
ing the allowable limits, an interlock (incorporating the
pH alarms with the well pump(s) magnetic starter) de-
energizes the well pump(s). Simultaneously, the chemi-
cal pumps shut down as their controls are interlocked with
the well pump(s) power circuitry. The failsafe pH override
automatically prevents any treated water for which pH is
out of tolerance from entering the distribution system. In
the event of such an excursion, the operator manually
controls the well pump(s) to divert the unacceptable
water to waste, determine the cause of the deviation,
and make corrections prior to placing the treatment sys-
tem back on line. Probable causes for treated water pH
deviations are: change in water flowrate, change in acid
48
-------
flowrate, change in caustic flow/rate, and change in raw
water chemistry.
As breakthrough occurs in the lead column, there is a
long period of slowly increasing arsenic concentration in
the lead column effluent. This period increases the arse-
nic loading on the media of the lag column and results in
lower operating costs. It should be noted that the higher
the raw water arsenic level, the greater the adsorption
(driving force) capacity. Because many other factors can
affect this capacity, the precise amount is difficult to
predict. The operator should be cognizant of the fact that
the more water treated during a run, the lower the oper-
ating cost.
In raw waters where the arsenic level is very low, part of
the raw water can bypass treatment and be blended
back with the treated water. A skilled operator may be
able to develop many techniques such as this to mini-
mize operating costs.
High iron content in raw water can cause problems dur-
ing a treatment run. The iron oxidizes, precipitates, and
is filtered from solution by the adsorptive media. This
results in increased pressure drop, and shortened treat-
ment runs. Raw water iron content greater than 0.3 mg/L
is cause for concern. However, if the iron concentration
is above 0.3 mg/L, the secondary MCL, an iron removal
process should be considered as the treatment process
for arsenic removal in place of the adsorptive media
process because of the capability of the process to
remove arsenic.
5.3.2 Backwash Mode
It is important that the bed be backwashed with raw
water after each treatment run prior to regeneration for
two reasons. First, any suspended solids that have been
filtered from the raw water by the treatment bed tend to
blind the bed. Therefore, these particles should be
removed from the bed. Second, even though filtration
may have been negligible, the downward flow tends to
pack the bed. An upflow backwash will expand the bed,
and break up any tendency towards wall effects and
channeling. A backwash rate of 7 gpm/ft2 will expand the
-28, +48 mesh activated alumina bed approximately 50%,
which is recommended. For other adsorptive media,
backwash flowrate requirements should be provided by
the manufacturer. As mentioned in prior chapters of this
manual, the backwash rate may vary with grain size,
material density, and water temperature. Care must be
taken to avoid backwashing granular bed material out of
the treatment unit. Normally backwashing lasts 10 min or
until all suspended solids are removed from the treat-
ment media.
Refer to Table 5-1 for valve positions for the backwash
mode. The basic flow schematic for the backwash mode
is illustrated in Figure 5-2. For most effective backwash,
it is recommended that the vessel be drained prior to
backwash. As backwash water flows into a drained bed,
it lifts the entire bed approximately 1 ft prior to the bed
fluidizing. This action provides an efficient scouring action
without excessive abrasion to the adsorptive media
grains. Backwash water samples should be inspected
frequently to determine that filtered material is still being
removed and treatment media is not being washed out
of the bed. Excessive backwash causes abrasion that
wears down the adsorptive media grains, and also wastes
raw water and increases the wastewater disposal vol-
ume. Therefore, backwash volume should be minimized.
It is prudent to periodically inspect the media level of
each treatment bed to determine whether bed volume
has changed.
5.3.3 Regeneration Mode
The most efficient, cost-effective method of regenerating
an activated alumina treatment bed upon completion of a
treatment run includes two discrete regeneration steps.
The first step is upflow following draining of the bed after
the backwash mode. The upflow regeneration is followed
by an upflow rinse. The unit is then drained to the top of
the treatment bed prior to the second regeneration step
(which is downflow). Both steps use a 5% (by weight)
NaOH solution. For regeneration procedures for other
adsorptive media, consult the manufacturer.
The object of regeneration is to remove all arsenic ions
from the media before any part of the media is returned
to the treatment mode. Arsenic ions lose their attraction
(adsorptive force) and become repelled by the alumina
when the pH rises above 10.5. The higher the pH, the
faster and more efficient the regeneration. However, too
high a pH not only costs more (because of higher caustic
for regeneration and acid for neutralization consump-
tion), but is also increasingly aggressive to the alumina.
The 5% NaOH solution is the maximum concentration
required for high efficiency regeneration (recovery of
total arsenic capacity). A skilled operator might be able
to reduce the concentration of the NaOH to 4% with the
same high efficiency performance. However, below 4%,
efficiency deteriorates rapidly. This lower caustic con-
centration can reduce caustic consumption for regenera-
tion up to 20%. As described in Chapter 3.0, the dilution
of the caustic takes place at an injector in the regenera-
tion water piping. Both the raw water and the 50% NaOH
are metered prior to injection into the regeneration main.
The accuracy of the metering ranges from ±2% to ±5%
depending on the type of flow instrumentation.
49
-------
The rule of thumb for the volume of 5% caustic solution
required per activated alumina regeneration step is
15 gal/ft3 of treatment media. Because there are two
regeneration steps (upflow and downflow), the actual
regeneration time exclusive of draining, flushing and
neutralization is 2 hr. The minimum time recommended
per step for the solution to flow through the bed is
60 min. The maximum time of 90 min for each step is
recommended. For a 5-ft-deep treatment bed, a flow of
1.25 gpm/ft2 for a period of 60 min for each regeneration
step is sufficient. This equates to 1 gal 50% NaOH per
cubic foot of treatment media for each regeneration step
(upflow and downflow).
For the valve position during each step of the regenera-
tion mode, refer to Table 5-1. The basic flow schematics
for the regeneration modes are illustrated in Figure 5-2.
After backwash, prior to the upflow regeneration step,
the bed will be drained to remove water, which dilutes
the caustic concentration. Upon completion of the upflow
regeneration, the caustic feed pump is turned off and the
caustic soda day tank refilled. The raw water continues
to flow for 60 min at 2.5 gpm/ft2 flowrate upward through
the bed, flushing out the arsenic. After this rinse step is
completed, the vessel is drained to the top of the treat-
ment bed, again to remove dilution water. The downflow
regeneration then takes place for 60 min. The downflow
regeneration is followed by draining fluid down to the top
of the bed prior to the start of the neutralization mode.
5.3.4 Neutralization Mode
The neutralization mode is critical to the success of the
following treatment run. The object of this mode is to
return the bed to the treatment mode as rapidly as possi-
ble without dissolving the activated alumina. The pH of
the treatment media after completion of the regeneration
is 13+. It should be adjusted down to pH 5.5, and there-
fore will pass through pH ranges where ions that com-
pete for absorption sites on the alumina will be adsorbed
onto the bed. The minimum pH that can be safely
exposed to the granular activated alumina is 2.5. A pH
lower than that is too aggressive and is not recom-
mended. For neutralization procedures for other adsorp-
tive media, consult the media manufacturer.
At the start of the downflow neutralization mode, the
valves are positioned according to Table 5-1, and the
flow is adjusted to the normal treatment mode rate. The
basic flow schematic for the neutralization mode is illus-
trated in Figure 5-2. After 15 min the acid pump is
started, and the pH of the raw water is adjusted to 2.5.
Acid feedrate again varies with the alkalinity of the raw
water. The raw water flowrate may have to be reduced
to achieve pH 2.5 at the maximum acid pump feedrate.
As the neutralization mode proceeds, the pH of the
treated water gradually drops below 13. The rate of pH
reduction increases at an increasing rate. As the treated
water pH drops below 10, the treated water arsenic level
begins to drop below that of the raw water. At the point
where the arsenic level drops below the MCL, the water
becomes usable and can be directed to storage. When
the treated water pH drops to 8.0, the raw water pH is
adjusted up to 4.0 as the bed rapidly neutralizes. When
the treated water pH drops to 6.5, the raw water pH is
adjusted up to 5.5 where it remains through the duration
of the treatment cycle. The regenerated treatment unit
now starts the next cycle in the treatment mode. Prior to
placement of the regenerated treatment unit into service
in the lag position, the operator should open the manway
in the top head of the vessel to check the level of the
treatment media. Approximately 5% of the activated alu-
mina will be dissolved during regeneration. The operator
should replace the lost activated alumina by adding an
equal amount to bring the bed back to the original level.
The operator should backwash the bed with water
adjusted to pH 5.5 for 30 min. The regenerated treat-
ment unit will then be placed into service in the lag posi-
tion. It remains there until the treatment vessel in the
lead position is removed for regeneration.
A summary of the regeneration process for the activated
alumina process is shown in Table 5-3. For similar infor-
Table 5-3. Typical Process Conditions for Regeneration of an Activated Alumina Treatment System"
Step
No.
1
2
3
4
5
6
7
Step
Backwash
Regeneration
Rinse
Regeneration
Neutralization
Neutralization
Neutralization
Liquid
Raw water
5% NaOH
Raw water
5% NaOH
Raw Water adjusted to pH 2.5
Raw water adjusted to 4.0
Raw water adjusted to 5.5
Flow
Direction
Upflow
Upflow
Upflow
Downflow
Downflow
Downflow
Downflow
Rate
(gpm/ft2)
7
1.2
2.5
1.2
Varies
Varies
Varies
Time
(minutes)
10
60-90
60
60-90
Time to achieve pH of 8.0
Time to achieve pH of 6.5
Time to achieve pH of 5.5
Total
Wastewater
(gai)
30
15
30
15
240
330
(a) Consult manufacturer for similar information on other adsorption media.
50
-------
mation on other adsorptive media with regeneration capa-
bility, the manufacturer should be contacted.
The volume of wastewater produced during the regener-
ation of a treatment bed will vary with the physical/chem-
ical characteristics of the raw water. A rule of thumb that
can assist the operator in his logistical handling is that
300-400 gal of wastewater is produced per cubic foot of
activated alumina during each regeneration. Typical vol-
umes of wastewater generated per cubic foot of activated
alumina during each regeneration step for a hypothetical
treatment bed are shown in Table 5-3.
Operational experience at a specific treatment plant will
present deviations from these quantities.
5.4 Treatment Process with Spent
Treatment Media Replacement
Treatment systems that are designed to replace spent
adsorptive media undergo the same initial startup proce-
dure as those that are designed for regeneration of
spent media. For those procedures, see Section 5.2.
5.4.1 Treatment Mode
The treatment mode for systems that replace spent
adsorptive media is identical to that described in Section
5.3.1 for systems that employ treatment process pH
adjustment with the exception that it is also applicable to
systems that do not employ treatment process pH
adjustment. For those systems that do not employ pH
adjustment, the treatment mode merely deletes all refer-
ence to pH adjustment from the treatment process. The
duration of treatment cycles for the systems without
treatment process pH adjustment is greatly reduced
(Rubel, 1984) depending upon the adsorptive media.
The relative performance is a function of the adsorptive
media, and the raw water chemistry for each individual
water treatment system. High concentrations of ions
including but not limited to silica, alkalinity, hardness,
fluoride, and sulfate as well as high pH may adversely
affect the adsorptive media arsenic capacity as well as
the percent removal of arsenic.
