Appendix
B
Ozone
CT
Methods
B.
1
INTRODUCTION
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
B.
1.1
BACKGROUND
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
B.
2
SELECTION
OF
METHODS
FOR
CALCULATING
INACTIVATION
CREDIT
.
.
.
.
.
.
4
B.
3
OZONE
CONTACTOR
CONFIGURATIONS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6
B.
4
EXTENDED­
CSTR
APPROACH
FOR
OZONE
CONTACTORS
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9
B.
4.1
INTRODUCTION
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9
B.
4.2
OVERVIEW
OF
SYSTEM
EVALUATION
AND
MONITORING
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9
B.
4.3
EXTENDED­
CSTR
APPROACH
­
OZONE
CONTACTORS
WITHOUT
A
TRACER
TEST
9
B.
4.3.1
CLASSIFICATION
OF
THE
CHAMBERS
AND
CONTACTOR
ZONES
.
.
.
.
.
.
.
.
.
.
10
B.
4.3.2
CALCULATING
LOG
INACTIVATION
ACROSS
AN
EXTENDED­
CSTR
ZONE
.
.
12
B.
4.3.2.1
Determining
the
Value
of
k*
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
13
B.
4.3.2.2
Determining
the
Value
of
C
in
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
15
B.
4.3.2.3
Quality
Assurance
for
Extended­
CSTR
Calculations
.
.
.
.
.
.
.
.
.
.
.
.
16
B.
4.4
EXAMPLE
OF
EXTENDED­
CSTR
APPLICATION
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
Example
­
Extended­
CSTR
Approach
for
a
Multi­
chamber
Contactor
With
In­
situ
Sample
Ports
and
One
Dissolution
Chamber
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
REFERENCES
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
22
Abbreviations
 
Molar
absorbance
expressed
as
M­
1
cm­
1.
BrO
3­
Bromate
ion
C
eff
T
10
Chamber
effluent
ozone
residual
in
mg/
L
times
chamber
T
10
time
in
minutes
Co­
current
A
chamber
in
an
ozone
contactor
where
the
water
is
flowing
upward
and
the
chamber
ozone
gas
bubbles
are
rising.
The
direction
of
flow
of
the
water
and
the
gas
is
the
same.
Counter­
current
chamber
A
chamber
in
an
ozone
contactor
where
the
water
is
flowing
downward
and
the
ozone
gas
bubbles
are
rising.
The
direction
of
flow
of
the
water
is
in
the
opposite
direction
of
the
gas
flow.
CSTR
Completely
Stirred
Tank
Reactor
­
fully
mixed
volume
CT
The
product
of
Concentration
and
Time
in
mg/
L­
min
DBP
Disinfection
byproduct
Half­
life
or
HL
The
time
that
it
takes
for
the
ozone
residual
to
decrease
by
50%.
It
is
calculated
as:
,
where
k*
=
first­
order
ozone
decay
coefficient
(
)
*
0.5
Ln
HL
k
=

HDT
Hydraulic
detention
time
calculated
as
the
volume
divided
by
the
flow.
When
volume
is
expressed
in
gallons,
and
flow
expressed
in
gallons/
minute,
then
the
calculated
HDT
is
in
minutes
In­
situ
sample
ports
Sample
ports
that
take
a
sample
from
the
flow
of
the
chamber,
typically
through
tubing
that
projects
into
the
flow
k*
The
first­
order
ozone
decay
coefficient,
min­
1.
k
10
Log­
base­
10
value
of
the
lethality
coefficient
for
the
inactivation
of
Cryptosporidium,
Giardia
or
virus
with
ozone.
The
units
of
k
10
in
this
document
are
L/
mg­
min.­
Log
(
I/
I
o)
Log
inactivation.
Negative
log­
base­
10
of
the
survival
rate
(
N/
N
o)
of
the
microorganisms,
where
I
o
is
the
number
of
viable
organisms
entering
the
contactor,
and
I
is
the
number
of
viable
organisms
leaving
the
contactor.
Q
Water
flow
­
usually
expressed
in
gallons
per
minute
(
gpm)
or
million
gallons
per
day
(
MGD)
Up
flow
chamber
A
chamber
within
an
over­
under
baffled
bubble­
diffuser
ozone
contactor
in
which
the
direction
of
water
flow
is
upward.
V
Volume
of
the
contacting
zone
in
question
­
usually
expressed
in
gallons
or
million
gallons.

B.
1
Introduction
B.
1.1
Background
Appendix
O
of
the
Surface
Water
Treatment
Rule
(
SWTR)
Guidance
Manual
(
USEPA
1991)
includes
a
description
of
different
methods
for
determining
inactivation
credit
using
an
ozone
contactor.
These
methods
differ
in
the
level
of
effort
associated
with
them
and,
in
general,
the
ozone
dose
needed
to
achieve
a
given
level
of
inactivation.
This
appendix
provides
guidance
to
help
water
systems
select
the
more
appropriate
methods
for
their
ozone
process.
More
importantly,
it
builds
on
the
information
presented
in
the
SWTR
Guidance
Manual
with
detailed
descriptions
of
the
extended
continuous
stirred
tank
reactor
(
CSTR)
method.
Appendices
D
and
E
compliment
this
appendix
with
descriptions
of
ozone
residual
sampling
and
laboratory
analysis
(
Appendix
D)
and
derivations
of
equations
used
in
the
extended
CSTR
and
SFA
approaches
(
Appendix
E).

The
three
methods
for
calculating
LT2ESWTR
ozone
inactivation
credit,
presented
in
Chapter
11
and
this
appendix,
are
described
below.

