Note: Descriptions are shown in the official language in which they were submitted.
104ZS75
BACKGROUND 0- ~HE I~VENII~N
The present invention relates to a proce~s or regen-
erating an ion e~change bed used to remove ammonium ions from
wastewater.
Ammonium ions interfere with many of the important
uses of water. They are toxic to fish, corrosive to metals
and concrete, and a matter of concern when consumed by man.
These ions enter water supply systems from a variety of
sources, one major input being the discharge of municipal
and industrial wastewaters with high ammonium ion concentra-
tions. Most sanitary discharges contain 15 to 30 parts per
million (ppm) ammonia as nitrogen (typically 20 ppm). A
generally accepted goal of the industry is to reduce this
ion level to the lowest value commensurate with economy and
good engineering practice (0-3 ppm). This level can be
achieved by the regeneration process embodied in the pre~sent
invention.
Traditionally, em~hasis in wastewater treatment has
been placed on the removal of biologically degradable organic
material, suspended solids, and floating substances. The
objectives of such treatment are to produce a clear effluent,
which, when mixed with the receiving water, will produce
minimal oxygen depletion and no gross signs of pollution or
objectionable odors. The physical and biological treatment
processes developed to achieve these objectives do not reduce
ammonium ion concentrations to desirable levels. Therefore,
some form of ammonia re val is necessary prior to the dis-
charge of the wast~water. Considerable attention has been
directed to the effective and economic removal of ammonia
nitrogen from wastewater streams.
.- --1-- .
1042575
Nitrogen can be removed through microbial action in
conventional biological waste treatment plants. Removal
using the standard activated sludge process requires a suffi-
cient mean cell residence time to allow nitrification bacteria
to become established in the system. The required aeration
period length negates the economic advantages of high rate
biological systems. Even in a compartmentalized system,
treatment periods are long and problems develop in maintaining
systems with different biological functions. In addition,
algal harvesting or stripping requires re land than other
plant processes. Biological nitrification-denitrification
has proven erratic and inadequate to meet water quality criteria.
The uncertainties and costs of biological removal have
stimulated the investigation of physical/chemical removal of -
ammonia nitrogen by an ion exchange process. Ion exchange
ammonia removal is more amenable to control than biological
processes ~nd more adaptable to the 1uctuating flows and
- concentrations of municipal wastewater systems. As a unit
process, ion exchange is easily controlled to achieve almost
20 - any desired product quality. In the ion exchange process,
the ion exchange material exchanges the ammonium ion (NH4+)
in the waste stream for ions originally in the ion exchange bed.
The ion exchange process to which the present invention
is directed, uses a zeolite bed, such as clinoptilolite (a
natural zeolite), through which a clarified sewage effluent
flows. As the influent passes through the bed, NH4+ ions
lodge on active sites on the zeolite. The exchange medium
preferentially absorbs ammonium ions in the presence of
sodium, calcium and magnesium ions.
-- -- .
1042S75
A complete discussion of the use oE a clinoptilolite
ion exchange bed for the removal of ammonia may be Eound in a
report approved for publication by the Environmental Protection
Agency entitled: Optimization of Ammonia Re val by Ion Exchange
Usin~ Clinoptilolite by the Sanitary Engineering Research
Laboratory, College of Engineering and School of Public Health,
University of California, Berkeley (September 1971). In tests
discussed in this report, the clinoptilolite columns achieved
an ammonia removal of 95.7%. The effluent ammonia concentra-
tion average was 0.75 ppm for runs which ranged from 120 to
180 in bed volumes in length. The regeneration process of
the present invention permits the achievement of equivalent
or superior results.
The removal of ammonium ions through ion exchange hasproven to be the simplest part of the job. The technically
challenging aspects come in regenerating the ion exchange bed ;
and disposing of the regenerant. When the clinoptilolite bed
becomes exhausted with ammonia, it is necessary to regenerate
the bed by passing an appropriate regenerant solution there-
through. Heretofore proposed regenerant solutions have been
composed of various concentrations of NaCl or CaC12, and
NaOH or Ca(OH)2 adjusted to a pH of about 10.5. The regen-
erant passes through the exhausted bed to replace the ammonium
ions with sodium or calcium ions. This regeneration also
chemically changes the NH4+ to NH3. Heretofore, it has been
the practice to either discard the regenerant solution after
one use or reuse it only after stripping the ammonia from it.