5.4.2 Media Replacement Mode
The media replacement mode includes removal of spent
media for disposal and replacement with fresh (virgin)
adsorptive media for the next treatment cycle. Several
methods are available for spent media from treatment
vessels. The method used will vary with the size of the
treatment vessel. Typical removal methods are discussed
in Section 5.1 and Appendix E. Installation of replace-
ment adsorptive media should repeat the procedures
described in Section 5.2.
5.5 Operator Requirements
A qualified operator for an arsenic removal water treat-
ment plant should have thorough arsenic removal pro-
cess training, preferably at an existing treatment plant.
The operator should be able to service pumps, piping
systems, instrumentation, and electrical accessories.
The operator should be fully informed about the safety
requirements and physical/chemical characteristics of
both acid and caustic in all concentrations. Corrosive
chemical safety requirements as to clothing, equipment,
antidotes, and procedures must be thoroughly under-
stood. The operator should be thoroughly trained to run
routine water analyses including the method for deter-
mining arsenic levels. The operator should be well
grounded in mathematics for operation cost accounting
and treatment run recordkeeping. The operator, above
all, should be dependable and conscientious.
5.6 Laboratory Requirements
In addition to the O&M Manual, the treatment plant should
have the latest edition of Standard Methods for the
Examination of Water and Wastewater prepared jointly
by the American Public Health Association-American
Water Works Association-Water Environment Federa-
tion (APHA-AWWA-WEF, 1995). This manual supplies
the plant operator with necessary information for accept-
able methods for analyzing water. A recommended list of
items for analysis is illustrated in Figure 3-1. The primary
requirement is accurate analysis for arsenic and deter-
mination of pH. As long as pH meters are calibrated and
cleaned regularly, high precision measurements are
easily obtained. Care should be exercised to prevent
contamination of pH buffers.
Total arsenic can be effectively preserved in field sam-
ples and analyzed by several analytical methods down
to the MCL of 10 ug/L or less. Preservation of total
arsenic is accomplished by acidifying the sample to pH
<2. The Arsenic Rule lists four U.S. EPA approved ana-
lytical methods: inductively coupled plasma-mass spec-
troscopy (ICP-MS), graphite furnace atomic absorption
(GFAA), stabilized temperature platform (STP) GFAA,
and gaseous hydride atomic absorption (GHAA). These
methods are U.S. EPA-approved for compliance require-
ments and require expensive analytical equipment that is
found only at extremely large water treatment plants.
During the past several years, several companies have
developed portable test kits for field analysis of arsenic.
Several arsenic tests kits have been evaluated under the
U.S. EPA Environmental Technology Verification (ETV)
program by the Advanced Monitoring Systems Center
managed by Battelle in partnership with U.S. EPA.
These kits were tested for monitoring arsenic in the 1 to
100 ug/L range. Information on the test kits can be found
51
-------
on the Internet (http://epa.gov/etv/verifications/vcenter1-
21.html). Although these test kits may be adequate for
monitoring process performance, they are not U.S. EPA-
approved methods for use in reporting MCL compliance
data. For regulatory data, water samples must be ana-
lyzed by U.S. EPA/state-certified testing laboratories
employing U.S. EPA-approved methods.
5.7 Operating Records
A system of records should be maintained on file at the
treatment plant covering plant activity, plant procedures,
raw water chemical analyses, plant expenditures, and
inventory of materials (spare parts, tools, etc.). The plant
operator should have the responsibility of managing all
aspects of the treatment plant operation. The operator is
accountable to the water system management. The
recommended record system should include, but not be
limited to, the items described in the following sub-
sections.
5.7.1 Plant Log
A daily log should be maintained in which the plant oper-
ator records daily activities at the plant. This record should
include a listing of scheduled maintenance, unscheduled
maintenance, plant visitors, purchases, abnormal weather
conditions, injuries, sampling for state and other regula-
tory agencies, etc. This record should also be used as a
tool for planning future routine and special activities.
5.7.2 Operation Log
The operator should maintain a log sheet for each treat-
ment run for each treatment unit. Thereby, a permanent
plant performance record will be on file. Figure 5-4 illus-
trates a copy of a suggested condensed form.
5.7.3 Water Analysis Reports
It is recommended that the plant operator run an analy-
sis of raw and treated arsenic levels once each week for
each unit, and should run a total raw water analysis once
per month. Changes in raw water may necessitate
changes in the treatment process. Raw water changes
that can impact the treatment process include, but are
not limited to, pH, alkalinity, iron, manganese, hardness,
phosphate, silica, sulfate, sodium, TDS, and turbidity.
Figure 3-1 illustrates a copy of a suggested form. A
permanent file of these reports can be a valuable tool.
5.7.4 Plant Operating Cost Records
Using accounting forms supplied by the water system's
accountants, the plant operator should keep a complete
record of purchases of all spare parts, chemicals, labora-
tory equipment and reagents, tools, services, and other
sundry items. This should be supplemented by a file of
up-to-date competitive prices for items that have been
previously purchased.
5.7.5 Correspondence Files
The plant operator should retain copies of all corre-
spondence pertaining to the treatment plant in chrono-
logical order. Included would be intradepartmental notes
and memos, in addition to correspondence with other
individuals and/or organizations.
5.7.6 Regulatory Agency Reports
The plant operator should maintain a complete file of
copies of all reports received from state, county, or other
regulatory agencies pertaining to the treatment plant.
5.7.7 Miscellaneous Forms
The operator should have an adequate supply of accident
and insurance forms.
5.8 Treatment Plant Maintenance
The maintenance concept for the arsenic removal water
treatment plant is to isolate the equipment to be serviced
by means of shutoff valves, vent and drain lines (as
required), repair or replace equipment, fill lines, open
valves, and start service. To accomplish this, all equip-
ment items are equipped with isolating valves, and all
piping systems have vents at high points and drains at
low points.
Equipment manufacturers' recommended spare parts
should be stocked at the treatment plant to avoid lengthy
maintenance shutdowns.
If the entire treatment plant needs to be shut down and
the plant has bypass, the plant itself can be bypassed.
This can be done by closing the butterfly valves in the
raw water and treated water line and then opening the
butterfly valve in the bypass line. This would result in
untreated water with excessively high arsenic being
pumped to distribution, an event that should not occur
without the approval of the water system manager and
the regulatory agency.
5.9 Equipment Maintenance
Equipment manufacturer's maintenance instructions
should be included in the Suppliers Equipment Instruc-
tions section of the O&M Manual.
52
-------
ARSENIC REMOVAL WATER TREATMENT PLANT
OPERATION LOG
Unit# Run#
TREATMENT TO RESERVOIR
Meter End
BYPASS TO RESERVOIR
Meter End
BACKWASH TO WASTE
Meter End
REGENERATION TO WASTE
Upflow:
Meter End
Downflow:
Meter End
RINSE TO WASTE
Meter End
Date Start
Meter Start
Meter Start
Meter Start
Meter Start
Meter Start
Meter Start
Date End
Total Treated
Total Treated
Total
Total
Total
Total
NEUTRALIZATION RINSE TO WASTE
Meter End
TOTAL WASTEWATER SUMMARY
Total to Tank
Meter Start
k-gal. PERCENT WASTE
Total
k-aal.
k-aal.
k-aal.
k-aal.
k-aal.
k-aal.
k-aal.
%
TREATED WATER LOG
Date
Treatment
Meter
(k-gal)
A Meter
(k-gal)
X A Meter
(k-gal)
Raw As
(mg/L)
Treated
As*
(mg/L)
A As
(mg/L)
As
Removed
(mg)
XAs
Removed
(mg)
Average treated water arsenic.
Figure 5-4. Arsenic Removal Water Treatment Plant Operation Log
53
-------
5.10 Treatment Media
Maintenance
The plant operator should inspect the surface of each
treatment bed at least once a month. If the level of a bed
lowers more than 8 inches, makeup adsorptive media
should be added. Makeup adsorptive media should be
evenly distributed. There should be a minimum depth of
2.0 ft of water above the surface of the existing bed
through which the makeup adsorptive media will be
added. The vessel should be closed immediately and
backwashed at 7 gpm/ft2 (or at rate recommended by the
manufacturer) for at least 30 min. It is very important to
flush the fines out of the virgin activated alumina as soon
as it is wetted.
It is important that the treatment beds should not remain
in the drained condition for more than an hour. Treat-
ment units not in use should remain flooded.
5.11 Treatment Chemicals Supply
The operator should carefully monitor the consumption
of liquid chemicals and reorder when necessary. The
operator should have a method of determining the depth
of liquid in the storage tank (e.g., dipstick) and equating
that to the volume of liquid in the tank. Figure 5-5 illus-
trates a liquid depth versus volume curve for a 5,000-gal
horizontal cylindrical tank with dished head.
5.12 Housekeeping
The plant operator should wash down all equipment at
least once per month. Floors should be swept. Bathroom
and laboratory fixtures should be cleaned once per
week. All light bulbs should be replaced immediately
upon failure. Emergency shower and eyewash should be
tested once per week. Any chemical spill should be neu-
tralized and cleaned up immediately. Equipment should
be repainted at least once every five years.
i
5
1 4
5
S
o
-------
6.0 Central Treatment Plant Operating Cost
6.1 Introduction
The prime objectives in central treatment plant design
are to provide the client with a low-capital cost installa-
tion that works efficiently and reliably; is simple to oper-
ate; and is inexpensive to operate. Operating costs nor-
mally are passed directly on to the water user in the
monthly water bill. These costs include the following:
1. Treatment chemical costs
2. Operating labor costs
3. Utility costs
4. Replacement treatment media costs
5. Replacement parts and miscellaneous materials
costs
6. Waste disposal cost (not included in this manual).
As the consumer's water bill normally is based on met-
ered water consumption, the costs for treatment are pro-
rated on the unit of volume measurement. The units of
volume are usually 1,000 gal, or 100 ft3 (750 gal). The
rate units employed in this design manual are 0/1,000
gal. Some systems do not meter consumption; instead,
they charge a flat monthly rate based upon the size of
the branch connection to the water main. Although this
latter mode of distribution saves the cost of meters as
well as the reading of meters, it does not promote water
conservation. Therefore, far more water is pumped,
treated, and distributed, resulting in a net increase in
operating cost.
The common denominator that applies to both the oper-
ating cost and the bill for water consumption is the unit of
volume, 1,000 gal. Each operating cost factor can be
reduced to cost/1,000 gal. The sum total of the annual
operating costs based on total water production yields
the cost per 1,000 gal.
6.2 Discussion of Operating Costs
Similar to capital cost, many variables affect operating
cost. This manual indicates the types of operating cost
variables that are evaluated during each stage of the
design phase of the project and during the operation of
the treatment plant. The example method employed in
this manual provides the user with the ability to design
the treatment system with maximum capability and flexi-
bility. The system includes adsorptive treatment media
with spent media regeneration and pH adjustment capa-
bilities (with manual or automatic operation) that is appli-
cable primarily to activated alumina.
Manufacturers of other adsorptive media indicate that
their products are not as pH-sensitive as activated alu-
mina, and therefore do not require pH adjustment. How-
ever, some of these materials are vulnerable to a loss of
arsenic removal capacity when the treatment process pH
adjustment is not provided, due to competition from com-
peting ions such as silica, phosphate, and sulfate. Manu-
facturers also indicate that As(lll) requires oxidation to
As(V) to accomplish total arsenic removal by their prod-
ucts, and that those products have such a large arsenic
removal capacity that spent media regeneration is not
considered necessary. Some of these products are not
capable of regeneration, and, therefore, must be replaced
upon exhaustion of capacity. Under these parameters,
those products do not require operating cost for pH
adjustment chemicals.