1.
T10
­­
calculates
CT
through
a
contactor
assuming
hydraulic
conditions
similar
to
plug
flow
and
can
be
used
with
or
without
tracer
study
data.
T
10
is
the
time
it
takes
for
90
percent
of
the
water
to
pass
the
contactor.
Even
in
well­
baffled
contactors,
the
T
10
is
most
often
less
than
65
percent
of
the
average
hydraulic
detention
time
(
HDT)
through
the
contactor,
and
generally
underestimates
the
true
CT
achieved.
(
The
T
10
approach
is
described
in
Chapter
11,
section
11.3.)

2.
CSTR­­
calculates
log
inactivation
credit
using
hydraulic
detention
time.
It
is
applicable
to
contactors
that
experience
significant
back
mixing
or
when
no
tracer
study
data
are
available.
EPA
recommends
using
this
method
(
or
the
Extended
CSTR)
when
no
tracer
study
data
are
available.
(
The
CSTR
approach
is
described
in
Chapter
11,
section
11.3.)

3.
Extended
CSTR­­
a
combination
of
the
CSTR
and
SFA
approaches.
It
utilizes
the
hydraulic
detention
time
for
the
contact
time
and
incorporates
the
ozone
decay
rate
to
calculate
concentration.
It
is
not
applied
to
chambers
into
which
ozone
is
introduced.

While
this
guidance
manual
describes
three
methods,
other
methods
or
modifications
to
these
methods
may
be
used
at
the
discretion
of
the
State.
A
fourth
method,
the
Segmented
Flow
Analysis
approach,
is
under
consideration
by
EPA,
but
the
details
of
the
approach
are
not
final.
EPA
is
requesting
comment
on
the
approach
and
any
appropriate
safety
factors
to
ensure
the
inactivation
credit
calculated
using
the
method
is
actually
achieved.

B.
2
Selection
of
Methods
for
Calculating
Inactivation
Credit
Selecting
the
appropriate
methods
to
use
depends
on
the
configuration
of
the
ozone
contactor
and
amount
of
process
evaluation
and
monitoring
that
a
water
system
is
willing
to
undertake.
It
is
also
possible
that
combinations
methods
can
be
used.
For
contactors
with
multiple
segments
it
is
likely
that
the
CT
of
one
or
two
segments
would
be
calculated
using
either
the
T
10
or
CSTR
methods,
while
the
CT
for
the
remaining
segments
would
be
calculated
with
the
Extended­
CSTR.

Of
the
three
methods
described
in
the
previous
section,
the
Extended
CSTR
is
the
most
complex
method.
The
Extended­
CSTR
approach
requires
measurements
of
the
ozone
concentration
at
a
minimum
of
three
points
within
the
contactor
to
develop
a
predicted
ozone
concentration
profile
through
the
contactor.
The
contact
time
is
based
on
the
hydraulic
detention
time
of
the
contactor
and
an
assumption
of
completely
mixed
flow.
While
many
mathematical
principles
are
discussed
in
these
methods,
their
implementation
is
fairly
straightforward.
In
fact,
the
methods
presented
in
this
appendix
can
be
programmed
into
a
conventional
spreadsheet
or
a
plant
computer
control
system.

The
following
tables
define
the
types
of
chambers
potentially
present
in
an
ozone
contactor
and
show
the
recommended
methods
for
calculating
the
inactivation
credit
achieved.
Only
the
T
10
or
CSTR
methods
can
be
applied
to
dissolution
chambers.
However,
they
can
be
applied
to
the
reactive
chambers
as
well.
In
general
the
T
10
method
should
be
used
unless
significant
back
mixing
occurs
in
the
chamber.
If
no
tracer
test
data
are
available,
it
is
recommended
that
the
CSTR
method
be
used.
The
Extended­
CSTR
method
is
applied
over
a
minimum
of
three
consecutive
reactive
chambers.
Table
B.
1
shows
the
recommended
methods.

Table
B.
1
Applicable
Methods
and
Terminology
for
Calculating
the
Log
Inactivation
Credit
Section
Description
Terminolog
y
Method
for
Calculating
Log
Inactivation
Restrictions
N
o
T
r
a
c
Chambers
where
ozone
is
added
e
r
D
a
t
a
First
chamber
First
Dissolution
Chamber
No
log
inactivation
credit
is
recommended
None
O
th
e
r
c
h
a
m
b
e
rs
Co­
Current
or
Counter­
Current
Dissolution
Chambers
CSTR
Method
in
each
chamber
with
a
measured
effluent
ozone
residual
concentratio
n
No
credit
is
given
to
a
dissolution
chamber
unless
a
detectable
ozone
residual
has
been
measured
upstream
of
this
chamber
Reactive
Chambers
>
3
consecutive
reactive
chambers
Extended­

CSTR
Zone
Extended­
CSTR
Method
in
each
chamber
Detectable
ozone
residual
should
be
present
in
at
least
3
chambers
in
this
zone,
measured
via
in­
situ
sample
ports.
Otherwise,
the
CSTR
method
should
be
applied
individually
to
each
chamber
having
a
measured
ozone
residual
<

3
c
o
n
s
e
CSTR
Reactive
Chamber(
s)
CSTR
Method
in
each
chamber
with
a
measured
effluent
None
c
ut
iv
e
r
e
a
ct
iv
e
c
h
a
m
b
e
rs
ozone
residual
concentratio
n
W
it
h
T
r
a
c
e
r
D
a
t
a
Chambers
where
ozone
is
added
First
chamber
First
Dissolution
Chamber
No
log
inactivation
is
credited
to
this
section
Not
applicable
O
th
e
Co­
Current
or
Counter­
Current
Dissolution
T
10
or
CSTR
Method
in
each
No
credit
will
be
given
to
a
dissolution
chamber
r
c
h
a
m
b
e
rs
Chambers
chamber
unless
a
detectable
ozone
residual
has
been
measured
upstream
of
this
chamber
Reactive
Chambers
>
3
consecutive
chambers
with
in­
situ
sample
ports
Extended­

