Water and Wastewater Engineering
Подождите немного. Документ загружается.
26-1
26-5 CARBON ADSORPTION
26-6 CHAPTER REVIEW
26-7 PROBLEMS
26-8 REFERENCES
TERTIARY TREATMENT
26-1 INTRODUCTION
26-2 CHEMICAL PRECIPITATION
OF PHOSPHORUS
26-3 GRANULAR FILTRATION
26-4 MEMBRANE FILTRATION
26
CHAPTER
26-2 WATER AND WASTEWATER ENGINEERING
26-1 INTRODUCTION
The need for treatment of wastewater beyond that which can normally be accomplished in sec-
ondary treatment is based on the recognition of one or more of the following:
1 . Increasing population pressures result in increasing loads of organic matter and sus-
pended solids to rivers, streams, and lakes.
2. The need to increase the removal of suspended solids to provide more efficient disinfection.
3. The need to remove nutrients to limit eutrophication of sensitive water bodies.
4. The need to remove constituents that preclude or inhibit water reclamation.
Initially, in the 1970s, these processes were called “ad
vanced wastewater treatment” because
they employed techniques that were more advanced than secondary treatment methods. In the
last three decades many of these technologies have either been directly incorporated into the
second ary processes, for example nu
trient removal, or they are so inherent in m eeting stringent
discharge standards that they have become c onventional. These processes include chemical
precipitation, granular filtration, membrane filtration, and carbon adsorption. As conventional
processes , they are better termed tertiary treatment processes rather than an advanced treatment
process. In current practice, the employment of air stripping, ion exchange, NF or RO treatment,
and other similar processes to meet water quality requirements is correctly termed advanced
wastewater treatment. A dvanced wastewater treatm ent technologies are, fund amentally, those
employed to treat water for reuse.
The di
scussion in this chapter focuses on tertiary treatment processes: chemical precipitation,
granular filtration, membrane filtration, and carbon adsorption. The emphasis in this chapter is on
the application of these technologies in tertiary treatment of wastewater.
26-2 CHEMICAL PRECIPITATION OF PHOSPHORUS
Because phosphorus is a c ritical element in the promotion of eutrophication, restrictions on
discharge concentrations are establis hed for many NPDES permits. Before the development
of biological phosphorus removal (BPR) technology, chemical precipitation was the primary
means of removing phosphorus. In many c
ases, it is still the only practical method of achieving
standards because of space or economic constraints. In addition, it is often provided in BPR
plants as a prudent backup in case of process upset or because stringent standards cannot be met
with BPR alone.
The theory of pho
sphorus precipitation and design strategies are discussed in this section.
The design of mixing systems and settling tanks are discussed in Chapters 6 and 10.
Theory
All polyphosphates (molecularly dehydrated phosphates) gradually hydrolyze in aqueous solu-
tion, and revert to the ortho * form
(PO
4
3
)
from which they were derived. Phosphorus is typically
found as monohydrogen phosphate
(HPO
4
2
)
in wastewater.
* Ortho i s the terrm used to designate the highest degree of hydration of the salt. For benzene rings, ortho refers to the number
2 and 6 positions.
TERTIARY TREATMENT 26-3
The removal of phosphorus to prevent or reduce eutrophication is accomplished by chemi-
cal precipitation using one of three compounds. The precipitation reactions for each are shown
below.
U sing ferric chloride:
FeCl HPO FePO H Cl
3 4
2
4
3
T
(26-1)
U sing aluminum sulfate:
Al SO 2HPO AlPO H SO
243 4
2
44
2
223()
T
(26-2)
U sing lime:
5 336
24
2
5 4 3 2
Ca OH HPO Ca PO OH H O OH() ( )
T
(26-3)
Ferric chloride and alum reduce the pH while lime increases it. The effective range of pH for
alum and ferric c hloride is between 5.5 and 7.0. If there is not enough naturally occurring
alkalinity to buffer the system to this range, then lime must be added to counteract the forma-
tion of H
.
Example 26-1. If a wastewater has a soluble orthophosphate concentration of 4.00 mg/L as P,
what theoretical amount of ferric chloride will be required to remove it completely?