This manual discusses systems that are capable of pro-
viding spent media regeneration and treatment process
pH adjustment. By including these capabilities in the sys-
tem design, the operation of the treatment plant has the
flexibility to include or exclude those functions. If the
system includes these capabilities, the operator may still
elect to replace the spent adsorptive media with virgin
activated alumina (or a different adsorptive media) instead
of regenerating the spent media. If a different adsorptive
media replaces the original adsorptive media, the pH
adjustment can also be added to or eliminated from the
operation. Therefore, the operator has the option of
replacing the spent adsorptive media or regenerating it.
Size of system is another variable that impacts the mode
of operation. Except for replacement of spent media,
operating labor requirements do not vary with the size of
the system, but do vary with the type of operation; the
smaller system will tend to employ the simplest oper-
ation. Replacement of spent treatment media in place of
regeneration is the main factor to consider. Spent media
replacement requires removal and disposal of spent
media, placement and conditioning of virgin media in
place of the regeneration process, and processing and
disposal of regeneration wastewater and waste solids.
Besides treatment system size, other items that influ-
ence the mode of operation are the feedwater arsenic
55
-------
concentration and the arsenic removal capacity of the
adsorptive media. The arsenic removal capacity of an
adsorptive medium increases as the arsenic concentra-
tion increases. The arsenic adsorptive capacity vs. arse-
nic concentration also may vary between media. The
costs of the adsorptive media vary. These factors are
evaluated in selection of the treatment concept and the
adsorptive media. The frequency of spent media replace-
ment/regeneration, cost of treatment chemicals, cost of
adsorptive media, waste disposal costs, and cost/avail-
ability of operating personnel not only vary with geo-
graphic locations but also are sensitive to price volatility.
Therefore, the operational flexibility provided in Chapter
3.0 of this manual allows the system to adapt to the
optimum adsorptive media and operating method at any
time.
The manual method is satisfactory for each operation
mode of the adsorptive media arsenic removal process.
If spent adsorptive media regeneration is included in the
operation, automatic operation also should be evaluated.
If the spent adsorptive media is replaced in place of re-
generation, automatic operation is not a practical option.
Media replacement is a manual function. As the feed-
water arsenic concentration increases, the frequency of
spent adsorptive media regeneration increases. As the
size of the system increases, automatic operation be-
comes more attractive. Therefore, automatic operation
will be beneficial for larger systems with high feedwater
arsenic concentration requiring more frequent regenera-
tion and stringent limits on operator time.
The following subsections discuss each of the operating
costs previously listed.
6.2. 1 Treatment Chemical Costs
The treatment chemicals discussed are limited to sulfuric
acid (HjSO^ and caustic (NaOH). Both are highly corro-
sive, hazardous liquid chemicals that require compatible
materials of construction, containment provisions, safety
provisions, weather protection, and operator training.
Although special precautions and training are required,
they are routinely accomplished. Other acids and bases
can be substituted for those chemicals, but they are usu-
ally more costly and therefore rarely considered. Other
chemicals also are used for other requirements such as
corrosion inhibition, precipitation of regeneration waste-
water solids, dewatering of precipitated solids in waste-
water, and disinfection; however, these are site-specific
requirements that are not covered in this manual.
The chemicals used for treatment of water for public
consumption require NSF/ANSI STD 60 certification by
most state regulatory agencies. It also is recommended
that the chemical supplier be required to certify that the
containers used to store and deliver the chemicals have
not been used for any other chemical; or if they have,
that they have been decontaminated according to pro-
cedures required by the governing regulatory agency.
Chemical costs are variable; recently these costs have
been volatile. Like all commodities, there is sensitivity to
the supply and demand fluctuation of the marketplace.
The geographic location of the treatment plant site in
relation to that of the supplier has an impact on the deliv-
ered cost. In some cases, the delivery costs are greater
than the cost of the chemical. The conceptual design
evaluates the chemical logistics and determines the
most cost-effective mode of procurement as well as
whether chemicals for pH adjustment are economically
feasible.
Chemical costs are sensitive to the volume and contain-
ment mode of the commodity purchased. Because com-
modity handling is minimized, bulk tank truck quantities
entail the least cost. Tank truck quantities are normally
48,000 Ib. Bulk deliveries require chemical storage tanks
within containment basins located at the treatment plant
site with necessary safety provisions and weather pro-
tection. The same commodities can be routinely pur-
chased in drums (55-gal or 30-gal), totes, carboys, gal-
lon jugs, etc. These packaged quantities result in much
higher unit prices than bulk quantity. The drum and other
small container prices also depend on the quantity pro-
cured at one time. Small containers also introduce addi-
tional handling requirements for the treatment plant
operator. For very small treatment systems, bulk pro-
curement and storage is not justified unless the feed-
water arsenic and alkalinity concentrations are extremely
high. In special low flowrate systems where high arsenic
and high alkalinity are present in the feedwater and drum
quantity costs are significantly higher than bulk quantity
costs, the increased chemical consumption could justify
bulk purchase.
The chemistry of the raw water to be treated is the most
significant factor affecting treatment chemical consump-
tion and cost. Arsenic and alkalinity are the key ions in
the raw water; the higher the concentration of either ion,
the higher the chemical consumption and cost per
1,000 gal of treated water.
6.2.1.1 Acid Cost
The most cost-effective, commercially available chemical
for lowering pH is concentrated sulfuric acid. Hydrochloric
acid also is applicable, but it is more difficult to handle,
increases chlorides (i.e., is corrosive), and usually is
more costly. The chemical designation of commercially
available sulfuric acid is 66°B' H2SO4. Its concentration is
93.14%. The remaining 6.86% is water (plus other ions).
The other ions that could be present should be eval-
uated and could result in a slight increase in their
56
-------
concentration in the treated water. Frequently, small
quantities of iron and trace amounts of heavy metals are
present. For water treatment service, there are stringent
limits on the levels of contaminants in the acid which will
be rigidly enforced. NSF certification of the acid for use
as an additive in drinking water is required.
The most economical method of procuring acid is in bulk
tank truck quantities (48,000 Ib) which are 3,100 gal
each. The tank trucks are loaded at each acid manufac-
turer's site or at a distribution storage site and delivered
directly to the treatment plant where the acid is trans-
ferred to the acid bulk storage tank. Transfer is accom-
plished by means of compressed air, which is provided
by an air compressor on the truck (unless the treatment
plant can provide the compressed air). In addition to the
lower commodity price resulting from minimum handling
and storage of the chemical, there is minimum chance of
contamination. At large treatment plants where there is
potential for high acid consumption, rail tank car quantity
(200,000 Ib) delivery, which is cheaper, may be justified.
Capital expenditures for a 16,000-gal (minimum) storage
tank and a rail spur with unloading equipment then are
required.
The delivered cost of bulk tank truck quantities of sulfuric
acid normally ranges from 4.5 to 60/lb depending on the
geographic location of the treatment plant. Drum quantity
costs are normally 10 to 120/lb higher.
The acid is consumed in three possible locations in the
treatment process at arsenic removal treatment plants
utilizing adsorptive media with pH adjustment of process
water and regeneration of spent media. First, it is used
to adjust the raw water pH to the treatment requirement;
second, it is used to neutralize the treatment bed imme-
diately after regeneration; finally, it may be used for pH
adjustment of the regeneration wastewater. In plants that
replace the spent adsorptive media rather than regen-
erate it, only the first acid feed location is required. The
raw water alkalinity dictates the amount of acid required
for the pH adjustment step. For treatment plants that do
not adjust treatment process pH, acid storage and feed
equipment is not required unless it is determined that
provisions for future pH adjustment capability is desirable.
The acid consumption for pH adjustment can be accu-
rately projected by running a titration on a raw water
sample. The cost of acid required for pH adjustment is
then determined by extending the acid addition in mg/L
to the weight (Ib) required per 1,000 gal and multiplying
by the commercial cost for the acid.
For the design example presented in Appendix B, a
hypothetical feedwater analysis includes the following:
Total alkalinity (M) = 220 mg/L (as CaCCg
Arsenic (As) =0.100 mg/L
pH =8.0.
Based upon determination by titration, the quantity of
66°B» H2SO4 required to adjust the pH to 5.5 is
205 mg/L. The amount of acid required per 1,000 gal
treated water is as follows:
205 mg 10bkg
— x — x-
Ib
mg 0.4545 kg
1.71lb/1,000gal
x 1000 gal x
3.785L
gal
Therefore, for an acid bulk quantity price of 50/lb, the
acid cost per 1,000 gal treated water is 8.50. If the acid
had been procured in drum quantities at 160/lb, the
resulting cost would be 270/1,000 gal. Conversely, if the
feedwater total alkalinity had been 100 mg/L as CaCO3
and the pH 7.5, then the resulting acid required to adjust
pH to 5.5 would be 92.4 mg/L. That equates to 0.77
lb/1,000 gal, or 3.90/1,000 gal (for acid bulk quantity
price of 50/lb). The acid requirement used in the esti-
mated operating cost estimate example is 8.50/1,000 gal.
The acid consumption for neutralization of regeneration
wastewater is a function of the caustic concentration
employed during regeneration and the raw water alka-
linity. This quantity varies from site to site. The consump-
tion also is a function of the raw water arsenic level,
which dictates the frequency of regeneration, and the
volume of water over which this cost is distributed. The
higher the arsenic level, the fewer gallons treated per
treatment cycle. The weight of acid required for neutrali-
zation after regeneration is normally in the range of
10 Ib/ft3 of treatment media.
For the design example presented in Appendix B using
activated alumina, the arsenic removal capacity is
38,940 mg/ft3 (600 grains/ft3) and the feedwater arsenic
concentration is 0.100 mg/L.
Then, the number of gallons of water from which total
arsenic is removed is
38,940
0.1/mg/L
389,400 Lft3
ft3
Then, using 10 Ib 66°B» H2SO4 per neutralization per
cubic foot regenerated adsorptive media, the cost of the
acid is
10lbacid/ft3x 50/lb = Q
103(1,000 gal)/ft3
57
-------
Therefore, for the example provided in Appendix B, acid
cost is as follows:
1. Activated alumina with spent media replacement and
without pH adjustment = 00/1,000 gal
2. Activated alumina with spent media replacement
with pH adjustment = 8.50/1,000 gal
3. Activated alumina with spent media regeneration
and pH adjustment = 90/1,000 gal.
6.2.1.2 Caustic Cost
Caustic (NaOH) can be procured in either solid (100%
NaOH) or liquid (50% NaOH or lower). The 50% NaOH
is the standard concentration that is handled and applied
to water treatment applications. That concentration is a
byproduct of the chlorine manufacturing process. There-
fore, it requires minimum handling to place it into a
48,000-lb bulk tank truck (3,850 gal). The problem with
50% NaOH concentration is that it freezes at 55°F; it is
also very viscous and difficult to transfer at temperatures
below 70°F. Therefore, it normally requires heating.