CSTR
Zone
Extended­
CSTR
Method
in
each
chamber
Detectable
ozone
residual
should
be
present
in
at
least
3
chambers
in
this
zone,
measured
via
in­
situ
sample
ports.
Otherwise,
the
T
10
or
CSTR
method
should
be
applied
to
each
chamber
having
a
measured
ozone
residual
<

3
c
o
n
s
e
c
ut
iv
e
c
h
a
m
b
e
rs
T
10
or
CSTR
Reactive
Chamber(
s)
T
10
or
CSTR
Method
in
each
chamber
None
B.
3
Ozone
Contactor
Configurations
Ozone
contactors
are
designed
in
a
wide
variety
of
configurations.
Different
configurations
are
adaptable
to
the
Extended­
CSTR
approach,
but
implementation
details
vary
with
contactor
configuration.
It
is
important
for
a
water
system
to
identify
the
type
of
configuration
and
become
familiar
with
the
terminology
used
in
this
guidance
manual.

Figure
B.
1
shows
configurations
with
multiple,
consecutive
well­
defined
reactive
chambers.
The
water
flow
pattern
in
such
contactors
can
be
an
"
over­
under"
pattern,
a
"
serpentine"
pattern,
or
a
combination
of
both.
Gaseous
ozone
is
added
to
the
water
by
one
of
two
procedures.
Gaseous
ozone
can
be
injected
into
the
influent
water
before
the
water
enters
the
contactor,
a
process
often
called
"
in­
line"
ozone
addition
(
see
schematic
B
&
D
in
Figure
B.
1).
Alternatively
ozone
enriched
gas
can
be
bubbled
into
one
or
more
chambers,
a
process
called
"
in­
chamber"
ozone
addition
(
see
schematic
A
&
C
in
Figure
B.
1).
In­
chamber
ozone
addition
takes
place
in
chambers
that
have
an
over­
under
flow
pattern
and
not
in
chambers
that
have
a
serpentine
flow
pattern
(
Figure
B.
1­
C)
in
order
to
ensure
full
and
complete
ozone
dissolution
into
all
the
water
flow.
These
so­
called
bubble
columns
can
be
counter­
current
or
cocurrent
describing
the
directional
flow
of
the
water
with
respect
to
the
upward
flowing
bubbles.
Note,
Figure
B.
1
only
shows
example
configurations;
size
and
geometry
of
the
chambers
will
vary.
Figure
B.
1
Schematics
of
Typical
Configurations
of
Ozone
Contactors
with
Multiple
Chambers
In
contrast
to
the
multi­
chamber
configuration,
ozone
contactors
may
also
be
comprised
on
only
one
or
two
reactive
chambers.
Examples
of
such
contactors
are
shown
in
Figure
B.
2,
which
include
a
closed­
pipe
contactor
(
see
schematic
A)
and
two
open­
channel
contactors
(
see
schematics
B
&
C).
All
three
contactors
include
a
long
and
narrow
water
flow
path
that
promotes
plug­
flow
hydraulic
characteristics.
As
with
multi­
chamber
contactors,
ozone
can
be
added
inline
or
in­
chamber.
Contactors
A
and
B
illustrate
in­
line
ozone
addition.
Contactor
C
illustrates
in­
chamber
ozone
addition.
Figure
B.
2
­
Schematics
of
Example
Single­
or
Dual­
Chamber
Ozone
Contactors
B.
4
Extended­
CSTR
Approach
for
Ozone
Contactors
B.
4.1
Introduction
The
method
described
in
this
chapter
represent
a
more
sophisticated
approach
to
calculating
inactivation
credit
in
an
ozone
contactor
as
compared
to
the
T
10
and
CSTR
approaches.
This
approach
could
potentially
provide
a
higher
and
more
accurate
estimate
of
the
level
of
Cryptosporidium
inactivation
than
that
obtained
using
the
T
10
approach.
The
potential
benefits
of
using
these
more
sophisticated
measures
are
lower
ozone
doses
and
lower
ozonation
disinfection
byproducts,
(
e.
g.
bromate).
However,
as
a
consequence
of
this
added
sophistication,
a
higher
degree
of
system
evaluation
and
monitoring
is
needed
for
a
given
inactivation
credit.
Whether
use
of
these
more
sophisticated
approaches
actually
benefit
the
utility
depends
on
many
factors
including
the
sought­
after
level
of
inactivation,
the
reactor
configuration,
and
the
water
quality.

The
approach
described
in
this
chapter
is
called
the
Extended­
CSTR
Approach.
Certain
aspects
of
this
methodology
was
introduced
in
Appendix
O
of
the
SWTR
Guidance
Manual.
However,
the
material
presented
here
greatly
expands
upon
the
SWTR
Guidance
Manual,
and
may
provide
beneficial
new
tools
for
the
utility.

B.
4.2
Overview
of
System
Evaluation
and
Monitoring
The
Extended­
CSTR
approach
relies
on
modeling
ozone
decay
reactions
through
ozone
contactors.
In
principal,
the
kinetics
of
ozone
decay
in
the
contactor
is
modeled
in
concert
with
the
hydrodynamics
of
the
ozone
contactor,
which
is
assumed
to
be
that
of
an
ideal
CSTR.
This
approach
is
applied
only
to
"
reactive
chambers"
within
a
contactor.

