-------
SECTION VII
CONTROL AND TREATMENT TECHNOLOGY
. .. ;
-2-vsa-S
END-QF-PIPE TREATMENT TECHNOLOGIES
demonstrated
r -^'
135
-------
Discussion of end-of-pipe treatment technologies is divided into
Sh?SS pa?t2i the majo? technologies; the effectiveness of ma3or
technologies; and minor end-of-pipe technologies.
MAJOR TECHNOLOGIES
5
^
affected by the precipitation operation and are
^^
combination with a solids removal operation.
1. Chemical Reduction Of Chromium
Ttocsrrintion of the Process. Reduction is a chemical reaction in
S5g£ eiSctrons^are-Transf erred to the chemical being reduced
frSSthS chemical initiating the transfer (the reducing agent).
Sulfur dioxide, sodium bisulfite, sodium metabisulf ite, and
ferrous sulf ate form strong reducing agents in aqueous solution
and are often used in industrial waste treatment facilities for
valent £°™ The
an are
?Ke reducton of hexavalent chromium to the trivalent .
reduction allows removal of chromium from solution in C°n3u^tion
with other metallic salts by alkaline precipitation. Hexavalent
chromium is not precipitated as the hydroxide.
esgo of
Reduction using othe/ reagents is chemically similar. The
reactions involved may be illustrated as follows:
3 SO2 + 3 H2O ---- > 3 H2SO3
3 H2S03 + 2H2Cr04 ---- > Cr2(S04)3 + 5 H20
The above reaction is favored by low pH. A pH of from 2 to 3 is
normal for situations requiring complete, reduction. At pH levels
above 5, thS reduction rate is slow. Oxidizing agents such as
136
-------
process^ minheduc^ "ith "» reduction
dioxide is metered to the rearf?™ J-ii, Z ' * . Gaseo"s sulfur
.
shows a continuous chromiumreducion'systlm
L
0.05 mg/1 are readily ttairt^X lnal c°"«ntrations of
are conlldered II be attflnlble LC°nCentr?tions °f °-01 mS/l
operated equipment Becauie JL yh Pf°P?rly maintained and
'
equipment Becaue h
chromium conve?s Ion ' coaM ngs Ire sLijSr""1^1 =ys^em? usfid £or
of chromium Is applicable"?! llnmaW^^istewaJeS^31 ceduc«°"
ssa. --ssga w
obtainable from many
ttat forhghcontra?Tonaoferhr™?n °f,he«a^le"t chromium is
chemicals may be p?ohlbl?iw ShfS J2?f '^ cost of treatment
^
dioxide iS somewhat hazlrdous' Stora«e and handling of sulfur
137
-------
practicing chromium reduction.
2. Chemical Precipitation
are commonly used to effect this precipitation:
eecp
Hydroxides. P[?mePalso may precipitate phosphates as calcium
phosphate and fluorides as calcium fluoride.
2) Both "soluble" sulfides such as hydrogen sulfide ^ sodium
2> sutfidf and "insoluble" sulfides such as ^roussul fide may
be used to precipitate many heavy metal ions as metal
sulfides.
3) Ferrous sulfate, zinc sulfate or both (as is ^quired) may
• be used to precipitate cyanide as a ferro or zinc
ferricyanide complex.
i> Carbonate precipitates may be used to remove metals either
' by direct ^precipitation using a carbonate reagent such as
calcium carbonate or by converting hydroxides into
carbonates using carbon dioxide.
These treatment chemicals may be added to a flash
e
idjS?men may b4 required to reduce the high pH created by the
alkaline treatment chemicals.
Chemical precipitation as a mechanism for removing metals from
-a SHE~«.
SStewatSafS? precipitation is complete. The amount of
138
-------
a-
a
^EEiication and Performance. Chemical precipitation is used in
canmaking for precipitation of dissolved metals. It can be u^S
1. Maintenance of an alkaline PH throughout the
precipitation reaction and subsequent settling?