When the regenerant solution is used only once, it adds sig-
nificantly to the cost of the process. The stripping of
ammonia from the regenerant solution may be accomplished by
.,
; . . :
. .
104ZS75
air stripping, steAam stripping, or electrolysts. A11 of these
stripping procedures are effective in removing ammonia from the
regenerant, but they also add to the cost of the ammonium ion
removal system.
SUMMARY OF THE INVENTION
One aspect of the invention pertains to a method of
regenerating an exhausted zeolite ion exchange bed used for re-
moving ammonium ions from wastewater which comprises passing
a previously used aqueous regenerant solution of an alkali
metal salt, ammonia having an ammonia concentration in the range
of 600 to 8,000 ppm, and an alkali metal hydroxide adjusted to
a pH i~ excess of 10.5 through the bed; and readjusting the pH
of the used regenerant solution to a pH in excess of 10.5 by
adding an alkali metal hydroxide for reuse in a subsequent
regeneration cycle. -~
Another aspect of the invention comprehends a method
of removing ammonium ions from wastewater, which includes passing
.
the wastewater through a zeolite ion exchange bed until a pre-
determined ammonium ion concentration has been reached in the ~-
effluent and discontinuing the flow of wastewater through the
bed. The bed is then regenerated by passing a previously used
aqueous regenerant solution of an alkali metal salt, ammonia
having an ammonia concentration in the range of 600 to 8,000 ppm,
and an alkali metal hydroxide adjusted to a pH in excess of 10.5
through the bed to remove ammonium ions from the bed. The pH of
the used regenerant solution is readjusted to a pH in excess of
10.5 by adding an alkali metal hydroxide for reuse in a subsequent
regeneration cycle.
The invention in another aspect comprehends a method
of eliminating the necessity of removing ammonia from an aqueous
regenerant solution of NaCl, ammonia, and NaOH having an ammonia
concentration in the range of 600 to 8,000 ppm prior to its reuse
in regenerating a zeolite bed wherein the method comprises the
step of adjusting the pH of the regenerant solution to in excess
- 4 -
~......
' , , ~ ' ' ' '~ ' . ~ ,, ',
~042575
of 10.5 by adding NaOH~
More particularly, the present invention is directed
to a method of regenerating a clinoptilolite bed which
eliminates the need for either discarding the regenerant
solution after each use or stripping the ammonia from the
solution prior to each reuse. The regenerant solution of
the present invention consists of approximately a lN NaCl
solution which contains a buffer system of NH40H and NaOH - -~
for pH adjustment. It should be understood that the present
invention contemplates the use of other alkali metal salts
in place of NaCl and other alkal~ metal hydroxides in place
of NaOH. It has been discovered that when the p~ of such
a regenerant is maintained in excess of 10.5, the concentrat-
ion of NH3 in the solution does not materially affect the
ability of the regenerant to remove ammonia from the ~ - -
clinoptilolite. The Na ions displace Ca and Mg ions ~
from the clinoptilolite and the OH in the buffer removes -
NH4 $rom the clinoptilolite by conversion to ~H3. The ~ -
presence of the buf$er system gives the effect of having
greater amounts of available OH at a moderate pH. It -
also permits the use of a smaller regenerant volume since
the resistance of the buffer to changes in pH allows
higher p~ regenerant to be maintained through the bed. In
accordance with the present invention, the regenerant may
be reused as often as desired until the ammonia concentration
reaches a level of approximately 8,000 to 10,000 ppm. This
is in contrast with all other work done in this area which
has discarded the regenerant or stripped the ammonia from
the regenerant either after each run or when the ammonia
concentration has exceeded approximately 600 ppm.
. .
:~.
B ~ 5 ~
.
., ,~, .
, . .
..... . . . . .