Also, because it is 50% water by weight, the freight is a
cost factor. Solid caustic in bead or flake form is also
readily available in drums or bulk. Its freight cost is
roughly half that of the liquid, but getting it into solution is
difficult and dangerous. Regardless of the economics,
solid caustic is not recommended for this application.
Commercially available caustic in the 25% NaOH con-
centration has a freezing point of 0°F; however, freight
costs for shipping this material are high (75% water).
Capital cost for larger storage and pumping require-
ments also are increased. Even though heating and
temperature protection are required, the 50% NaOH is
recommended. Transferring caustic from tank trucks to
storage tanks is accomplished with compressed air simi-
lar to the method for acid.
The delivered cost of bulk tank truck quantities of 50%
NaOH presently ranges from 10 to 150/lb depending on
the geographic location of the treatment plant. Drum
quantity cost are normally 10 to 120/lb higher.
For the activated alumina adsorptive media with treat-
ment process pH adjustment and spent media regenera-
tion, the caustic is consumed at two locations in the
treatment process. First, it is used to raise the pH of the
treated water to the level desired for distribution; second;
it is used to raise the pH of the raw water to the level
required for treatment media regeneration. The first
requirement may be reduced or replaced by aeration of
the treated water to strip free CO2 from the treated water.
The volume of 50% NaOH required for a 5% NaOH con-
centration regeneration (includes upflow and downflow
requirements) is 2 gal/ft3 per regeneration. As with the
acid required for neutralization, the caustic consumption
is a function of the raw water arsenic level which dictates
the frequency of regeneration and the volume of water
over which this cost is distributed. This varies from treat-
ment system to treatment system.
The caustic consumption for treated water pH adjust-
ment is also a function of raw water alkalinity and the
desired treated water pH. The concentration of free CO2
in the water after the initial pH adjustment with sulfuric
acid will determine the caustic requirement. The con-
sumption requirement is again accurately determined by
continuing the original titration required for acid to lower
the pH to the treatment level of 5.5; then adding the 50%
NaOH required to raise the pH to the desired level (e.g.,
7.5). The cost of caustic required then is determined by
extending the caustic addition in mg/L to the weight
required per 1,000 gal and multiplying by the commercial
price for the delivered caustic.
For the design example presented in Appendix B for
which the feedwater pH had been adjusted to 5.5 for
treatment, the treated water pH is readjusted back to a
desired level (for example, pH 7.7). For the Appendix B
example, the 50% NaOH requirement determined by
titration is 210 mg/L. The required quantity of 50% NaOH
per 1,000 gal treated water is as follows:
210x10 6 ppm x 1,000 gal (8.34 Ib/gal) =
1.75lb/1,000gal
Therefore, at a caustic bulk quantity price of 12.50/lb, the
caustic cost per 1,000 gal is 21.90/1,000 gal. If the
caustic had been procured in drum quantities at 230/lb,
the cost would be 400/1,000 gal. The caustic used in the
estimated operating cost example is 21.90/1,000 gal.
Using the same activated alumina arsenic capacity
(38,940 mg/ft3 [600 grains/ft3]) and volume of water
treated per treatment cycle (102,600 gal) discussed in
Section 6.2.1.1, the cost of caustic soda is as follows:
2 gal x 12.7 Ib/gal (50% NaOH) x 12.50/lb _
103 (x 1,000 gal treated water)
3.10/1,000 gal treated water
Therefore, for the example provided in Appendix B,
caustic soda cost is as follows:
1. Activated alumina with spent media replacement
without pH adjustment = 00/1,000 gal
2. Activated alumina with spent media replacement
with pH adjustment = 21.90/1,000 gal
58
-------
3. Activated alumina with spent media regeneration
with pH adjustment = 250/1,000 gal.
6.2.2 Operating Labor Costs
Operating labor cost is difficult to quantify. The operator
is required to be dependable and competent; however,
the position is not always full-time. Depending on the
size of the system and the other duties available for the
operator, the operator's time should be distributed over
several accounting categories. Except for days when
spent media regeneration or replacement takes place,
the treatment plant normally requires less than 1 hr per
day of operator attention. During regeneration, the oper-
ator may be required to spend approximately 8 hr over a
12-hr period. Where spent media replacement is imple-
mented, the operator time requirement is a function of
the size of the system.
On routine operating days, the operator checks the sys-
tem to see that pH is being controlled, takes and analyzes
water samples, checks instruments (flow, temperature,
pressure), and makes entries in daily logs. The only
exceptions to the normal routine include special activities
including but not limited to arsenic analyses in treatment
plant lab, equipment maintenance, and chemical tank
truck deliveries. During the remainder of the time, the
operator is able to operate and maintain other systems
(distribution, pumps, storage, etc.), read meters, or handle
other municipal responsibilities (e.g., operate sewage
treatment plant). There should always be a second oper-
ator available to take over in case of an emergency; that
individual should be well versed in the operation of the
plant.
Using the example treatment plant presented in Appen-
dix B, the cost of operational labor will be as follows (it is
assumed that the hours not used for treatment plant
operation will be efficiently used on other duties):
Given:
Flowrate: = 570 gpm
Annual average utilization: = 50%
Number of regenerations per year: = 4
Operator annual salary: = $30,000
Overhead and fringe benefits: = 30%
Available hours per year: = 2,000/man
Then:
Number of hours on
regeneration/year: 4x8 = 32 hr
Number of hours on routine
operations/year: 1 x (365-4) = 361 hr
Number of hours on extra
tasks/year: 50 x 3 hr = 150 hr
Total plant operator time: = 543 hr
Operator hourly rate: 30,000/2,000= $15.00/hr
30% (overhead and fringe benefits): = $ 4.50/hr
Operator Rate: $19.50/hr
Total operator cost: 543 hr/year x $l9.50/hr =
$10,589/year
Total gallons water produced:
0.5(570 gpm) x 1,440 min/day x 365 days/year =
149,800,000 gal/year
Labor cost/1,000 gal: $10,589/149,800 (1,000 gal) =
$0.07/1,000 gal.
If the operator had no other responsibilities and the oper-
ator's entire salary were expended against this treatment
plant operation, the operating labor cost would become
$0.25/1,000 gal. Obviously, there are many variables,
which can be controlled in different ways. Depending on
the operational philosophy of the designer/planner/man-
ager, the operating labor cost can be minimized or
maximized over a very broad range. In the case of a
very high production plant, the operating labor require-
ment is not significantly larger than that for a very small
treatment plant. Therefore, depending on relative sala-
ries, the resulting cost per 1,000 gal can range from a
few cents to more than a dollar. In proper perspective,
the operating labor cost should fall in the $0.02 to $0.30/
1,000-gal range.
If the treatment plant in the example in Appendix B had
used automatic operation in place of manual operation,
the operating labor costs might be lower. However,
because a higher skilled operator is required to maintain
and calibrate the more sophisticated instrumentation and
control equipment, the operating labor cost may not be
lower. Therefore, no reduction of operating labor cost is
assumed for systems with automatic operation.
For the example presented in Appendix B, there are
three additional operational concepts for which labor
costs should be considered. They are as follows:
The first concept applies to the activated alumina meth-
od with spent media replacement and pH adjustment.
For that operational concept, the treatment runs are the
same duration and the day-to-day operator requirements
are the same. However, the media replacement effort for
a large treatment vessel is larger. The resulting labor
requirement and resulting costs are as follows:
Number of hours on spent media
replacement/year: 4 x 20 = 80 hr
Number of hours on routine
operations/year: 1 x (365-4) = 361 hr
Number of hours on extra
tasks/year: 50 x 3 hr = 150 hr
59
-------
Total plant operator time:
591 hr 6.2.3 Utility Cost
Total labor cost: 591 hr/year x $19.50/hr =
$11,525/year
Labor cost/1,000 gal:
$11,525/149,800 (1,000 gal) = $0.08/1,000 gal.
The second concept applies to the activated alumina
example method with spent media replacement without
pH adjustment. This operational concept entails much
lower media arsenic capacity. For this example, the acti-
vated alumina media capacity reduces from 38,940 mg/ft3
(600 grains/ft3) to 5,192 mg/ft3 (80 grains/ft3). The spent
media replacement frequency increases from 4/year to
30/year.
The resulting labor requirements and tasks are as
follows:
Number of hours on spent media
replacement/year: 30x12 = 360 hr
Number of hours on routine labor
requirements/year: 1 x (365-30) = 335 hr
Number of hours on extra tasks/year:
20 x 3 hr = 60 hr
Total plant operator time: 755 hr
Total labor cost: 755 hr/year x $19.50/hr =
$14,723/year
Labor cost/1,000 gal:
$14,723/149,800 (1,000 gal) = $0.098/1,000 gal.
The third concept applies to the other adsorptive media
that can be applied to arsenic removal treatment system
with spent media replacement without pH adjustment.
Furthermore, the arsenic removal capacity may be such
that the spent media need only be replaced once per
year. The resulting labor and cost requirements are as
follows:
Number of hours on spent media
replacement/year: 1 x 20 = 20 hr
Number of hours on routine
operations/year: 1 x (365-1) = 364 hr
Number of hours on extra tasks/year:
20x3hr = 60 hr
Total plant operator time: 444 hr
Total labor cost: 444 hr/year x $19.50/hr =
$8,658/year
Labor cost/1,000 gal:
$8,658/149,800 (1,000 gal) = $0.06/1,000 gal.
The utility cost is normally electric utility. However, there
also can be telephone and natural gas (or oil) utility
costs. Telephone service to the treatment building is
recommended as a safety precaution in case of accident
as well as operator convenience. Cost for that service
should be the minimum available monthly rate. Depend-
ing upon the local climate, the cost for heating can vary.
The purpose of the building is to protect the equipment
from elements (primarily freezing), not for operator com-
fort. Normally the treatment units act as heat sinks, main-
taining an insulated building at a temperature near that
of the raw water. In cold climates, the building should
have an auxiliary heat source to prevent freezing of
pipes in the event that the water is not flowing. If the
client determines that the treatment building is to serve
additional functions, heating to a comfort temperature
could be an additional required cost.
Electric power will be needed for the following functions:
1. Chemical pumps
2. pH controls
3. Caustic storage tank immersion heater
4. Lighting
5. Convenience receptacle
6. Aeration unit blower (optional)
7. Repressurization pump (optional)
8. Extra load on well pump for regeneration/backwash
wastewater, and loss of head through the treatment
system.
Items 1, 2, 4, and 5 are negligible. Item 3 is a function of
the climate and the heat losses through the insulation.
Provisions to conserve energy for this function should be
incorporated. Item 6 is a relatively small load (1-3 hp
blower motor). Item 7 is potentially the biggest electrical
load. This requirement only exists when aeration is used
to adjust treated water pH, and the water is pumped to
an elevated storage tank. This electrical load can be
equal to the well pump motor load. However, when
repressurization is a requirement, then the well pump
should be modified to reduce its discharge pressure
capability to only that which is required to pump the raw
water through treatment into the clean/veil in place of the
pressure to pump to the elevated storage tank. Then the
net increase of electrical energy consumption is nearly
negated. Item 8 amounts to 3-5% of the well pump elec-
trical energy consumption.
60
-------
The electrical utility rates also vary considerably from
one geographic location to another. In August 2001,
rates varied from $0.03 to $0.20/kWh. The electrical util-
ity cost can range from $0.005 to $0.02 per 1,000 gal
under normal conditions. Under abnormal conditions, the
cost could be 50/1,000 gal or higher.