Solution:
a . From Equation 26-1 , note that one mole of ferric chloride is required for each mole of
phosphorus to be removed. The pertinent gram molecular weights are as follows:
FeCl
3
162.2 g/mole
P 30.97 g/mole
b. With a PO
4
-P concentration of 4.00 mg/L, the theoretical amount of ferric chloride is
()400
162 2
3097
2.
.
.
mg/L
g/mole
g/mole
⎛
⎝
⎜
⎞
⎠
⎟
0095 21 0..or mg/L
Comment. Because of side reactions, solubility product limitations, and day -to-day variations,
the actual amount of chemical to be added must be d etermined by jar tests on the wastewater.
Design Strategies
Selection of Chemicals. The principal chemicals used in precipitating phosphorus are alum
(Al
2
(SO
4
)
3
· 14 H
2
O) and ferric chloride. Lime use has been sharply curtailed over the last few
decades because of the substantial increase in sludge production, pH control requirements,
26-4 WATER AND WASTEWATER ENGINEERING
and operation and maintenance problems with handling, storage and feeding. Alum and ferric
chloride present storage and handling issues because they are corrosive. Because of numerous
side reactions, the actual dose of alu m or ferric chloride to achieve a high degree of removal is
significantly larger than the
stoichiometric dose. Table 26-1 summarizes some actual experience
for alum. Actual dosages will be determined by the operator based on jar test results. From a
design perspective, a conservative dose estimate and a large turndown ratio for the feed equip-
ment is re
commended.
The actual ferric chloride dose will be 1.5 to 3 times the theoretically calculated amount for
90 percent removal.
Polymers may be added to enhance settling of the precipitate.
Preprecipitation. The addition of a precipitating chemicals upstream of the primary settling
tank is c
alled preprecipitation. The influent structures of the primary tank mix the chemicals
with the wastewater. The primary tank serves as both the reaction basin and the settling basin for
the precipitant. The precipitated phosphate is removed with the primary sludge. This improves
the efficiency of suspended
solids removal in the primary tank but may deprive the biological
processes of needed nutrients.
Coprecipitation. The addition of precipitating chemicals that are removed with the biological
sludge in the secondary clarifier is called coprecipitation. They may be added in the effluent
from
the primary clarifier, the return mixed liquor, or the efflu ent from the biologic al treatm ent
process before the secondary clarifier.
When ferric c hloride and alum are used, the chemicals may be added directly to the
aeration tank in the activated sludge system. Thus, the aeration tank serves as
a reaction basin.
The precipitate is then removed in the secondary clarifier. This is not possible with lime
because the high pH required to form the precipitate is detrimental to the activated sludge
organisms.
Postprecipitation. The addition of precipitating chemicals after secondary clarification is
called postprecipi
tation. This arrangement requires separate mixing and settling facilities and/or
filtration.
Sludge Production. The chemical precipitation of phosphorus will significantly increase the
sludge production from the facility.
Source: U.S. EPA, 1976.
Phosphorus
reduction, %
Alum:P
weight ratio
Al:P
weight ratio
75%13:1 1.2:1
85% 16:1 1.5:1
95% 22:1 2.0:1
TABLE 26-1
Typical alum dosages to achieve various levels
of phosphorus reduction
TERTIARY TREATMENT 26-5
Hints from the Field. Experience from the field has revealed the following sugggestions when
chemical precipitation is to be employed:
• The quality of the metals salts is an important consideration in selecting a supplier. For
example, FeCl
3
from waste pickling liquor may contain high concentrations of toxic
chemicals such as chromium. Alum may contain mercury. These contaminants may cause
an NPDES violation for the discharged wastewater or limit the ability to land apply the
biosolids.
• Ferrous salts fed in the primary clarifier will be oxidized in the aeration tank. This may
have adverse effects on fine bubble diffusers or membrane diffusers.
26-3 GRANULAR FILTRATION
Filtration Objectives and Performance
Filtration is used when the effluent limit for total suspended solids (TSS) is equal to or less than
10 mg/L. Average day effluent concentrations that filtration can achieve for secondary settled
effluent are shown in Table 26-2 .