104Z575
BRIEF DESCRIPTION OF THE DRAWINGS
The novel features of the present invention are set
forth with particularity in the appended claims. The present
invention, both as to its organization and manner of operation,
may best be understood by way of a general description and a
disclosure of the example as illustrated in the accompanying ~ -
drawings, in which:
FIG. 1 schematically illustrates the equipment used
to test regeneration efficiencv in an exemplary embodiment
in accordance with the present invention;
FIG. 2 iS a graph using the test data from the
example plotting the ammonia leakage in ppm NH3 as N vs. the
number of bed volumes of service flow at the 1,500 ppm NH
concentration level; and ` -~
FIG. 3 is a graph using data from the example plotting -
the average service leakage as ppm NH3 vs. the regenerant NH
concentration.
EXAMPLE AND TEST EMBODIMENT
The invention will be more fully explained and exem-
plified inthe following example and test embodiment, Theexample illustrates the comparative effectiveness of regen-
erating clinoptilolite with regenerants of varying ammonia
concentrations and a constant pH.
The test equipment set up is schematically illustrated
in Fig . 1 . Three acrylic columns (1" by 6'), indicated at 1,
2, and 3, are provided having influent tubes 4, 5 and 6 and
effluent tubes 7, 8 and 9-respectively. Three-way valves 10,
;,..
.
104'~575
, 12, 13, 14, and 15 are provided to control the Elow
through the respective influent and effluent tubes. Three
regene~ant supply tanks 16, 17, and 18 are positioned at
elevations respectively above the valves 10, 11, and 12 for
delivering regenerant therefrom through corresponding tubes
19, 20, and 21, valves 13, 14, and 15, and tubes 7, 8, and
9, into columns 1, 2, and 3. Three used regenerant collection
tanks 22, 23, and 24 are provided to receive the used regen-
erant which flows from columns 1, 2, and 3 through valves 10,
11, and 12 and tubes 25, 26, and 27 thereinto. A large feed
tank 28 supplies feed water through a splitter 29, down
influent tubes 4, 5, and 6 and through valves 10, 11, and 12,
into columns 1, 2, and 3. The feed ~ater exits columns 1, 2,
and 3 and flows through effluent tubes 7, 8~ and 9 and valves
13, 14, and 15 into effluent collection tanks 30, 31, and 32.
Flowmeters 33, 34, and 35 are provided to control the eed
water flow rate through effluent tubes 7, 8, and 9 and con-
sequently through columns 1, 2, and ~. This equipment set up
permits test runs to be conveniently conducted in groups of
three runs.
At the start of each test run, the columns 1, 2, and
3 were packed with clinoptilolite in accordance with the fol-
lowing procedure. The clinoptilolite was received in the
form o~ 20 to 50 mesh particles. The material was packed in
the columns 1, 2, and 3 to a tamped depth oE 36". This gave
a bed volume of 463 cc. They were then backwas~ed to remove
fines. After backwashing, the clinoptilolite was allowed to
settle and was tapped to a constant height (36"+0.5"). Since -
the clinoptilolite was received in an undetermined ionic
form, feed water containing approximately 20 ppm NH3 as N
. ~ - ., , ,:
. .. ~ . .
.' . ' ' . ,, ' : . . .
.
, ..: ' . ,' ~ . ~:; .
104Z5~5
direc~ed through fe~d tank 28, spli~ter 29, influen~
tubes 4, 5, and 6, columns 1, 2, and 3, and effluent tubes
7, 8, and 9 into effluent collection tanks 30, 31, and 32
until the NH3 concentration entering tanks 30, 31, and 32
equaled the NH3 concentration of the feed water. This ensuredi
that the clinop~ilolite was in an exhausted condition with
respect to NH3.
A regenerant was prepared which consisted of a one
normal NaCl solution in which was dissolved a predetermined
weight of NH4Cl. The amounts of NH4Cl added to the regen-
erant were set up to produce five different levels of ammonia
in the regenerant; O ppm, 50 ppm, 500 ppm, 1,500 ppm, and
3,000 ppm. Four liter batches of these solutions were titrated
with 20% NaOH to a pH 11 end point. Three batches of regen-
erant at each ammonia concentration ievel were prepared in -~
this manner. ~ -
The following test procedure was conducted for each
ammonia concentration level of regenerant. Columns 1, 2,
and 3 were packed with exhausted clinoptilolite in accordance
with the above mentioned procedu~e. The regenerant at each
concentration level was introduced into the regenerant tanks
16,~ 17, and 18 and directed through the corresponding tubes
19, 20, and 21 and 7, 8, and 9 into the bottom of columns 1, -
2, and 3 at a flow rate o 50 ml/min. The spent regenerant
leaving the top oE columns 1, 2, and 3 passed through the
corresponding tubes 4, 5, and 6 and 25, 26, and 27 and was
collected in the used regenerant tanks 22, 23, and 24~ The
equivalent of 6.5 bed volumes of regenerant was collected in
each of the used regenerant tanks 22, 23, and 24. Due to
dead space in the columns l, 2, and 3 over the clinoptilolite,
-- 8 --
.