6.2.4 Replacement Treatment
Media Cost
The consumption of treatment media per regeneration
for a system with process water pH adjustment and spent
media regeneration in a well-operated activated alumina
arsenic removal water treatment plant should be 5% of
the bed volume. However, there are additional ways in
which the media can be lost.
The loss of media occurs during regeneration. In order to
remove virtually all of the arsenic from the grains of acti-
vated alumina with a 5% NaOH regenerant solution, a
small amount of aluminum is dissolved. This is a process
requirement because the attractive forces between the
arsenic and the alumina are extremely strong.
During regeneration and neutralization, excessively high
and/or low pH contact will attack the treatment media. If
the pH of the regeneration solution exceeds the recom-
mended 5% NaOH, the solution becomes increasingly
aggressive to the activated alumina. Similarly, if the pH
of the neutralization solution is lower than pH 2.0, a
more severe dissolving of the alumina takes place. Sam-
ples taken during the regeneration cycle should period-
ically be analyzed for aluminum.
Backwash, if conducted carelessly, also can result in
media carry over. An excessive backwash rate can
expand the treatment media by an amount that carries the
adsorptive media out of the vessel resulting in loss of
media. Monitoring the backwash water will detect and
provide prevention of that. If backwash water flows into
the wastewater surge tank, the lost media can be recov-
ered.
A final way for the media to be lost is through the effluent
underdrain (collection system) within the bed. If media
grains ever appear in the treated effluent, the treatment
unit should be immediately taken out of service for
inspection (and repair) of the collection system.
Media replacement costs are difficult to predict. Signifi-
cant media replacement can occur at a treatment plant
where backwash at an excessive rate for an extensive
period has been required to remove filtered solids from
the media. A plant in which suspended solids in the raw
water require frequent extended backwashing is vulner-
able to loss of media problems. For systems encoun-
tering such conditions an upstream filter (e.g., bag filter)
should be evaluated.
A typical pricing structure for a representative activated
alumina product suitable for arsenic removal is provided
in Table 6-1.
Table 6-1. Price for Typical -28, +48 Mesh
Activated Alumina
Quantity
Price1"
2,000-1 0,000 Ib
1 2,000-20,000 Ib
22,000-38,000 Ib
40,000 Ib and over
$1 .00/lb
0.90/lb
0.75/lb
0.70/lb
(a) August 2001 prices.
A conservative bed replacement estimate is 20% per
year. In the example in Appendix B where two 380 ft3
beds are used, the media replacement will be:
2 x 380 ft3 x 45 Ib/ft3 x $.70/lb x 0.2 = $4,788/year
$4,788/149,800 (1,000 gal) = $0.032/1,000 gal.
As discussed in Section 6.2.2, there are three additional
operational concepts for which replacement media costs
will be considered, they are as follows:
The first concept applies to the activated alumina exam-
ple method with spent media replacement and pH adjust-
ment. As pointed out, four spent treatment beds will be
replaced per year.
Therefore, the media replacement cost for this treatment
mode is:
4/year (380 ft3) (45 Ib/ft3) x $0.70/lb = $47,880/year
The replacement treatment media cost/1,000 gal =
$47,880/year
149,800 (1,000 gal)
= $0.32/1,000 gal
The second concept applies to the activated alumina
example method with spent media replacement without
pH adjustment. This operational concept entails very low
media arsenic capacity. The spent media replacement
frequency increases from 4/year to 30/year. Therefore,
the media replacement cost for this treatment mode is:
30/year (380 ft3) (45 Ib/ft3) x $0.70/lb = $359,100/year
The replacement treatment media cost/1,000 gal is
61
-------
$359,100/year
149,800 (1,000 gal)
= $2.40/1,000 gal
The third concept applies to the other adsorptive media
that can be applied to arsenic removal water treatment
systems with spent media replacement without pH
adjustment. Furthermore, the arsenic removal capacity
may be so much greater than activate alumina that the
spent media need only be replaced once per year.
These media have been reported to cost from $1 to
$4/lb.
Therefore, if that arsenic removal capacity is verifiable,
then the media replacement cost using $1/lb for this
treatment mode is:
1/year (380 ft3) (45 Ib/ft3) x $1.00/lb = $17,100/year
The replacement treatment media cost/1,000 gal is:
$17,100/year
149,800 (1,000 gal)
= 0.110/1,000 gal
replacement pump heads) should be kept in stock in the
treatment plant, to prevent extended plant shutdown in
the event a part is required. Also included are consum-
ables such as laboratory reagents (and glassware), and
recordkeeping supplies. An operating cost allowance of
$0.01/1,000 gal of treated water is conservative.
6.3 Operating Cost Summary
The range of adsorptive media arsenic removal water
treatment plant operating costs discussed above are
summarized in Table 6-2. As has been pointed out, the
range of costs is very broad.
For adsorptive media arsenic removal water treatment
plants in which flowrates, raw water arsenic concentration,
raw water analyses (pH, alkalinity, silica, sulfate, etc.),
adsorptive media, labor rates, and utility rates vary from
the values used in the example in Appendix B, the oper-
ating costs will deviate from those indicated in Table 6-2.
The information included in this subsection provides a
method for the determination of an operating cost esti-
mate for any adsorptive media arsenic removal water
treatment plant.
6.2.5 Replacement Parts and
Miscellaneous Material Costs
This is a very small operational cost item. Replacement
parts (e.g., chemical, pump diaphragms, seals and
Table 6-2. Operating Cost Tabulation for an Activated Alumina Plant"
Operating Cost Items
Flowrate: 570 gpm
Dollars/1,000 Gal Treated Water
Manual Operation
Treatment Chemicals - acid
- caustic
Operating Labor
Utility
Replacement Treatment Media
Replacement Part and Misc. Material
Total
Activated Alumina
with Spent Media
Replacement
without pH Adjustment
0.00
0.00
0.10
0.01
2.40
0.01
2.52
Activated Alumina
with Spent Media
Replacement
with pH Adjustment
0.08
0.22
0.08
0.02
0.32
0.01
0.73
Activated Alumina
with Spent Media
Regeneration
with pH Adjustment
0.09
0.25
0.07|bl
0.02
0.03
0.01
0.47
Other Adsorptive Media
with Spent Media
Replacement
without pH Adjustment10'
0.00
0.00
0.06
0.01
0.11
0.01
0.19
(a) Wastewater and waste solids, processing and disposal not included.
(b) Applicable to automatic operation.
(c) Cost to oxidize As(lll) to As(V) not included.
62
-------
7.0 References
American Public Health Association-American Water
Works Association-Water Environment Federation.
Standard Methods for the Examination of Water and
Wastewaters, 19th ed. A.E. Greenberg (Ed.), American
Public Health Association, Washington, DC.
Battelle. 2002. Cosf Estimating Program for Arsenic
Removal by Small Drinking Water Facilities. Developed
for the United States Environmental Protection Agency,
National Risk Management Laboratory, Cincinnati, OH.
December.
Clifford, D., L. Ceber, and S. Chow. 1983. "Arsenic(lll)/
Arsenic(V) Separation by Chloride-Form Ion-Exchange
Resins." Proceedings of the XI AWWA WQTC.
Edwards, M., S. Patel, L. McNeill, H. Chen, M. Frey,
A.D. Eaton, R.C. Antweiler, and H.E. Taylor. 1998. "Con-
siderations in As Analysis and Speciation." J. AWWA
(March): 103-113.
Ficklin, W.H. 1982. "Separation of Arsenic (III) and Arse-
nic (V) in Groundwaters by Ion Exchange." Talanta,
30(5): 371-373.
Ghurye, G., and D. Clifford. 2001. Laboratory Study on
the Oxidation of Arsenic III to Arsenic V. EPA/600/R-01/
021. United States Environmental Protection Agency,
National Risk Management Laboratory, Cincinnati, OH.
Lowry, J.D., and S.B. Lowry. 2002. Oxidaton ofAs(lll) by
Aeration and Storage. EPA/600/R-01/102. United States
Environmental Protection Agency, National Risk Man-
agement Laboratory, Cincinnati, OH.
Rubel, F. 1981. Report on Investigation of Alcoa F-1
Activated Alumina for Removal of Excess Arsenic from
Potable Water. Aluminum Company of America, P.O.
No. TC331114TC. December 11.
Rubel, F. 1984. Concept Design Report of Arsenic
Removal Water Treatment Plants at Fallon, Nevada.
Department of the Navy, Contract No. N62474-81-C-
8532. May 14.
Rubel, F., and F. Williams. 1980. Pilot Study of Fluoride
and Arsenic Removal from Potable Water. EPA-600/2-
80-100. August.
Wang, L., A. Chen, T. Sorg, and K. Fields. 2002. "Field
Evaluation of Arsenic Removal by IX and AA." J. AWWA,
94:4: 161.
63
-------
-------
Appendix A
Summary of Subsystems Including Components
The items that are designated as "optional" are not man-
datory requirements. Some of those items may already
be included in systems other than treatment and there-
fore, would be redundant. Other items, though desirable,
are not mandatory. Automatic and semiautomatic opera-
tion is optional. Therefore, for each instrument and con-
trol item, though not indicated for clarity, there is an
automatic option.
For Schematic Flow Diagram, see Figure A-1.
1. Raw Water Influent Main
a. Flow control
b. Flowrate measurement, flow total
c. Acid injection for pH adjustment
d. In-line static mixer
e. pH measurement, indicator, alarm, and fail-safe
control
f. Pressure indicator
g. Pressure control (optional)
h. Backflow preventer
i. Sample before pH adjustment piped to sample
panel (optional)
j. Sample after pH adjustment piped to sample
panel (optional)
k. Isolation valve
I. Temperature indicator (optional)
2. Intervessel Pipe Manifold
a. Process control valves
b. Pressure indicators
c. Sample piped to sample panel (optional)
3. Treated Water Effluent Main
a. Caustic injection for pH adjustment
b. In-line static mixer
c. pH measurement, indicator, alarm and fail-safe
control
d. Sample after pH adjustment piped to sample
panel (optional)
e. Pressure indicator
g.
h.
Aeration subsystem(optional)
i. Air blower (optional)
ii. Clean/veil (optional)
Booster or repressurization pump (optional)
Disinfection injection (optional)
Isolation valve
4. Raw Water Bypass Main
a. Flow control
b. Flowrate measurement, flow total
c. Backflow preventer
d. Isolation valve
5. Backwash/Regeneration Feed Main (optional)
a. Flow control
b. Flowrate measurement, flow total
c. Caustic injection for pH adjustment
d. Acid injection for pH adjustment
e. In-line static mixer
f. pH measurement
g. Sample after pH adjustment piped to sample
panel (optional)
h. Backflow preventer
i. Isolation valve
6. Wastewater Main (optional)
a. Backflow preventer
b. Process isolation valves
c. Acid injection for pH adjustment
d. Coagulation chemical injection
e. In-line static mixer
f. Sample after chemical injection piped to sample
panel (optional)
7. Treatment Unit
a. Pressure vessel
b. Treatment media
c. Internal distribution and collection piping
d. Pressure relief valve
e. Air/vacuum valve
f. Operating platform and/or ladder (optional)
65
-------
[-»—
i
i
' i
T
(-'-
5
; -*•
i
*$
A
1
'"A /^
Ox^
— — *
\=J
NO OH
Q«f TANK
i
f™"| ,— -~ i
* /^\
| L(PHE)-
" Hi * 10 1
**HS '
?