2' ^fi^io|? ofa sufficient excess of treatment ions to
drive the precipitation reaction to completion;
3* afdiiron ojra" fde?uat? s«PPly of' sacrif leal ions (such
as iron or aluminum) to ensure precipitation and
removal of specific target ions; and ^^"on ana
4. Effective removal of precipitated solids (see
Smovll"? technol°9i€s discussed under "Solids
Control of pji. Irrespective of the solids removal technoloav
r°er contro1 of PH is absolutely essenMa? f or
, ?f - , , Palpitation-sedimentation
^
-obtained from
MMeline and New Source PeTfori^
o^ n r.
TS; Si'S^V^" ™. "0/1-74/033, November, 19747
-3 was plotted from the sampling data from sev4ral facilitiel
139
-------
with metal finishing operations. It is partially illustrated by
data obtained from 3 consecutive days of sampling at one metal
processing plant (47432) as displayed in T^le VII- page 216)
Flow through this system is approximately 49,263 1/hr (I3,uuu
gal/hr).
This treatment system uses lime precipitation (pH adjustment)
followed by coagulant addition and sedimentation. Samples were
taken before (in) and after (out) the treatment system. The best
treatment for removal of copper and zinc was achieved on day^one,
when the pH was maintained at a satisfactory level. The poorest
treatment was found on the second day, when the pH slipped to an
unacceptably low level; intermediate values were achieved^on the
third day when pH values were less than desirable but in between
those of the first and second days.
Sodium hydroxide is used by one facility (plant 439) for pH
adjustment and chemical precipitation followed by settling
(sedimentation and a polishing lagoon) of P^P^ed solids
Samples were taken prior to caustic.addition and following the
polishing lagoon. Flow through the system is approximately
22,700 1/hr (6,000 gal/hr) displayed in Table VII-2 (page 216).
These data for this plant indicate that the system was operated
efficiently. Effluent pH was controlled within the range of 8.6-
9.3 and, while raw waste loadings were not unusually high, most
toxic metals were removed to very low concentrations.
Lime and sodium hydroxide (combined) are sometimes used to
precipitate metals. Data developed from plant 40063, a facility
with a metal bearing wastewater, exemplify efficient operation of
a chemical precipitation and settling system. Table VII?3 (page
217) shows sampling data from this system, which uses lime and
sodium hydroxide for pH adjustment and chemical precipitation,
polyelectrolyte flocculant addition, and sedimentation. Samples
were taken of the raw waste influent to the system and of the
clarifier effluent. Flow through the system is approximately
19,000 1/hr (5,000 gal/hr).
At this plant, effluent TSS levels were below 15 mg/1 on each
day, despite average raw waste TSS concentrations of over 3500
mq/i. Effluent pH was maintained at approximately 8, lime
addition was sufficient to precipitate the dissolved metal ions,
and the flocculant addition and clarifier retention served to
remove effectively the precipitated solids.
Sulfide Precipitation is sometimes used to precipitate metals
resulting—in improved metals removals. Most metal sulfides are
less soluble than hydroxides and the precipitate^are frequently
more dependably removed from water. Solubilities for selected
metal hydroxide, carbonate and sulfide precipitates are shown in
140
-------
-
Cr03 + FeS f 3H20 - --- > Fe(OH)3 + Cr(OH)3 + S
.
The solubxUt most metal carbonates is intermediate befwlen
141
-------
hydroxide and sulfide solubilities,, in addition, carbonates form
easily filtered precipitates.
phenomenon .
tomprove the remova! of toxic etals
In so*e
nf treatment. The iron functions to improve toxic metal
been used.
Co-precipitation using large amounts of ferrous iron salts is
known afferrite co-precipitation because magnetic iron oxide or
ferrite co-precipitation is shown in Table VII-7 (page 22U).
Advantages and Limitations. Chemical precipitation^as^proven^to
=— ^-- : - •~ for removing many
operates at ambient
blockinq of the lines, which may result from a buildup of solids.
hyd?oxide precipitation usually makes recovery of the
142
-------
precipitated metals difficult, because of the heterogeneous
nature of most hydroxide sludges.