1~)4'~:S75
~his amounted to approximately 7.7 bed volumes thro-lgh the
bottom of the bed. Feed water, having an ammonia concentra-
tion of 19.1 ppm, was then directed from feed tank 28 through
the splitter 29 and tubes 4, 5, and G and then through the
corresponding colu~ls 1, 2, and 3 and tubes 7, 8, and 9 Eor
collection in effluent tanks 30, 31, and 32. The fLow of
rinse water was continued through th columns until the
sampled effluent water dropped to an ammonia concentration
of 3 ppm. The number of bed volumes of feed water required
to so rinse the columns 1, 2, and 3 was recorded. The feed
water was then continued through thè columns 1, 2, and 3
until the ammonia concentration again reached the 3 ppm
level. The effluent from the columns 1, 2, and 3 was period-
ically sampled for ammonia concentra~ion and recorded along
with the corresponding bed volume count until the ammonia
concentration level rose past 3 ppm. The ammonia concentra-
tions were measured by ASTM D 1426 non referee.
Graphs plotting the ppm NH3 ln the effluent vs.
number of bed volumes of service flow at each regenerant
;
concentration level best indicate the results of the three
test runs at each level; an example of which is illustrated
in FIG. 2 for the 1,500 ppm NH3 concentration level. A curve
is sketched to approximate the relationship be~een the
ammonia leakage in the effluent vs. the run length of thè
column. The average effluent ammonia leakage was then cal-
culated over the run length from the point where the ammonia
concentration decreased to 3 ppm during the rinse down cycle
until it increased to 3 ppm during the service cycle.
The results of the above test are summari~ed in ~he
following table:
. .~ .
:, ........................ .
`'` ` ~04;~S75
COLU~ PE~FO~ANCE DArA
ppm NH3 Average LeakageAverage Run Length
In Re~enerantppm as N Bed Volumes
0 0.76 185
0.76 186
500 0.85 180
1,500 0.92 185
3,000 0.95 184
Referring to FIG. 3, a graph of the above data,
plotting the average service leakage (ppm NH3) vs. the
regenerant NH3 concentration, follows a logarithmic relation-
ship of the form: ~
ammonia leakage (ppm) = 0.113 log (NH3 as N (ppm)) + 0.558 -~;
Between 50 and 3,000 ppm, this relation is linear with a
standard deviation of 0.01 ppm. Usi~g sound engineering
procedures, this relationship can be extrapolated to show
that the regenerant can reach a level of 8,000 to 10,000 ppm
ammonia concentration without exceeding an average ammonia -;
leakage of 1.0 ppm over the test run. It should be noted -~
that, if the average ammonia leakage acceptable is greater ~
than 1.0 ppm, then the level of NH3 in the regenerant may be `-.
increased in accordance with the above formula. This extrap-
olation is shown by an extension of the straight line in FIG. 3.
; In summary, it has been discovered that by maintaining
the pH of the regenerant in excess of 10.5, the concentration
of NH3 in the solution may reach approximately 8,000 to 10,000
ppm without resulting in an average ammonia leakage in excess
of 1.0 ppm over the length of the service run. It should
also be noted tha~ the average length of the service run is
not materially affected by the concentration of NH3 in the
regenerant solution.
- 10 -
,~3 ' .
:; , . ' . . ' ' " . . ' . , . ` '
104Z~75
It should be understood that the present disclosure
is for the purpose oE illustration only~ and that this inven-
tion includes all modifications and equivalents which fall
within the true spirit and scope oE the inven~ion as defined
by the appended claims.
. .
-- 11 --
r;
'''' .. ' ' ' ~' ':,, ~ . . ,
',