1
?
i
»
M
-
]
H^"? £
(FT)
X (OPTIONAL)
xq— (f)
""""
r^i
0 RAW (FEED) \^
*^** /r-fN. HTS-S-SS *""' !
\__/ Jko«5aj |
x /*""\ r~i $
jr— fpHEH A 4
- — nrH I vy | 5
ill! t 2
H2S04 S
»V
f H J
1
___„
T :
s I
"~y
UJ
R
S
"1™
t/i
s
i£
CJ
s
1
f
i
WNK
i
J t
r CK
i *A ^
1 r~^ ^\
il J^-^-Xi
TO
W*SI£ __
f PI j
TREATMENT T''
VESSFL
;
^_^«s
. T Js-g
i— — »
i ifa
1™™™™™ p*
— — IX
4 tx
A3
v A/V
i--J 0-
__i-^
"^
A f
T
1 s
^^ ^
IK, j» ii[-
^ •""" "•" "— i>Q p ^ ^ ^^^
TREA^ENT
VESSEL
E
^^
X rC>O"
©
1
i
S
— 1
1
S
r
*
/
1
1
ffAl 9CKI
m]6«s
T
;-• T '•'- iQ^'^y-^-^
i i /~\ n> i — -n
1 f-(p«)J f V
i f *Tu) ^ m
[ t «RMS i aj
^ S H . SO .
!
i
3
r-^
rfl f!talf!
^ |^«y
;
"s
5
F
3^—
3™
|l
TG
WST
•c*<
"™1
f
,
TO TANK m*m*
BLENDED
TO
................... fc, fx-x-x-m ik
^
;
->
k
S
r
i
Of,
i , i
i —
5
S
a
LJ
-------
8. Sample Panel (optional)
a. Manifolds
i. Influent manifold (influent main sample and
raw water samples from each treatment
vessel after pH adjustment)
ii. Effluent manifold (effluent main sample after
pH adjustment, treated water samples from
each treatment vessel and wastewater
manifold sample after pH adjustment and
chemical injection)
iii. pH indicator (influent sample manifold and
effluent sample manifold)
iv. Sample collection spigots with drain
b. Wet chemistry laboratory bench with equipment,
glassware, reagents, etc.
9. Acid Storage and Feed Subsystem
a. Emergency shower and eyewash, signage
b. Acid storage tank (outside treatment building)
i. Fill, discharge, drain, vent, and overflow
piping
ii. Liquid level sensor (optional)
iii. Desiccant air dryer in vent (optional)
iv. Weather protection
v. Containment basin
c. Acid day tank (inside treatment building)
i. Fill pipe float valve
ii. Drain valve
iii. Containment basin
d. Acid pumps
i. Treatment unit pH adjustment
ii. Neutralization pH adjustment
iii. Wastewater pH adjustment (optional)
e. Acid piping (interconnecting piping)
i. Between storage tank and day tank
ii. Between feed pumps and raw water
injection point
iii. Between feed pumps and regeneration feed
and wastewater mains injection points
(optional)
iv. Backflow prevention
10. Caustic Storage and Feed Subsystem
a. Emergency shower and eye wash, signage
b. Caustic storage tank (outside treatment building)
i. Fill, discharge, drain, vent, and overflow
piping
ii. Liquid level sensor (optional)
iii. Immersion heater with temperature control
iv. Weather protection
v. Containment basin (optional)
c. Caustic day tank (inside treatment building)
i. Fill line float valve
ii. Drain valve
iii. Containment basin (optional)
d. Caustic piping (interconnecting piping)
i. Between storage tank and day tank
ii. Between feed pump and, regeneration feed
main injection point (optional)
iii. Between feed pump and treated effluent
main injection point (optional)
iv. Backflow prevention
11. Backwash Water Disposal System (optional)
a. Surge tank (optional)
b. Unlined evaporation pond (optional)
c. Sewer (optional)
d. Drainage ditch (optional)
e. Other discharge method (optional)
12. Toxic Regeneration Wastewater Disposal System
a. Surge tank (optional)
b. Wastewater reclamation system (optional)
c. Other discharge method (optional)
67
-------
-------
Appendix B
Treatment System Design Example
This design example is applicable to a specific manually
operated activated alumina arsenic removal water treat-
ment system employing treatment process pH adjust-
ment and regeneration of spent treatment media. This
design example is adaptable to any other arsenic removal
adsorptive media treatment system by deletion of equip-
ment and/or adjustment of equipment size as described
in Chapter 3.0. This example is applicable to any of the
following combinations of options:
1. Replacement of spent media in place of regeneration
2. Deletion of treatment process pH adjustment
3. Application of other adsorptive media in place of
activated alumina
4. Adjustment of EBCT
5. Adjustment of flowrate
6. Adjustment of arsenic concentration
7. Adjustment of raw water chemical analysis
8. Automatic operation in place of manual operation.
Given:
q (flowrate) = 570 gpm
N (number of treatment trains) = 1
n (number of treatment vessels/train) = 2
Raw water arsenic concentration = 0.100 mg/L
Arsenic MCL = 0.010 mg/L
Treated water arsenic design concentration =
0.008 mg/L (max)
Activated alumina arsenic removal capacity =
1,376 g/nf (600 grains/ft3)
(Note: Indicated capacity applies only to system with
raw water 0.100 mg/L arsenic concentration and
treatment process pH adjusted to 5.5)
Md (media density) = 45 Ib/ft3
EBCT = 5 min
Pipe material-Type I Schedule 80 PVC,
v (pipe velocity) = 5 ft/second (max.)
p (system pressure): 50 psig (max.)
T (ambient temperature): 95°F (max.)
Tw (water temperature): 85°F (max.)
1. Vessel and Treatment Bed Design (reference:
Figure 3-5)
Solve for: h (treatment bed depth)
d (treatment bed diameter)
A (treatment bed horizontal surface area)
V (treatment bed volume)
Mw (total weight of treatment media)
D (vessel outside diameter)
H (vessel overall height)
When EBCT = 5 min, then flowrate = 11/2 gpm/ft3
media.
Then, q = 570 gpm; therefore
1.5 gpm/ft
Then, when h = 5 ft,
V 380 ft3
=380ft3
_
~"~~
5ft
= 76 ft2
Then, d2=
71
4x76ft
QQJQ ft2
Then, d = 9.83 ft = 9-10-
Then, D = d + 1« = 9» 11«, therefore use D = 10» 0»
(then, A = 77.2 ft2)
Then,
= ft3
Then, Mw = 2 vessels x 386 ft3 x 45 Ib/ft3 = 34,800 Ib
Because the media quantity is almost a 40,000 Ib
truckload, it is prudent to procure a truckload quantity.
Then the treatment vessel dimensions (see Figure 3-
5) are as follows:
69
-------
H = h + h/2 +6- +(2)0/4 + 1- =
60« + 30« + 6« + 2 (12O/4)+ 1« = 157- = 13- 1«
D = 10-0-
2. Pipe Sizing
Solve for: Sizes for all water pipe mains
Mains: q = 570 gpm (max)
Try 6», v = 6.5» / sec. > 5« / sec., therefore NG
Try 8», v = 3.6» / sec. < 5» / sec., therefore OK
Use 8« Schedule 80 PVC
Backwash rate is not to exceed rate required for
50% treatment bed expansion.
Then, backwash rate = A x 7 gpm/ft2 = 77.2 ft2 x
7 gpm/ft2 = 540 gpm <570 gpm, therefore OK. (Note:
The backwash rate is sensitive to water temperature.)
3. Acid Subsystem Design
(Note: This subsystem is not applicable for systems
that do not include treatment process pH adjustment.)
a. Storage Tank Size
Storage tank size is based upon logistical require-
ments which are a function of treatment plant
acid consumption rate and bulk tank truck deliv-
eries of acid. The tank truck can deliver up to
48,000 Ib of 66°B« H2SO4. The density of this
liquid is 15.5 Ib/gal. Therefore, a delivery con-
tains 3,100 gal.
In this example the peak treatment flow is
570 gpm, and it is assumed that the acid con-
sumption (determined by titration) is 0.05 gal/
1,000 gal treated water. Then the acid consump-
tion is 1.71 gal/hr. Then, a tank truckload would
supply a minimum of 1,800 hr of treatment oper-
ation. Acid consumption for raw water pH reduc-
tion, which is a function of total alkalinity and
free CO2, is discussed in Appendix C.
A 5,000-gal acid storage tank provides capacity
for more than VA bulk tank truckloads of 66°B»
H2SO4. Therefore, when half a truckload has
been consumed (providing capacity for the next
truckload delivery), there is a minimum of a
900-hr (37 days) acid supply available in storage
before the acid supply is exhausted.
b. Day Tank Size
The storage tank supplies a polypropylene day
tank located inside of the treatment building. A
100-gal day tank will satisfy more than 200% of
the maximum treatment process pH adjustment
acid requirements (41 gal/day) for maximum
treatment flow of 820,800 gal for one day.
c. Acid Pump Size
The acid feedrate required for the treatment
process pH adjustment function is: 570 gpm x
60 min/hr x 0.05 gal acid/1,000 gal water =
1.71gph
The acid feedrate required for the treatment
process pH adjustment function (1.71 gph) is
satisfied by a positive displacement diaphragm
pump that has a maximum flowrate of 2.5 gph
@ 50 psig with a 1,000:1 turndown capability
(materials of construction to be recommended
for 66°B» H2SO4 service).
For neutralization of the treatment bed after
completion of regeneration and the regeneration
wastewater flowing from the treatment vessel to
the regeneration wastewater surge tank two
additional acid feed pumps are required (Note:
For systems that replace spent media in place of
regeneration, this equipment is not applicable.)
The rule of thumb relating to the volume of acid
required to be applied to accomplish both func-
tions is 1 gal/ft3 (activated alumina), or 386 gal/
regeneration. The acid feed for these two func-
tions will take place over a period of 4 to 6 hr.
The first pump feeds acid into the regeneration
feedwater main to adjust the pH initially to 2.5,
then to 4.0, and finally at completion of the
neutralization to 5.5. The second pump feeds
acid into the wastewater main at a rate required
to adjust the pH of the entire wastewater batch
to a range of 6.0 to 6.5. This latter acid feed
requirement can take place at a constant rate
that will provide the necessary wastewater pH
for the volume of the entire wastewater batch
(thoroughly mixed in the wastewater surge tank)
at the conclusion of the regeneration process.
The two acid feed pumps required for the two
functions can be identical air-operated dia-
phragm pumps with maximum flowrate of 2 gpm
at 50 psig with a 100:1 turndown capacity
(materials of construction to be recommended
for 66°B» H2SO4 service).
A 5-hp air compressor with a 60-gal receiver
capable of supplying 14.7 cfm at 175 psig com-
pressed air. The air compressor will supply com-
pressed air for both air-operated diaphragm acid
70
-------
feed pumps, the air-operated diaphragm caustic
soda feed pump, and (for automatic operation)
the pneumatic-operated process control butterfly
valves. If there is a wastewater sludge dewater-
ing system, the air compressor will be available
to operate the air-operated diaphragm pump (for
sludge transfer) and the plate and frame filter
press.