B.
4.3
Extended­
CSTR
Approach
­
Ozone
Contactors
without
a
Tracer
Test
In
the
event
that
an
approved
set
of
tracer
test
results
is
unavailable
for
an
ozone
contactor,
the
utility
may
choose
one
of
the
following
two
options:

1.
Use
the
CSTR
method
to
calculate
the
log
inactivation
across
each
individual
chamber.

2.
Use
the
Extended­
CSTR
approach
to
calculate
the
log
inactivation
across
each
individual
chamber.

The
choice
of
using
the
CSTR
approach,
the
Extended­
CSTR
approach,
or
a
combination
of
the
two
greatly
depends
on
the
reactor
configuration
and
the
manner
in
which
the
measurement
of
ozone
residuals
is
attained.
Briefly,
for
CSTR
approach,
concentrations
are
measured
for
each
chamber
where
log
inactivation
is
calculated.
In
contrast,
for
the
chambers
in
the
Extended­
CSTR
approach,
ozone
concentrations
of
each
chamber
are
calculated
through
modeling
of
the
ozone
decay.
This
section
describes
the
appropriate
application
of
the
CSTR
approach
and
Extended­
CSTR
approach
to
calculate
the
log
inactivation
credit
across
the
contactor.

B.
4.3.1
Classification
of
the
Chambers
and
Contactor
Zones
The
contactor
should
be
divided
into
specific
sections,
or
zones,
to
properly
calculate
the
inactivation
credit
across
a
conventional
contactor.
To
ensure
clarity,
certain
terminology
is
adopted
for
unique
sections
of
an
ozone
contactor,
as
presented
in
Table
B.
1.

Figure
B.
3
shows
an
example
schematic
of
a
10­
chamber
over­
under
baffled,
multichamber
ozone
contactor
with
in­
chamber
ozone
addition.
Ozone
is
being
added
in
Chambers
1
and
4
only
in
this
example.

Chamber
1
is
classified
as
a
"
First
Dissolution
Chamber"
and
it
is
recommended
that
no
disinfection
credit
be
granted
for
this
chamber.
Rapid,
initial
ozone
reactions
and
the
transitional
development
of
the
ozone
residual
occur
in
the
first
dissolution
chamber.
As
such,
a
representative
dissolved
ozone
profile
is
difficult
to
estimate
without
multiple
sample
ports
along
the
bubble
column.
The
second
and
third
chambers
in
the
contactor
shown
in
Figure
B.
3
are
reactive
chambers
through
which
ozone
is
decaying.
These
chambers
are
called
"
CSTR
Reactive
Chambers".
The
CSTR
method
is
used
to
calculate
log
inactivation
across
CSTR
Reactive
Chambers
when
ozone
residual
values
are
available
from
the
effluent
of
the
chamber.
The
CSTR
method
is
described
in
Chapter
11.

Figure
B.
3
­
Names
of
the
Various
Sections
of
a
Multi­
Chamber
Over­
Under
Ozone
Contactor
The
fourth
chamber
in
the
contactor
shown
in
Figure
B.
3
includes
ozone
addition.
This
chamber
is
called
a
"
Co­
Current
Dissolution
Chamber".
It
should
be
emphasized
that
a
chamber
is
given
the
"
Dissolution
Chamber"
notation
only
when
ozone
residual
has
been
detected
at
any
point
upstream
of
the
influent
to
that
chamber.
In
other
words,
chamber
4
in
Figure
B.
3
can
be
classified
as
a
Dissolution
Chamber
only
if
ozone
residual
has
been
detected
at
the
effluent
of
either
chamber
1,
2,
or
3.
The
CSTR
method
is
used
to
calculate
the
log
inactivation
credit
across
a
Dissolution
Chamber.
If
no
ozone
residual
was
detected
upstream
of
this
chamber
location,
then
chamber
4
takes
on
the
classification
of
a
"
First
Dissolution
Chamber"
and
as
with
chamber
1,
no
log
inactivation
credit
is
granted.

Chambers
5
through
10
in
the
contactor
pictured
in
Figure
B.
3
represent
the
"
Extended­
CSTR
zone"
since
they
meet
the
criterion
of
containing
a
minimum
of
three
consecutive
reactive
chambers.
Since
tracer
data
are
unavailable,
the
Extended­
CSTR
approach
is
used
to
calculate
the
log
inactivation
across
each
chamber
in
this
zone.
Modeling
is
used
to
calculate
the
ozone
residual
concentration
at
the
effluent
of
each
chamber
within
the
Extended­
CSTR
zone.
This
modeling
requires
an
accurate
estimation
of
the
ozone
decay
coefficient,
k*,
and
the
initial
ozone
residual
at
the
entrance
to
the
zone,
C
in.
Estimation
of
these
two
parameters,
which
is
discussed
in
sections
B.
4.3.2.1
and
B.
4.3.2.2,
requires
the
measurement
of
three
ozone
residual
values
across
the
minimum
span
of
three
chambers.

In
the
case
of
a
contactor
with
in­
line
ozone
addition,
the
entire
contactor
potentially
becomes
an
Extended­
CSTR
zone.
If
the
contactor
has
at
least
three
chambers
equipped
with
insitu
sample
ports
and
a
measurable
ozone
residual
then
the
requirements
for
calculating
k*
and
C
in
have
been
met
and
the
entire
contactor
can
be
treated
as
an
Extended­
CSTR
zone.
Care
should
be
taken
in
locating
the
first
ozone
sample
port
such
that
enough
reaction
time
is
allowed
for
the
immediate
ozone
demand
to
be
fully
met
before
the
sample
port.

