United States September
Environmental Protection 1986
Agency
-
Design,
Construct, and
Operate for
-------
Disinfection with Ultraviolet
Introduction
The use of ultraviolet radiation for the disinfection of
wastewater is an effective and economical alternative
to chlorination and ozonation. Ultraviolet (UV)
disinfection, like chlorination, involves the addition of a
germicidal agent to a wastewater to inactivate the
bacteria. Like ozonation, disinfection with UV radiation
requires on-site generation of the germicidal agent. The
principal advantage of UV disinfection over chlorination
is that UV leaves no residual in the wastewater that
may affect the receiving waters.
As shown in Table 1, there are a number of UV
installations currently operating in the U.S. The list
reflects the distribution of treatment plants with regard
to size. The operating plants are predominantly small,
although those in the planning or construction stages
are larger.
Size (Design)
(m3/day) (MGD)
<380 (<0.1)
380-1900 (0.1-0.5)
1900-3800 (0.1-1.0)
3800-19000(1.5)
19000-38000 (5-10)
In
Operation
15
17
7
11
-
Under
Construction
—
10
5
14
-
38000-190000 (10-50) 1
>1 90000 (>50)
Total
Note: List compiled
-
51
Spring 1984.
1
30
Being
Designed
7
10
4
11
-
2
-
34
Table 1. Summary of UV Installations in the United States.
The Process
Disinfection by UV radiation relies on the transference
of ultraviolet electromagnetic energy from a lamp
source to an organism's cellular material to prevent cell
replication. The effectiveness of the radiation depends
on the dose and the organism's exposure time.
The most efficient and effective artificial source of UV
energy is the mercury lamp. It can be submerged in or
suspended above the wastewater. If submerged, it is
inserted into a quartz sleeve to minimize the cooling
effects of the water. Figure 1 is a schematic
representation of a UV disinfection unit showing a
submerged lamp placed perpendicular to the direction
of the wastewater flow. Lamps may also be placed
-------
The estimation of K and Np involves generating a set
of data specifically for this purpose during the design,.
phase of the project. The rate of bacterial inactivation
(K) is a function of intensity as shown on Figure 2.
Increasing the intensity of the UV radiation increases "
the rate, K. The coliform density associated with
particulates, Np, is a function of the suspended solids in
the wastewater. When these relationships are
established, the performance of the UV system can be
approximated for different equipment configurations and
process variations.
Design
The degree of pretreatment received before the !
disinfection step affects the sizing and performance of
the UV system. The five wastewater parameters that
most affect UV design and performance are flow (Q),
suspended solids (ss), initial bacteria density (N0),
density of bacteria associted with particulates (Np), and
the UV absorbence of the wastewater. r
The performance of a UV disinfection system is directly
related to the initial density (N0) of the indicator
organisms . Performance is measured as the log of the;
survival ratio, N/N0 (see Figure 2). The initial density
cannot be predicted based on the type of pretreatment
received and should be measured prior to design.
The aggregation of bacteria and particulates, expressed
as Np, significantly affects the efficiency of UV
disinfection. The level of Np is related to the suspended
solids content. !
alt,
watts/cm)
to the initial density (N0) of the indicator organisms.
I for a given residence time.
-------
The UV "demand" of the chemical constituents of the
wastewater must also be measured since this
absorbence will affect the intensity of the radiation
within the reactor. In specific design situations, the level
of absorbence will affect the sizing of a system and
possibly the configuration of the lamps.
Design Considerations
• Low pressure mercury arc lamps are currently the
most efficient source of UV radiation.
• Temperature effects should be minimized. In quartz
systems, O-ring spacers should be slipped over the
lamps to prevent direct contact with the cooler quartz
sleeve.
• Fittings holding the quartz sleeve should be tight and
leakproof to prevent fouling on the inside of the
sleeve.
• The control panel should be remote from the UV
reactor.
• The ballasts must be properly mated with the lamps
and thermally protected to shut down if they
overheat.
• The electrical wiring should be properly sized and be
resistant to UV radiation effects.
• Large debris should be prevented from entering the
UV system.
• A feature that has been found not to be successfully
applied is the use of microprocessor controlled
automatic lamp bank shut-off systems. Many plants
that originally had automatic systems have
since reverted to manual control. The use of a simple
mechanical cam timer that shuts off selected lamps
or banks of lamps to reduce UV output during
selected time periods is recommended.
Operational Considerations
• Cleaning
Periodic chemical or detergent cleaning of surfaces
that come in contact with the wastewater is required,
particularly when the wastewater has a high oil and
grease content or has a high hardness content.
Magnesium and calcium carbonate deposits are a
major cause of fouling of quartz surfaces. However,
acidification of the reactor water will usually restore
the surface. Organic fouling by oil and grease must
-------
rate for Success
N = (N'0 + Np) exp (-Kit) + Nf
(2)
where: N'0 = initial, non-aggregate density
(conforms per 100 ml)
Np = coliform density associated with
particulates (coliforms per 100 ml)
Generally N'0»NP, and N0 = N0' + Np; therefore.
Equation 2 can be rewritten as Equation 3.
N = N0 exp (-Kit) + Np
(3)
Equation 4 incorporates the dispersive properties of a
reactor under steady-state conditions.