Because a fraction of the TSS is biomass, and because a fraction of the biomass is biodegradable,
rem
oval of TSS will reduce the effluent BOD. BOD concentrations in the range of 4 to 10 mg/L
may be achieved. In addition, the use of filtration in combination with chemical coagulation can
reduce the effluent
PO
4
3
concentration to 0.1 mg/L (WEF, 1998). It is also possible to combine
nitrate removal with filtration. Up to 90 percent NO
3
-N can be removed (Savage, 1983).
Filtration Technologies
The five types of granular filters commonly used for wastewater filtration are (1) conventional
downflow filters, (2) deep-bed downflow filters, (3) deep-bed upflow continuous-backwash
filters, (4) pulsed-bed filters, and (5) traveling-bridge filters. The deep-bed upflow, pulsed-bed,
and traveling-bridge filters are proprietary. Because the
design details for the proprietary filters
are supplied by the manufacturer, the following discussion is limited to conventional downflow
and deep-bed downflow filters.
Without chemical
coagulation
With tertiary chemical
coagulation
Filter influent Effluent TSS, mg/L Effluent TSS, mg/L
Conventional activated sludge 3–10 0–5
Extended aeration 1–5 0–5
High-rate trickling filter 10–20 0–3
Two-stage trickling filter 6–15 0–3
A dapted from WEF, 1998.
TABLE 26-2
Typical average day effluent concentrations from granular media filtration
of secondary effluent
26-6 WATER AND WASTEWATER ENGINEERING
Although pressure filters are common for industrial applications, in practice their use for
municipal application is not common. They are not included in this discussion.
Design Practice
Process Train. Typically, filtration is used to remove residual biological floc from a secondary
settling tank effluent.
Pretreatment. Good design practice is to provide for the capability to add inorganic or organic
coagulants both upstream of the sedimentation tank that precedes the filter and to the filter influent.
Typical do
sages of polyelectrolyte are 0.5 to 1.5 mg/L to the settling tank influent and/or 0.05 to
0.15 mg/L to the filter influent. Dosages are determined by the operator based on jar tests.
If the average influent TSS to the filter is anticipated to be greater than 20 mg/L, upstream
pretreatment consisting of coagulation, flocculation, and sedimentation, or flotation, is required
to achieve a TSS le
ss than 3 mg/L (Metcalf & Eddy, 2003).
Filter Type. Most wastewater filters in the United States are d ownflow, dual-media or multi-
media units (WEF, 1998). Single-medium stratified beds are no longer designed for municipal
wastewater applications because of their unfavorable headloss buildup characteristic
s.
Number and Size. As in drinking water filters, multiple filter units are used to allow continuous
operation during backwashing. The design guidance given in Chapter 11 is appropriate for
selecting the number and size of units.
Filtration Rate and Terminal Headloss. Typical filtration rates range from 5 to 20 m/h with
terminal headlosses of 2.4 to 3 m.
Underdrains, Backwashing, and Wash Troughs/Gullet. The design guid
ance given in Chap-
ter 11 is appropriate for selecting the number and size of units.
Media. A summary of the guidance for the various media arrangements is given in the following
tables.
TABLE 26-3
Design criteria for dual-media filters used in tertiary treatment of wastewater
Parameter
Reported
range Typical
GLUMRB
recommendation
Anthracite coal on top
Effective size 0.8–2.0 mm 1.3
Uniformity
coefficient 1.3–1.6 1.5 1.7
Shape factor (f)
0.40–0.60
Porosity 0.56–0.60
Specific gravity 1.4–1.75
Depth of medium 360–900 mm 720 mm
(continued)
TERTIARY TREATMENT 26-7
(continued)
TABLE 26-4
Design criteria for trimedia filters used in tertiary
treatment of wastewater
Parameter
Reported
range Typical
Anthracite coal on top
Effective size 1.0–2.0 mm 1.4 mm
Uniformity
coefficient 1.4–1.8 1.5
Shape factor (f)
0.40–0.60
Porosity 0.56–0.60
Specific gravity 1.4–1.75
Depth of medium 240–600 mm 480 mm
Sand in middle
Effective size 0.4–0.8 mm 0.5 mm
Uniformity
coefficient 1.3–1.8 1.5
Shape factor (f)
0.7–0.8
Porosity 0.4–0.46
Specific gravity 2.55–2.65
Depth of medium 240–480 mm 300 mm
a
A ctual off-line time will be 30 min because of the time required to drain the filter and
gradually increase to the full backwash rate. An additional 30–40 minutes off-line is required for
“filter-to-waste” to clear the bed of wash water and dislodged turbidity. If air scour is provided,
the time will be even longer becaus
e of the necessity of sequencing the air scour and wash
water.