B.
4.3.2
Calculating
Log
Inactivation
across
an
Extended­
CSTR
Zone
Calculation
of
log
inactivation
across
an
Extended­
CSTR
zone
is
handled
in
much
the
same
manner
as
it
is
for
a
CSTR
Reactive
Chamber
as
discussed
in
Chapter
11.
The
Extended­
CSTR
zone
comprises
three
or
more
individual
chambers.
Inactivation
within
each
chamber
is
calculated
according
to
Equation
11­
1,
exactly
as
it
is
for
the
CSTR
chamber
above,
and
the
sum
of
the
log
inactivation
values
for
individual
chambers
gives
the
inactivation
across
the
whole
zone.
The
distinction
between
a
CSTR
Reactive
Chamber
and
a
chamber
that
is
a
component
of
an
Extended­
CSTR
zone
is
the
manner
in
which
the
value
for
C
is
obtained.
In
the
case
of
the
CSTR
Reactive
Chamber,
C
is
obtained
from
an
actual
measurement
of
the
dissolved
ozone
residual
at
the
exit
of
the
chamber
(
i.
e.,
C
out).
In
contrast,
C
for
a
chamber
in
an
Extended­
CSTR
zone
is
a
calculated
value.
The
procedure
for
calculating
C
for
an
Extended­
CSTR
zone
is
described
in
this
section.

The
value
of
C
for
an
Extended­
CSTR
is
calculated
using
the
first­
order
ozone
decay
coefficient,
k*,
and
the
ozone
residual
concentration
at
the
entrance
to
the
zone,
C
in.
Equation
B­
1
shows
how
to
calculate
the
ozone
residual
at
the
effluent
of
chamber
"
X"
in
an
Extended­
CSTR
zone:

(
B­
1)

where:
k*=
s
described
in
section
B.
4.3.2.1
C
in
=
STR
zone,
mg/
L,
calculated
as
described
in
section
B.
4.3.2.2
[
Volume]
0­
X=
to
the
effluent
of
chamber
"
X"
N
0­
X=
to
the
effluent
of
chamber
"
X"
Q
=
Water
flow
through
the
contactor,
gpm
Equation
B­
1
describes
the
Extended­
CSTR
zone
between
the
first
chamber
(
subscript
0)
and
chamber
X
as
a
series
of
equal­
volume
CSTR
reactors.
This
is
a
simplifying
assumption
that
is
based
on
a
balance
between
ease
of
implementation
and
consistency
with
other
provisions
within
this
guidance
manual.

Once
the
values
of
the
ozone
residual
concentrations
at
the
effluent
of
each
chamber
in
the
Extended­
CSTR
zone
are
calculated,
Equation
11­
1
can
then
be
used
to
calculate
the
log
inactivation
achieved
across
that
chamber.
The
total
log
inactivation
achieved
across
the
entire
contactor
is
equal
to
the
sum
of
the
log
inactivation
values
calculated
for
each
chamber.

­
Log
(
I/
I
0)
=
Log
(
1
+
2.303k
10
x
C
x
HDT)
Equation
11­
1
where:
­
Log
(
I/
I
0)
=
the
log
inactivation
k
10
=
log
base
ten
inactivation
coefficient
(
L/
mg­
min)
C
=
Concentration
from
Table
11­
2
(
mg/
L)
HDT
=
Hydraulic
detention
time
(
minutes)

Because
the
ozone
demand
in
the
water
is
constantly
changing,
the
values
of
k*
and
C
in
should
be
determined
every
time
log
inactivation
credit
is
calculated
(
i.
e.
at
least
daily).
These
parameters
are
calculated
using
three
measured
ozone
residuals
from
three
locations
within
the
Extended­
CSTR
zone.

B.
4.3.2.1
Determining
the
Value
of
k*

The
ozone
decay
coefficient,
k*
is
calculated
using
ozone
sample
measurements,
taken
from
in­
situ
sample
ports,
and
a
model
of
the
chamber's
hydrodynamics.
The
following
approach
assumes
that
the
individual
chambers
can
be
modeled
as
a
CSTR
(
or
equal­
volume
CSTR­
inseries
if
there
are
more
than
one
chamber
between
sample
ports).

Calculating
k*

The
steps
outlined
below
pertain
to
a
contactor
with
a
minimum
of
three
consecutive
chambers
with
measurable
ozone
residuals.
That
is,
there
should
be
at
least
three
in­
situ
sample
ports
from
the
Extended­
CSTR
zone
with
measurable
ozone
residual.
The
three
ozone
residual
measurements,
C
1,
C
2,
and
C
3,
are
needed
to
estimate
the
value
of
the
ozone
decay
coefficient,
k*.
For
example,
the
Extended­
CSTR
zone
in
the
contactor
shown
in
Figure
B.
3
includes
chambers
5
through
10.
The
ozone
residual
values
at
any
three
chambers
in
that
span
can
be
used
to
represent
C
1,
C
2,
and
C
3
in
this
analysis.
The
following
steps
should
be
followed
to
calculate
the
k*
value:

Step
1
­
Use
Equation
B­
2
and
residual
measurements
C
1
and
C
2
to
calculate
the
k*
value
representing
the
ozone
decay
between
locations
1
and
2,
.
(
A
derivation
and
explanation
of
*
2
1
 
k
Equation
B­
2
is
presented
in
Appendix
E):
(
B­
2)

where:
=
sampling
locations
1
&
2,
min­
1
*
2
1
 
k
C
1
=
one
residual
at
location
1,
mg/
L
C
2
=
one
residual
at
location
2,
mg/
L
[
Volume]
1­
2=
pling
locations
1
and
2,
gallons
N
1­
2=
een
sampling
locations
1
and
2
Q=
ow
through
the
contactor,
gpm
Step
2
­
Use
residual
measurements
C
1
and
C
3
along
with
Equation
B­
3
to
calculate
the
k*

value
representing
ozone
decay
between
sampling
locations
1
and
3,
:
*
3
1
 
k
(
B­
3)

where:
=
sampling
locations
1
&
3,
min­
1
*
3
1
 
k
C
1
=
one
residual
at
location
1,
mg/
L
C
3
=
one
residual
at
location
3,
mg/
L
[
Volume]
1­
3=
pling
locations
1
and
3,
gallons
N
1­
3=
een
sampling
locations
1
and
3
Q=
ow
through
the
contactor,
gpm
It
should
be
emphasized
that
sampling
location
1
should
not
be
at
the
entrance
to
the
Extended­
CSTR
zone,
but
should
be
at
least
one
chamber
into
the
zone.
For
example,
in
Figure
B.
3,
C
1
should
not
be
measured
at
the
entrance
to
chamber
5,
since
that
is
the
entrance
to
the
Extended­
CSTR
zone.
Instead,
the
first
Extended­
CSTR
zone
sampling
location
should
be
located
at
the
effluent
of
chamber
5,
or
downstream
of
that
location.
Section
O.
3.2
of
Appendix
O
of
the
SWTR
Guidance
Manual
provides
guidance
on
the
use
of
in­
situ
sample
ports
for
direct
ozone
measurements.