4. Caustic Subsystem Design
(Note: This subsystem is not applicable for systems
that do not include treatment process pH adjust-
ment)
a. Storage Tank Size
Storage tank size is based upon logistical require-
ments which are a function of treatment plant
caustic consumption rate and bulk tank truck
deliveries of caustic. The tank truck can deliver
up to 48,000 Ib of 50% NaOH. The density of
this liquid is 12.9 Ib/gal. Therefore, a delivery
contains 3,700 gal.
In this example the peak treatment flow is
570 gpm, and it is assumed that the caustic con-
sumption (determined by titration) is 0.135 gal/
1,000 gal treated water. Then the caustic con-
sumption is 4.6 gal/hr. Then, a tank truckload
would supply a minimum of 800 hr of treatment
operation.
A 5,000-gal caustic storage tank provides
capacity for more than 1% bulk tank truckloads
of 50% NaOH. Therefore, when 75% of a truck-
load has been consumed (providing capacity for
the next truckload delivery), a minimum of 900 gal
remains, which provides a 200-hr (8-day) caustic
supply available in storage before the caustic
supply is exhausted. Note: When the supply
remaining in the storage tank provides capacity
for a bulk tank truck delivery, spent media regen-
eration (if applicable) will be deferred until after
caustic delivery.
b. Day Tank Size
The storage tank supplies a polypropylene day
tank located inside of the treatment building. A
500-gal day tank will satisfy more than 200% of
the maximum treatment process pH adjustment
caustic requirements (110 gal/day) for maximum
treatment flow of 820,800 gal for one day as well
as the requirement for one step of the two-step
spent media regeneration.
c. Caustic Pump Size
The caustic feedrate required for the treatment
process pH adjustment function is: 570 gpm x
60 min/hr x 0.135 gal caustic/1,000 gal water =
4.6gph.
The caustic feedrate required for the treatment
process pH adjustment function (4.6 gph) is satis-
fied by a positive displacement diaphragm pump
that has a maximum flowrate of 5 gph @ 50 psig
with a 1,000:1 turndown capability (materials of
construction to be recommended for 50% NaOH
service).
For regeneration of the activated alumina treat-
ment bed two regeneration steps are required
utilizing 15 gal of 5% NaOH/ft3 per step. (Note:
For systems that replace spent media in place of
regeneration this equipment is not applicable.)
The following calculations provide the volume
and flowrate of 50% NaOH required per regen-
eration.
Given:
d, = density 5% NaOH = 8.8 Ib/gal
d2 = density 50% NaOH = 12.9 Ib/gal
v, = volume 5% NaOH/regeneration step-ft3 =
15gal/step-ft3
n = number of steps = 2 (upflow and downflow)
V = 386 ft3 (activated alumina)
Find:
w, = weight of 5% NaOH/step-ft3
v2 = volume 50% NaOH required/regeneration
step
Then: w, = v^d,) = 15 gal/ft3 x 8.8 Ib/gal =
132lb/step-ft3
Then: 100% NaOH = 132 Ib/step-ft3 x .05 =
6.6 Ib/step-ft3
Then: 50% NaOH =
100% NaOH x 2 = 13.2 Ib/step - ft3 =
13.2lb/step-ft:
12.9 Ib/gal
= 1 gal/step-ft;
Then: v2 = 1 gal/step-ft3 x 386 ft3 = 386 gal/step
71
-------
Then: If, step duration is 60 min,
50% NaOH flow/rate = 386gal = 6.4 gpm
60 minutes
Then: Total 50% NaOH required per regenera-
tion = v2 x n = 386 gal/step x 2 steps = 772 gal.
The caustic feed pump required for this function
will be an air-operated diaphragm pump with
maximum flowrate of 15 gpm at 50 psig with
100:1 turndown capability (materials of construc-
tion to be recommended for 50% NaOH service.
The recommended air compressor for the acid
air-operated diaphragm pumps also will provide
the compressed air for this function.
5. Regeneration Wastewater Surge Tank Design
Given:
Maximum volume of regeneration wastewater per
cubic foot media = 400 gal/ft3
Number of cubic feet of media per regeneration =
386ft3
Tank construction - epoxy interior lined carbon steel
Find:
Volume of wastewater per regeneration = 400 gal/ft3
x 386 ft3 = 155,000 gal = 20,600 ft3
Dimensions of surge tank (use height = 20 ft)
* ^2 4x20,600 ft3 . -,. _2
Then, (diameter) = ' = 1,310ft2
7rx20 ft
Then, diameter = 36 ft
Then tank dimensions = 36» $ x 20» h
Suggested Containment Basin Dimensions: length
80 ft, width 72 ft, height 4 ft; volume = 22,430 ft3 =
168,200 gal >155,000 gal.
72
-------
Appendix C
Discussion of Acid Consumption Requirements for pH Adjustment of Raw
Water
This manual discusses acid titration as the practical
method used to determine the acid feed requirement for
lowering the raw water pH to 5.5. However, this also can
be accomplished theoretically when a raw water analysis
is available and raw water samples are not. This method
requires the pH, the total alkalinity (M as mg/L CaCO3),
and/or the free carbon dioxide (CO2 as mg/L) from the
raw water analysis in addition to the graph illustrated in
Figure C-1. If only two of the three raw water analysis
items are available, the third is determined by the graph.
The pH curves illustrated in Figure C-1 were developed
from theoretical chemical formulae which integrate the
relationship between pH, alkalinity and free CO2.
Trial-and-error usage of these curves rapidly leads the
user to the acid feed requirement for the desired pH
adjustment. The objective is to determine the amount of
alkalinity reduction that is required to lower the pH to the
desired amount, and then to convert the alkalinity reduc-
tion to acid addition. The user should be aware of the
fact that the reduction in alkalinity coincides with the
corresponding increase in free CO2. The following exam-
ples best illustrate this method:
Example 1:
Given:
Raw water pH = 8.0
Raw water M = 220 mg/L as CaCO3
Raw water CO2 = 4 mg/L
Find:
1. M and free CO2 for pH adjusted to 5.5
2. 66°B» H2SO4 required feedrate to adjust pH
to 5.5
1. Try reducing M by 200 mg/L (as CaCO.) to 20 mg/L
(as Ca CO.,)
Then, increase in free CO2 (M multiplied by 0.88),
200x0.88 = 176 mg/L
Then, total free CO2 = 176 + 4 = 180 mg/L
Then, using graph we find that the pH is 5.4 when:
a. M = 20 mg/L (as CaCO.,)
b. C02 = 180 mg/L. Therefore, NG.
Therefore, too much alkalinity was removed. Try
reducing M by 196 mg/L (as CaCO3) to 24 mg/L (as
CaCO3).
Then, increase in free CO2 = 196 mg/L x 0.88 =
172.5 mg/L
Then, total free CO2 = 172.5 + 4 = 176.5 mg/L.
Then, using graph we find that the adjusted raw
water pH is 5.5 when:
a. M = 24 mg/L CaCO3
b. CO2 = 176.5 mg/L. Therefore, OK.
2. For each 100 mg/L (as CaCO3) reduction of total
alkalinity, 105 mg/L 66°B« H2SO4 will be added.
Therefore, reduce M by 196 mg/L (as CaCO3) by
feeding 1.96 mg/L (CaCO.,) x 105 mg/L H2SO4/mg/L
CaCO3 = 205.8 mg/L I-LSO, to adjust raw water pH
to 5.5. If we desire to find what acid feedrate would
be required per 1,000 gal of treated water, we find
that:
Feedrate = (205.8 x 10~6 mg/L) x
(1,000 gal x 8.34 Ib/gal) / (15.5 Ib/gal) =
0.11 gal H2SO4 /1,000 gal water
73
-------
1000
1 10 100
TOTAL ALKALAW1W AS CaCOj - mg/L
Figure C-1. Graph of pH as a Function of Total Alkalinity and Free Carbon Dioxide
1000
Example 2:
Given:
Raw water M = 100 mg/L (as CaCO.,)
Free CO2 = 6 mg/L
Find:
1. Raw water pH
2. M and free CO2 for pH adjusted to 5.5
3. 66°B» H2SO4 required feedrate to adjust pH
to 5.5
1. From graph we find raw water pH to be 7.5
2. Try reducing M by 80 mg/L (as CaCO3) to 20 mg/L
(as CaCO3)
Then, increase in free CO2 = 80 x 0.88 = 70.4 mg/L
Then, total free CO2 = 70.4 + 6 = 76.4 mg/L
Then, using the graph we find the adjusted pH to be
5.75 when:
a. M = 20 mg/L (as CaCO.)
b. CO2 = 76.4 mg/L. Therefore, NG.
Therefore, too little alkalinity was removed, try reduc-
ing M by 87 mg/L (as CaCO3) to 13 mg/L CaCO.,).
Then, increase in free CO2 = 76.5 + 6 = 82.5 mg/L
Then, using the graph we find the adjusted pH to be
5.55 when:
a. M = 13 mg/L (as CaCO.,)
b. CO2 = 82.5 mg/L. Therefore, NG.
Therefore, too little alkalinity was removed; try reduc-
ing M by 88 mg/L (as CaCO3) to 12 mg/L CaCO,).