Sources: Cleasby and Logsdon, 1999; GLUMRB, 2004; Metcalf & Eddy, 2003; WEF, 1988.
Sand on bottom
Effective size 0.4–0.8 mm 0.65
Uniformity
coefficient 1.2–1.6 1.5 1.7
Shape factor (f)
0.7–0.8
Porosity 0.40–0.47
Specific gravity 2.55–2.65
Depth of medium 180–360 mm 360 mm
Filtration rate 5–24 m/h 12 m/h
Backwash rate 48–72 m/h
Backwash duration
a
10–20 min
Surface wash rate
Revolving arms 1.2–2.4 m/h
26-8 WATER AND WASTEWATER ENGINEERING
TABLE 26-5
Design criteria for deep-bed monomedium filters used in tertiary
treatment of wastewater
a
A ctual off-line time will be 30 min because of the time required to drain the filter and
gradually increase to the full backwash rate. An additional 30–40 minutes off-line is required for
“filter-to-waste” to clear the bed of wash water and dislodged turbidity. If air scour is provided, the
time will be even longer becaus
e of the necessity of sequencing the air scour and wash water.
Sources: Cleasby and Logsdon, 1999; GLUMRB, 2004; Metcalf & Eddy, 2003; WEF, 1998.
Parameter
Reported
range of values Typical
Anthracite coal
Effective size 2–4 mm
Uniformity
coefficient 1.3–1.8 1.5
Shape factor (f)
0.40–0.60
Specific gravity 1.4–1.75
Porosity 0.56–0.60
Depth of medium 900–2,100 mm 1,500 mm
Filtration rate 5–24 m/h 12 m/h
Backwash rate 37–45 m/h
Backwash duration
a
15 min
Surface wash rate
Revolving arms 1.2–2.4 m/h
TABLE 26-4 (continued)
Design criteria for tri-media filters used in tertiary treatment of wastewater
a
A ctual off-line time will be 30 min because of the time required to drain the filter and gradually
increase to the full backwash rate. An additional 30–40 minutes off-line is required for “filter-to-waste”
to clear the bed of wash water and dislodged turbidity. If air scour is provided, the time will be even
longer becaus
e of the necessity of sequencing the air scour and wash water.
Sources: Cleasby and Logsdon, 1999; GLUMRB, 2004; Metcalf & Eddy, 2003; WEF, 1998.
Parameter
Reported
range Typical
Garnet on bottom
Effective size 0.20–0.6 mm 0.35 mm
Uniformity coefficient 1.5–1.8 1.5
Shape factor (f)
0.60–0.80
Porosity 0.42–0.55
Specific gravity 3.6–4.3
Depth of medium 50–150 mm 100 mm
Filtration rate 5–24 m/h 12 m/h
Backwash rate 48–72 m/h
Backwash duration
a
10–20 min 15 min
Surface wash rate
Revolving arms 1.2–2.4 m/h
TERTIARY TREATMENT 26-9
Denitrification Filters. Coarse-media deep-bed denitrification has been in practice for over
30 years. Denitrification occurs when the filter grains serve as a medium for attached growth
of denitrifying organisms. The filter operates anaerobically. A carbon source must be supplied.
Typically, methanol, at a dosage of approximately
3 mg/L per mg/L of NO
3
-N is used (Pickard
et al., 1985).
Operation of denitrification filters is similar to that for typical granular filters except for the
requirement to release nitrogen (called “bumps”). This is required because the nitrogen from
denitrification accumulates in the filter. This causes an increase in headloss. This headloss i
s
relieved by backwashing with water only for approximately 1 to 5 minutes. This backwash water
is not captured, and the backwash is not intended to clean the filter. The number of denitrification
bumps varies from 4 or 5 times per day, up to 14 to 16 times per day. It is a function of nitrate
load
ing, media, and underdrain type.