Step
3
­
The
value
of
k*
that
is
to
be
used
in
Equation
B­
1
will
be
calculated
as
the
average
of
and
as
shown
in
Equation
B­
4.
*
2
1
 
k
*
3
1
 
k
(
B­
4)








+
=
 
 
2
*
3
1
*
2
1
*
k
k
k
It
is
normal
for
the
individual
values
of
and
to
be
somewhat
different.
However,
*
2
1
 
k
*
3
1
 
k
it
is
recommended
that
they
be
within
the
range
of
80%
to
120%
of
the
average
k*
value
calculated
in
Step
3.
That
is,

If
they
are
outside
this
range,
the
measured
residual
values
should
be
rejected
and
new
samples
should
be
collected
until
this
quality
assurance
(
QA)
criterion
is
met.

Ozone
residual
measurement
at
the
three
locations
might
be
conducted
manually
using
the
Indigo
Trisulfonate
method,
or
continuously
using
on­
line
ozone
analyzers.
The
Quality
Assurance
protocols
discussed
in
Appendix
D
should
be
implemented
to
ensure
that
the
ozone
residual
measurements
are
accurate.

B.
4.3.2.2
Determining
the
Value
of
C
in
While
it
is
possible
to
measure
the
ozone
residual
at
the
entrance
to
the
Extended­
CSTR
zone
(
e.
g.,
an
in­
situ
sample
port),
it
is
not
recommended
that
the
measured
value
be
used
because
it
is
usually
higher
than
the
residual
predicted
by
the
first­
order
decay
profile
(
Amy
et
al.,
1997;
Carlson
et
al.,
1997;
Hoigné
and
Bader,
1994;
Rakness
and
Hunter,
2000;
Rouston
et
al.,
1998).
This
phenomenon
is
commonly
attributed
to
the
more
rapid
initial
ozone
decay,
which
is
followed
by
a
somewhat
slower
first­
order
decay
profile.
For
this
reason,
the
C
in
representing
the
ozone
decay
in
the
Extended­
CSTR
Zone
should
be
extrapolated
using
the
downstream
measured
ozone
residual
values.

The
value
of
C
in
can
be
calculated
once
the
value
of
k*
is
estimated
from
the
three
residual
ozone
measurements.
Maintaining
the
assumption
of
a
first­
order
decay
rate,
and
again
using
the
CSTR
(
or
equal­
volume
CSTR­
in­
series
if
there
are
more
than
one
chamber
between
sample
ports)
assumption,
Equations
B­
5
through
B­
7
can
be
used
to
estimate
the
value
of
C
in
from
the
three
measured
ozone
residual
concentrations:

(
B­
5)

(
B­
6)
(
B­
7)

where:
k*=
st­
order
decay
coefficient,
min­
1
C
1
=
one
residual
at
location
1,
mg/
L
C
2
=
one
residual
at
location
2,
mg/
L
C
3
=
one
residual
at
location
3,
mg/
L
N
0­
1=
R
Zone
and
sampling
location
1
N
0­
2=
R
Zone
and
sampling
location
2
N
0­
3=
R
Zone
and
sampling
location
3
[
Volume]
0­
1=
R
Zone
and
sampling
location
1
[
Volume]
0­
2=
R
Zone
and
sampling
location
2
[
Volume]
0­
3=
R
Zone
and
sampling
location
3
Q=
ow
through
the
contactor,
gpm
The
C
in
value
is
then
calculated
as
the
average
of
the
three
values
determined
by
Equations
B­
5
through
B­
7:

(
B­
8)

These
calculations
outline
the
methodology
of
the
Extended­
CSTR
approach.
A
systematic
example
of
the
Extended­
CSTR
approach
is
presented
in
section
B.
4.5
B.
4.3.2.3
Quality
Assurance
for
Extended­
CSTR
Calculations
The
Extended­
CSTR
method
depends
on
ozone
residual
measurements
and
an
assumption
that
the
contactor
hydrodynamics
can
be
modeled
as
a
CSTR
in
order
to
predict
ozone
concentrations
through
the
contactor.
To
ensure
that
the
predicted
concentrations
are
accurate,
both
the
measurements
and
assumptions
should
be
verified.
Therefore,
QA
controls
are
recommended
as
described
below.

The
predicted
ozone
residual
concentration,
the
parameter
C
in
Equation
11­
1,
encompasses
both
the
CSTR
assumption
and
ozone
measurements.
The
principal
QA
issues
focus
on
the
prediction
of
the
value
of
C.
As
seen
in
equation
B­
1,
C
depends
on
the
parameters
k*
and
C
in.
In
section
B.
4.3.2.1,
as
part
of
the
discussion
on
the
calculation
of
k*,
it
is
stipulated
that
the
individual
k*
values
(
i.
e.,
k*
1­
2
and
k*
1­
3)
should
be
within
20%
of
the
average
value.
This
QA
control
is
meant
to
ensure
that
ozone
residual
measurements
used
to
calculate
the
ozone
decay
profile
are
consistent
with
the
calculated
profile.
Since
the
calculation
of
C
in
(
Equations
B­
5
through
B­
8)
depends
on
k*,
as
well
as
the
measured
ozone
concentrations,
the
QA
criteria
for
k*
is
sufficient
for
C
in.
Therefore,
no
additional
QA
criteria
are
necessary
for
it.