N = N0 exp
N
(4)
where: z = 1 + 4KE/U2
x = length of reactor, cm
E = dispersion coefficient, cm2/sec
K = rate of bacteria I inactivation, sec;"1
u = velocity of wastewater, cm/sec '
where: u = xVv"1CT1
Vv = void volume of reactor, liters
Q = wastewater flow, liters/sec
I = Intensity
I I1>I2>I3
Time
Figure 2. The performance of a UV disinfection system as
Note: The rate K increases with increasing intensil
-------
be cleaned with detergents or a combination of
cleaning methods.
Accessory equipment is available to maintain the
surfaces. These include the mechanical wiper,
ultrasonic transducer, or a higher pressure spray
nozzle. While these methods are somewhat effective,
intermittent cleaning with chemicals is generally
required.
• Elements for Effective Maintenance
The reactor should be designed with drains that
allow complete and rapid emptying.
The system should be modularly designed to allow
isolation of any unit from the plant flow.
Strict inventories of lamp use and output should be
maintained.
The reactor should be drained when removed from""
service.
> Labor
The labor requirements of the UV system are divided
into three main categories: direct UV operation and
maintenance tasks, general maintenance, and
system overhaul. Overall, the total labor needs for the
UV process are relatively low, ranging from
approximately 40 persondays/year for a small 10 KW
1000
Approximate Number of Lamps (1.5 ARC)
60 120 300 600 1200 3000 6000
Note:
Labor Based on
Total System KW
40 60 100 200 4006001000
Figure 3. Estimate of Labor Requirements for the
Operation and Maintenance of UV Systems
Scheible, O. K., et. al., 1986.
-------
|ht - Design, Construct, and Ope
Influent
Control Box
UV Lamp Inside
Quartz Sleeve
Effluent
Figure 1. Ultraviolet Design Schematic.
parallel to the flow in some UV units. The intensity of
the radiation emitted by the lamp dissipates as the
distance from the lamp increases.
Process Model - UV Reactor Performance
A mathematical model was developed during a study2
at Port Richmond, NY to account for major process
variables in the design of a UV system. The derivation
of this expression is briefly presented here. Inactivation
of bacteria by UV radiation can be approximated by
the following equation:
N = N0 exp (-Kit)
0)
where: N = effluent bacterial density (coliforms per
100 ml)
N0 = influent bacterial density (coliforms
per 100 ml)
K = rate constant (cm2watT1sec~1)
I = intensity watt(cm2)"1 i
t = time of exposure (sec) ;;
Equation 1 is a good first approximation of a response"
to a given dose. However, direct testing on mixed |
cultures often shows a reduced efficiency with
increasing dose. In wastewater treatment, this has
been attributed to the aggregation of bacteria and
particulate matter. Ultraviolet light is unable to penetrate
this material to inactivate the bacteria; thus, the
continued elevation of dose will show a diminishing
response as residual active bacteria are protected in '
the particulates. Consequently, Equation 1 should be
modified (shown as Equation 2) to account for the
effect of particulate matter in the wastewater.
-------
(120 lamps) system to approximately 40
persondays/year for a 400 KW (5000 lamps) system.
Total yearly estimated labor requirements are
presented in Figure 3.
Summary
Studies have found UV to be very effective for
disinfection of secondary or tertiary effluent. In addition,
the studies showed quartz systems to be generally
more energy efficient than teflon systems.
Also, the efficiency of both systems drop significantly
as turbulence and suspended solids levels increase
above secondary levels.
The UV process is relatively simple and also offers the
advantages of system flexibility and capability of
responding to changes in demand. Since the process
leaves no residual in the wastewater that could impact
the receiving water, less rigorous control is necessary
than that associated with chlorine. In addition, a
mathematical model recently developed will aid in the
design of future UV disinfection systems.
Additional Reading
1. U.S. Environmental Protection Agency, Technology Transfer
Process Design Manual for Municipal Wastewater Disinfection,
U.S. Environmental Protection Agency, CERI, Cincinnati, OH, 1986.
2. Scheible, O. K., et. al., "Ultraviolet Disinfection of Wastewaters
from Secondary Effluent and Combined Sewer Overflows."
EPA-600/2-86-005, U.S. Environmental Protection Agency, WERL,
Cincinnati, OH, 1986.
3. White, S. C., et. al., "A Study of Operational Ultraviolet
Disinfection Equipment at Secondary Treatment Plants." Journal
of the Water Pollution Control Federation, March 1986, Vol. 58,
pp. 181-192.
Prepared by Environmental Resources Management, Inc.
For additional information contact:
EPA-OMPC (WH-595) EPA-WERL (489)
401 M Street, SW 26 West St. Clair Street
Washington, DC 20460 Cincinnati, OH 45268
(202) 382-7368/7369 (513) 569-7688
EPA Region 1
John F. Kennedy Federal BuMng
Boston, MA 02203
EPA Region 2
26 Federal Plaza
New York, NY 10278
EPA Region 3
841 Chestnut Street
Philadelphia, PA 19107
EPA Region 4
345 Courtland Street, NE
Atlanta, GA 30365
EPA Region 5
230 South Dearborn Street
Chicago, IL 60604
EPA Region 6
1201 Elm Street
Dallas, TX 75270
EPA Region 7
726 Minnesota Avenue
Kansas City, KS 66101
EPA Region 8
999 18th Street
Denver, CO 80202
EPA Region 9
215 Fremont Street
San Francisco, CA 94105
EPA Region 10
1200 6th Avenue
Seattle, WA 98101
------- |