Then, increase in free CO2 = 88 x 0.88 = 77.5 mg/L
74
-------
Then, total free CO2 = 77.5 + 6 = 83.5 mg/L 3. Therefore, reduce M by 88 mg/L (as CaCO3) by
feeding 0.88 x 105 mg/L H2SO4/100 mg/L CaCO3 =
Then, using the graph we find the adjusted raw 92.4 mg/L 66°B« KSO, to adjust raw water pH to 5.5
water pH to be 5.5 when:
Acid feedrate = (92.4 x irr6 mg/L) x
a. M = 12mQ/LfasCaCO..) (1,000 gal x 8.34 Ib/gal) / (15.5 Ib/gal) =
b. CO: = 83.5 ma/L. Therefore. OK. 0.05 gal H2SO4 /1,000 gal water
75
-------
-------
Appendix D
Tabulations of Estimated Capital Cost Breakdowns for Arsenic Removal
Water Treatment Plants by Means of the Activated Alumina Process
at Typical and Ideal Locations
Contents
D-1 Typical Locations with Manual Operation, Replacement of Spent Media, and Without Process Water pH
Adjustment
D-2 Typical Locations with Manual Operation, Replacement of Spent Media, and with Process Water pH Adjustment
D-3 Typical Locations with Manual Operation, Spent Media Regeneration, and with Process Water pH Adjustment
D-4 Typical Locations with Automatic Operation, Spent Media Regeneration, and Process pH Adjustment
D-5 Ideal Locations with Manual Operation, Replacement of Spent Media, and Without Process Water pH
Adjustment
D-6 Ideal Locations with Manual Operation, Replacement of Spent Media, and with Process Water pH Adjustment
D-7 Ideal Locations with Manual Operation, Spent Media Regeneration, and with Process Water pH Adjustment
D-8 Ideal Locations with Automatic Operation, Spent Media Regeneration, and Process pH Adjustment
77
-------
Table D-1. Estimated Capital Cost(a) Breakdowns for Central Arsenic Removal Water Treatment Plants at Typical
Locations by Means of the Activated Alumina Process With Manual Operation, Replacement of Spent
Media, and Without Process Water pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
7
4
N/A
N/A
40
19
6
5
30
N/A
26
8
34
11
115
100
31
7
9
4
N/A
N/A
51
24
6
6
36
N/A
29
9
38
13
138
200
38
13
13
5
N/A
N/A
69
25
7
8
40
N/A
35
11
46
16
171
300
55
20
21
6
N/A
N/A
102
29
8
9
46
N/A
35
12
47
20
215
400
62
25
21
6
N/A
N/A
114
29
9
9
47
N/A
35
13
48
21
230
500
71
30
32
7
N/A
N/A
140
31
10
10
51
N/A
40
14
54
25
270
600
76
32
32
7
N/A
N/A
147
31
10
10
51
N/A
40
14
54
26
278
700
80
32
32
7
N/A
N/A
151
31
10
10
51
N/A
40
14
54
26
282
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
Table D-2. Estimated Capital Cost(a) Breakdowns for Central Arsenic Removal Water Treatment Plants at Typical
Locations by Means of the Activated Alumina Process With Manual Operation, Replacement of Spent
Media, and With Process Water pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
8
8
N/A
3
48
21
8
6
35
N/A
34
9
43
13
139
100
31
7
10
8
N/A
4
60
27
8
6
41
N/A
37
10
47
15
163
200
38
13
15
9
40
4
119
29
10
7
46
N/A
48
12
60
23
248
300
55
20
24
10
40
5
154
35
13
11
59
N/A
58
13
71
29
313
400
62
25
24
10
40
6
167
35
15
11
61
N/A
58
14
72
30
330
500
71
30
36
11
40
6
194
43
17
13
73
N/A
62
15
77
35
379
600
76
32
36
11
40
6
201
43
17
13
73
N/A
62
15
77
36
387
700
80
32
36
11
40
6
205
43
17
13
73
N/A
62
15
77
36
391
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
78
-------
Table D-3. Estimated Capital Cost|a| Breakdowns for Central Arsenic Removal Water Treatment Plants at Typical
Locations by Means of the Activated Alumina Process With Manual Operation, Spent Media
Regeneration, and With Process Water pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
11
13
N/A
6
59
24
8
6
38
20
34
14
68
17
182
100
31
7
18
13
N/A
7
76
30
8
6
44
35
37
15
87
21
228
200
38
13
21
14
40
8
134
32
10
7
49
50
48
18
116
30
329
300
55
20
32
15
40
10
172
38
13
11
62
75
58
20
153
39
426
400
62
25
32
15
40
11
185
38
15
11
64
95
58
21
174
43
466
500
71
30
49
16
40
12
218
46
17
13
76
110
62
23
195
49
538
600
76
32
49
16
40
13
226
46
17
13
76
130
62
23
215
52
569
700
80
32
49
16
40
13
230
46
17
13
76
140
62
23
225
54
585
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
Table D-4. Estimated Capital Cost(a) Breakdowns for Central Arsenic Removal Water Treatment Plants at Typical
Locations by Means of the Activated Alumina Process With Automatic Operation, Spent Media
Regeneration, and Process pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
17
58
N/A
7
111
29
28
6
63
20
34
14
68
25
267
100
31
7
25
60
N/A
8
131
35
28
6
69
35
37
15
87
29
316
200
38
13
29
61
40
9
190
37
30
7
74
50
48
18
116
38
418
300
55
20
42
63
40
11
231
44
33
11
88
75
58
20
153
48
520
400
62
25
42
63
40
12
244
44
35
11
90
95
58
21
174
51
569
500
71
30
64
66
40
13
284
51
41
13
105
110
62
23
195
59
643
600
76
32
64
66
40
14
292
51
41
13
105
130
62
23
215
62
674
700
80
32
64
66
40
14
296
51
41
13
105
140
62
23
225
63
689
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
79
-------
Table D-5. Estimated Capital Cost(a) Breakdowns for Central Arsenic Removal Water Treatment Plants at Ideal
Locations by Means of the Activated Alumina Process With Manual Operation, Replacement of Spent
Media, and Without Process Water pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
7
4
N/A
N/A
40
17
3
0
24
N/A
3
0
3
7
74
100
31
7
9
4
N/A
N/A
51
22
3
5
30
N/A
3
0
3
9
93
200
38
13
13
5
N/A
N/A
69
23
4
7
34
N/A
3
0
3
11
117
300
55
20
21
6
N/A
N/A
102
27
5
7
39
N/A
4
0
4
15
160
400
62
25
21
6
N/A
N/A
114
27
6
7
40
N/A
4
0
4
16
174
500
71
30
32
7
N/A
N/A
140
29
7
8
44
N/A
5
0
5
19
208
600
76
32
32
7
N/A
N/A
147
29
7
8
44
N/A
5
0
5
20
216
700
80
32
32
7
N/A
N/A
151
29
7
8
44
N/A
5
0
5
20
220
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
Table D-6. Estimated Capital Cost(a) Breakdowns for Central Arsenic Removal Water Treatment Plants at Ideal
Locations by Means of the Activated Alumina Process With Manual Operation, Replacement of Spent
Media, and With Process Water pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
8
8
N/A
3
48
19
5
4
28
N/A
3
0
3
8
87
100
31
7
10
8
N/A
4
60
25
5
4
34
N/A
3
0
3
10
107
200
38
13
15
9
0
4
79
27
7
5
39
N/A
3
0
3
12
128
300
55
20
24
10
0
5
114
33
10
9
52
N/A
4
0
4
17
187
400
62
25
24
10
0
6
127
33
12
9
54
N/A
4
0
4
19
204
500
71
30
36
11
0
6
154
40
14
11
65
N/A
5
0
5
22
236
600
76
32
36
11
0
6
161
40
14
11
65
N/A
5
0
5
24
255
700
80
32
36
11
0
6
165
40
14
11
65
N/A
5
0
5
24
259
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
80
-------
Table D-7. Estimated Capital Cost(a) Breakdowns for Central Arsenic Removal Water Treatment Plants at Ideal
Locations by Means of the Activated Alumina Process With Manual Operation, Spent Media
Regeneration, and With Process Water pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
11
13
N/A
6
59
21
4
4
29
0
3
0
3
10
101
100
31
7
18
13
N/A
7
76
27
4
4
35
0
3
0
3
12
126
200
38
13
21
14
0
8
94
29
6
5
40
0
3
0
3
14
151
300
55
20
32
15
0
10
132
35
9
9
53
0
4
0
4
19
208
400
62
25
32
15
0
11
145
35
11
9
55
0
4
0
4
21
225
500
71
30
49
16
0
12
178
43
13
11
67
0
5
0
5
25
274
600
76
32
49
16
0
13
186
43
13
11
67
0
5
0
5
26
284
700
80
32
49
16
0
13
190
43
13
11
67
0
5
0
5
27
289
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
Table D-8. Estimated Capital Cost(a) Breakdowns for Central Arsenic Removal Water Treatment Plants at Ideal
Locations by Means of the Activated Alumina Process With Automatic Operation, Spent Media
Regeneration, and Process pH Adjustment (Multiply by $1,000)
Treatment Flowrate (gpm)
Process Equipment
Treatment Vessels
Treatment Media
Process Piping, etc.
Instrument and Controls
Chemical Storage Tanks
Chemical Pumps, Piping, etc.
Subtotal
Process Equipment Installation
Mechanical
Electrical
Painting and Miscellaneous
Subtotal
Misc. Installed Items
Wastewater Surge Tank
Building and Concrete
Site Work and Miscellaneous
Subtotal
Contingency 10%
Total
50
26
3
17
58
N/A
7
111
26
23
4
53
0
4
0
4
17
185
100
31
7
25
60
N/A
8
131
32
23
4
59
0
4
0
4
20
214
200
38
13
29
61
0
9
150
34
25
5
64
0
4
0
4
22
240
300
55
20
42
63
0
11
191
41
28
9
78
0
5
0
5
28
302
400
62
25
42
63
0
12
204
41
30
9
80
0
5
0
5
29
318
500
71
30
64
66
0
13
244
48
36
11
95
0
6
0
6
35
380
600
76
32
64
66
0
14
252
48
36
11
95
0
6
0
6
36
389
700
80
32
64
66
0
14
256
48
36
11
95
0
6
0
6
36
393
(a) August 2001 prices.
Note: Engineering, exterior utility pipe and conduit, wastewater and waste solids processing system, finance charges, real estate cost and taxes
not included.
81
-------
-------
Appendix E
Alternative Methods for Removing Media from Very Small System Tanks
1. Pressurized Canister
3. Inverter
Fabricate a special cap for the top of the adsorptive
media tank. Drill two holes in the cap approximately
1 inch in diameter. Screw the cap onto the top of the
tank. Attach a hose to each hole. Force raw water into
the tank through the first hose. Slowly lower the second
flexible plastic hose down through the other opening in
the cap to the top of the media level. Turn on the water
pressure so as to force media out of the second hose.
Pipe the water/media mixture to disposal barrels.
The depth of the escape pipe should be adjustable;
probably using a friction fitting through a rubber cap or
rubber washer. Movement capability ("wiggle") in the
vertical alignment of the escape pipe will allow media to
be removed from the lower sides of the media bed.
2. Industrial Wet/Dry Vacuum
Drain the water from the media. Hang a vacuum hose
from a support above the tank opening with the open
suction end hanging into the media. Vacuum out the
media. Remove media from vacuum compartment. For
this method, a high-powered motor/fan from an industrial
vacuum cleaner has been used, by mounting it on a
large barrel. When the first barrel was filled with media,
the motor/fan was remounted on the second barrel while
the first barrel was capped and made ready for pickup/
disposal.
Drain the water from the media. Construct a piece of
equipment out of 2-inch steel angles that is approxi-
mately one-half the height of the media tank. The media
tank should be strapped to the device and then inverted.
The media in the tank will partially fall out into a wide,
low-rise, pan. Use a hose stream to flush the inside of
the tank clean. Strain out the larger support gravel from
the flat pan and return it to the tank, along with new
adsorptive media, once the tank is replaced in an upright
orientation.
4. Gravity Discharge from a Sidewall Flange
With this process, a gate valve or bolted flange connec-
tion should be specified when the pressure tank was
being fabricated. The position of this fitting will be approx-
imately at the interface of the support media and adsorp-
tive media. When rebedding, the valve or flange is
opened and the media then falls, or is flushed, into a
low-rise decant tub, where the water and media are
separated. The media then is shoveled into the disposal
barrels. The media tanks must be elevated to allow the
decant tub to be placed below the outlet gate valve or
flange. A process water line should be mounted near the
top of the pressure tank to provide the wash water to
flush out the media. A small pump will be needed to
address the decant water.
83
-------
-------
Appendix F
English to Metric Conversion Table
English
Multiply by
Metric
Inch
Inch2
inch3
feet (ft)
ft2
ft3
gallon (gal)
gal
gal
grains (gr)
gr/ft3
pounds (Ib)
Ib/inches2 (psi)
Ib/ft2 (psf)
c/1 ,000 (gal)
0.0254
0.000645
0.000016
0.3048
0.0929
0.0283
0.2642
0.0038
0.0038
0.0649
2.2919
0.4545
0.00689
4.8922
0.2642
meter (m)
nf
m3
m
m2
m3
liter (L)
m3
kiloliter (kl_)
gram (g)
g/nf
kilogram (kg)
megapascals (MP)
kg/nf
c/1 ,000 L
85
------- |