The
accuracy
of
the
CSTR
assumption
cannot
be
completely
verified
without
conducting
a
tracer
study
through
the
contactor.
However,
it
is
recommended
that
ozone
residual
measurements
be
taken
at
different
flows
and
ozone
doses,
and
k*
and
C
in
be
calculated
at
the
different
conditions,
in
order
to
determine
the
impact
of
changing
conditions
on
the
predicted
ozone
decay
rate.

Finally,
one
of
the
most
important
aspects
of
any
application
of
a
model
towards
predicting
reactor
performance
is
the
confirmation
of
the
model's
prediction.
This
is
in
essence
"
model
validation."
Appendix
O
of
the
SWTR
Guidance
Manual
makes
several
points
to
this
effect.
Ideally,
model
validation
would
take
the
form
of
measuring
the
actual
disinfection
of
the
Cryptosporidium.
A
more
practical
alternative
is
to
compare
the
predicted
ozone
concentrations
to
measured
values.
The
general
recommendation
is
that
the
predicted
ozone
residual
should
not
be
greater
than
20%
of
a
measured
value.
Note
that
this
is
a
one­
sided
QA
control.

The
ozone
concentration
measurements
used
to
calculate
k*
and
C
in
cannot
be
compared
to
the
predicted
ozone
residuals,
since
they
are
interdependent.
It
is
recommended
that
ozone
samples
be
taken
from
other
sampling
locations
in
the
contactor,
and
those
values
compared
to
the
calculated
C.

B.
4.4
Example
of
Extended­
CSTR
Application
This
section
provides
an
example
calculating
the
log
inactivation
credits
using
the
Extended­
CSTR
approach.

Example
­
Extended­
CSTR
Approach
for
a
Multi­
chamber
Contactor
With
In­
situ
Sample
Ports
and
One
Dissolution
Chamber
Figure
B.
6
shows
a
schematic
of
a
12­
chamber
ozone
contactor.
The
contactor
is
treating
50
MGD
of
water
at
a
temperature
of
20
°
C.
The
volumes
of
the
individual
chambers
are
noted
on
the
schematic.
Ozone
is
added
to
the
first
chamber
only.
The
bottom
graph
in
Figure
B.
6
shows
the
values
of
the
ozone
residual
measured
at
the
effluents
of
chambers
2,
5,
and
8.

Figure
B.
6
­
Schematic
of
the
Ozone
Contactor
and
the
Measured
Ozone
Residual
Values
in
Example
1
The
Cryptosporidium
inactivation
credit
across
the
contactor
is
calculated
as
follows:

Chamber
1(
First
Dissolution
Chamber)
­
No
inactivation
credit
is
given
to
the
first
dissolution
chamber.

Chambers
2
through
12
(
Extended­
CSTR
zone)
­
This
zone
is
classified
as
an
Extended­
CSTR
zone.
The
Extended­
CSTR
calculations
(
Section
4.3)
are
applied
to
determine
the
log
inactivation
across
each
chamber.
The
following
steps
are
implemented
Step
1:
Calculate
k*
value
­
The
k*
value
is
calculated
as
described
in
section
B.
4.3.3.1
using
the
three
ozone­
residual
measurements,
C
1,
C
2,
and
C
3
that
are
shown
in
Figure
B.
6.
The
values
of
and
can
be
calculated
using
Equations
B­
2
and
B­
3
as
follows:
*
2
1
 
k
*
3
1
 
k
=
0.0670
min­
1
=
0.0785
min­
1
The
k*
value
is
then
calculated
as
the
average
of
and
as
follows:
*
2
1
 
k
*
3
1
 
k
=
0.0728
min­
1
A
QA
check
shows
that
the
values
of
and
are
within
8%
of
the
average
k*
value
of
*
2
1
 
k
*
3
1
 
k
0.0728
min­
1.
This
value
of
k*
is
within
the
recommended
maximum
variability
of
20%.

Step
2:
Calculate
C
in
value
­
The
value
of
C
in
is
calculated
using
the
approach
described
in
Section
4.3.3.2.
With
the
value
of
k*
calculated
at
0.0728
min­
1,
Equations
B­
5
to
B­
7
can
be
used
to
calculate
the
C
in
value
as
follows:

=
0.865
mg/
L
=
0.902
mg/
L
=
0.796
mg/
L
Therefore,

=
0.854
mg/
L
Step
3:
Calculate
the
value
of
k
10
­
The
value
of
k
10
for
the
inactivation
of
Cryptosporidium
with
ozone
at
the
measured
temperature
of
20
°
C
can
be
obtained
from
Table
11­
3
directly
and
equals
0.2537
L/
mg­
min.
Otherwise
the
value
for
k
10
could
be
determined
using
equation
11­
2.

Step
4:
Calculate
the
Ozone
Residual
at
the
Effluent
of
Each
Chamber
­
Knowing
the
values
of
C
in
and
k*,
the
ozone
concentration
at
the
effluent
of
each
chamber
within
the
Extended­
CSTR
zone
can
be
calculated.
These
values
are
calculated
using
Equation
B­
1:
where
C
X
is
the
calculated
concentration
at
the
effluent
of
chamber
"
X".
For
example,
the
residual
concentration
at
the
effluent
of
chamber
4
is
calculated
as:

=
0.473
mg/
L
Note
that
the
Extended­
CSTR
zone
begins
at
the
effluent
of
Chamber
1,
which
makes
the
subscript
to
[
Volume]
in
the
equation
above
depicted
as
"
1­
4".
Table
B.
10
lists
the
calculated
residual
values
for
each
chamber
using
the
same
approach,
beginning
with
chamber
2.