The design of the filter is based on empty bed contact time (EBCT). Typically, this is about
20 minutes for warm water (20 C) to about 60 minutes for cold water (10 C). The beds are about
1.8 m deep and are loaded at an average of 1.7 to 5 m/h (WEF, 1998). Typical
design parameters
are shown in Table 26-6 .
Hint from the Field. Experience has revealed that it is not necessary to fluidize a monome-
dium filter. However, the air scour must be fine tuned.
TABLE 26-6
Typical design parameters for denitrification filters
(continued)
Parameter Range Typical Comment
Sand monomedia
Effective size (sand) 1.8–2.3 mm 2.3 2.3 mm is the largest commercially available,
but larger media may be better
Uniformity coefficient 1.3
Sphericity 0.8–0.9 0.82 0.9 is preferred; less spherical is effective but
requires more frequent backwa
shing and bumps
Depth 1.2–2 m 1.6 m
Dual media
Effective size
Coal 2.38–3.65 mm 3.65 mm
Sand 1.8–2.3 mm 2.3 mm
Depth
Coal 0.3–0.9 m
0.6 m
Sand 0.9–1.2 m 1.2 m
General
Empty bed contact time 20–60 min 20 min Longer for cold water (10C) than warm water
(20C)
Hydraulic loading
20C 60–120 m/d 100 m/dm
/d m
3
/m
2
· d
10C 30–90 m/d 80 m/d
26-10 WATER AND WASTEWATER ENGINEERING
TABLE 26-6 (continued)
Typical design parameters for denitrification filters
Sources: Metcalf & Eddy, 2003; U.S. EPA, 1975; WEF, 1998.
26-4 MEMBRANE FILTRATION
Low-pressure microfiltration (MF) and ultrafiltration (UF) membranes are used to provide ter-
tiary treatment for effluent from municipal wastewater treatment plants . Although they may be
used instead of granular filtration, currently the largest use is to pretreat secondary effluent to
facilitate further treatment by reverse osmo
sis before aquifer recharge or indirect potable reuse.
Membrane filtration theory, properties of MF and UF membranes, and fundamental aspects
of the design of membranes for filtration are discussed in Chapter 12. This discussion is focused
on those design elements particular to wastewater practice.
Membrane Performance
A s shown in Table 26-7 , filtration with MF/UF membranes results in a high-quality effluent.
Because MF and UF membranes act as physical barriers, their removal efficiency for conventional
pollutants is dependent on the fraction of the pollutant that is associated with suspended matter.
No removal can be expecte
d in the absence of biological and/or chemical treatment that results
in the formation of floc.
Feed Water Quality
To achieve the performance shown in Table 26-7 , the influent to the MF/UF filter must, at a
minimum, meet the standards for secondary effluent, that is, BOD
5
30 mg/L, TSS 30 mg/L,
and fecal coliforms (FC) 200/100 mL (WEF, 2006).
Pretreatment
E xperience has shown that pretreatment of secondary effluent prior to tertiary MF/UF filtration is
essential to optimize the membranes’ performance. Pretreatment may include chemical coagula-
tion, chlorination or chloramination, screening with strainers, and flow equalization.
Parameter Range Typical Comment
NO
3
-N loading
20C 1.4–1.8 kg/m
2
· d 1.6 kg/m
2
· d
10C 0.8–1.2 kg/m
2
· d 1.0 kg/m
2
· d
Methanol to NO
3
-N ratio 2.0–3.5 3.0 Units are mg/L methanol per mg/L NO
3
-N
Backwash
Water 15–25 m/h 20 m/h m/h m
3
/m
2
· h. Duration is about 15 min
Air 19–120 m/h 100 m/h Duration is about 20–40 s
Nitrogen release (bump)
Water only 10–14 m/h 12 m/h Introduction of air scour inhibits denitrification
Duration 2–15 min 5 min
interval 1 to 6 h 2 h