Table
B.
10
­
Application
of
the
Extended­
CSTR
Method
to
the
Example
Step
4:
Calculate
Log
Inactivation
­
Knowing
the
values
of
C,
k
10,
and
k*,
Equation
11­
1
is
used
to
calculate
the
log
inactivation
achieved
in
each
chamber
in
the
Extended­
CSTR
Zone:
where
C
X
is
the
effluent
residual
concentration
at
Chamber
X
and
[
Volume]
X
is
the
volume
of
that
chamber.
For
example,
the
log
inactivation
achieved
in
chamber
4
is
calculated
as:

=
0.26
logs
Column
(
4)
in
Table
B.
10
lists
the
log
inactivation
values
calculated
for
chambers
2
through
12.
The
sum
of
the
log
inactivation
achieved
(
total
of
Column
4
in
Table
B.
9)
is
1.9
logs.
References
Amy,
G.
L.,
P.
Westerhoff,
R.
A.
Minear,
and
R.
Song.
1997.
Formation
and
Control
of
Brominated
Ozone
By­
Products.
AWWA
Research
Foundation,
Denver,
CO.

 
Carlson,
K.,
K.
Rakness,
and
S.
MacMillan.
1997.
Batch
Testing
Protocol
for
Optimizing
Ozone
System
Design.
Presented
at
AWWA
Annual
Conference
in
Atlanta,
GA
­
June
15­
19,
1997.

 
Froment,
G.
F.
and
K.
B.
Bischoff.
2nd
ed.
1990,
Chemical
Reactor
Analysis
and
Design.,
New
York:
John
Wiley
&
Sons.

 
Gordon,
G.,
R.
D.
Gauw,
Y.
Miyahara,
B.
Walters,
and
B.
Bubnis.
2000A.
"
Using
Indigo
Absorbance
to
Calculate
the
Indigo
Sensitivity
Coefficient,"
J.
AWWA,
92(
12):
96­
100.

Gordon,
G.,
B.
Walters,
and
B.
Bubnis.
2000B.
"
The
Effect
of
Indigo
Purity
on
Measuring
the
Concentration
of
Aqueous
Ozone,"
Conference
Proceedings:
Advances
in
Ozone
Technology,
Orlando,
FL.
International
Ozone
Association,
Pan
American
Group.

Guidance
Manual
for
Compliance
With
the
Filtration
and
Disinfection
Requirements
for
Public
Water
Systems
Using
Surface
Water
Sources.
March
1991
Edition.
USEPA
Office
of
Drinking
Water,
Cincinnati,
OH.

 
Hoigné,
J.
and
H.
Bader.
1994.
Characterization
of
Water
Quality
Criteria
for
Ozonation
Processes.
Part
II:
Lifetime
of
Added
Ozone.
Ozone:
Science
&
Engineering.
Vol.
16,
No.
2:
pp.
121­
134.

 
Levenspiel,
O.,
3rd
ed.
1999.
Chemical
Reaction
Engineering.
New
York:
John
Wiley
&
Sons.

Rakness,
K.
L.
G.
Gordon,
B.
Bubnis,
D.
J.
Rexing,
E.
C.
Wert,
and
M.
Tremel.
2001.
"
Underestimating
Dissolved
Ozone
Residual
Using
Outdated
or
Impure
Indigo,"
Conference
Proceedings:
International
Ozone
Association
15th
World
Congress;
London,
England;
International
Ozone
Association
­
September
2001).

Rakness,
K.
L.
and
G.
F.
Hunter.
2000.
"
Advancing
Ozone
Optimization
During
Pre­
Design,
Design
and
Operation."
AWWA
Research
Foundation,
Denver,
CO,
and
Electric
Power
Research
Institute­
Community
Environmental
Center,
St.
Louis,
MO.

 
Rakness,
K.
L.,
and
G.
F.
Hunter.
2001.
"
Monitoring
and
Control
of
Ozone
Disinfection
for
Crypto,
Giardia,
and
Virus
Inactivation."
Conference
Proceedings
of
International
Ozone
Association
World
Congress;
London,
England
­
September
2001.

Rakness,
K.
L.,
G.
Gordon,
D.
J.
Rexing,
and
E.
C.
Wert.
2002.
"
Reported
Ozone
Residual
Data
Might
Be
Undervalued."
Conference
Proceedings:
American
Water
Works
Association
Annual
Conference;
New
Orleans,
LA
­
June
2002).

Roustan,
M.,
H.
Debellfontaine,
Z.
Do­
Quang,
and
J.
Duguet.
1998.
Development
of
a
Method
for
the
Determination
of
Ozone
Demand
of
Water.
Ozone:
Science
&
Engineering.
Vol.
20,
No.
6:
pp.
513­
520.

 
Standard
Methods
for
the
Examination
of
Water
and
Wastewater,
20th
Edition.
1998.
(
American
Public
Health
Association,
American
Water
Works
Association,
and
Water
Environment
Federation),
pp.
4­
137
and
4­
138.

Teefy,
S.
and
P.
Singer.
1990.
Performance
and
Analysis
of
Tracer
Tests
to
Determine
Compliance
of
a
Disinfection
Scheme
with
the
SWTR.
Journal
AWWA,
82(
12):
88­
89.

Teefy,
S.
et
al.,
1996.
Tracer
Studies
in
Water
Treatment
Facilities:
A
Protocol
and
Case
Studies.
Final
Report.
American
Water
Works
Association
Research
Foundation.
American
Water
Works
Association,
Denver,
CO.