The major advantage of the sulf ide precipitation process is that
the extremely low solubility of most metal sulf ides promotes very
high metal removal efficiencies; the sulf ide process also has the
ability to remove chromates and dichromates without preliminary
reduction of the chromium to its trivalent state. in addition,
sulf ide can precipitate metals complexed with most complexing
agents. The process demands care, however, in maintaining the pH
of the solution at approximately 10 in order to prevent the gen-
eration of toxic hydrogen sulf ide gas. For this reason,
ventilation of the treatment tanks may be a necessary precaution
in most installations. The use of insoluble sulf ides reduces the
problem of hydrogen sulf ide evolution. As with hydroxide
precipitation, excess sulf ide ion must be present to drive the
precipitation reaction to completion. Since the sulf ide ion
itself is toxic, sulfide addition must be carefully controlled to
maximize heavy metals precipitation with a minimum of excess
sulfide to avoid the necessity of additional wastewater
treatment. At very high excess sulfide levels and high pH
soluble mercury-sulfide compounds may also be formed. Where
excess sulfide is present, aeration of the effluent stream can
* / residual sulfide to the less harmful sodium
sulf ate (Na2S04). The cost of sulfide precipitants is high in
comparison with hydroxide precipitants, and disposal of metallic
sulfide sludges may pose problems. An essential element in
effective sulfide precipitation is the removal of precipitated
solids from the wastewater and proper disposal in an appropriate
Si , * Sulfide precipitation will also generate a higher volume
of sludge, than hydroxide precipitation, resulting in higher
disposal and dewatering costs. This is especially true when
ferrous sulfide is used as the precipitant. '
Sulfide precipitation may be used as a polishing treatment after
hydroxide precipitation-sedimentation. This treatment
configuration may provide the better treatment effectiveness of
sulfide precipitation while minimizing the variability caused by
changes in raw wastewater composition and reducing the amount of
sulfide precipitant required.
Operational Factors. Reliability: Alkaline chemical
precipitation is .highly reliable, although proper monitoring and
similar reliabilit*"*' Sulfide Precipitation systems provide
Maintainability: The major maintenance needs involve periodic
upkeep of monitoring equipment, automatic feeding equipment,
mixing equipment, and other hardware. Removal of accumulated
143
-------
sludge is necessary for efficient operation of precipitation-
sedimentation systems.
Solid Waste Aspects: Solids which precipitate out are removed in
a subsequent treatment step. Ultimately, these solids require
proper disposal.
Demonstration Status. Chemical precipitation of metal hydroxides
ii—a—classic wastewater treatment technology used by most
industrial wastewater treatment systems. Chemical precipitation
of metals in the carbonate form alone has been found to be
feasible and is commercially used to permit metals recovery and
water reuse. Full scale commercial sulfide precipitation units
are in operation at numerous installations. As noted earlier,
sedimentation to remove precipitates is discussed separately.
Use in Canmakinq Plants. Chemical precipitation equipment is in
place at 42 canmaking plants.
3. Cyanide Precipitation
Cyanide precipitation, although a method for treating cyanide in
wastewaters, does not destroy cyanide ThVy3?i %h*i- dur?na
in the sludge that is formed. Reports indicate that during
exposure to sunlight the cyanide complexes can break down and
form free cyanide. For this reason the sludge from this
treatment method must be disposed of carefully.
Cyanide may be precipitated and settled out of wastewaters by the
addition of zinc sulfate or ferrous sulfate. In the presence of
iron, cyanide will form extremely stable cyanide complexes. The
addition of zinc sulfate or ferrous sulfate forms zinc
ferrocyanide or ferro and ferricyanide complexes.
Adequate removal of the precipitated cyanide requires that the pH
must be kept at 9.0 and an appropriate retention time be
maintained. A study has shown that the formation of the complex
is very dependent on pH. At pH's of 8 and 10 the residual
cyanide concentrations measured are twice those of the same
reaction carried out at a pH of 9. Removal efficiencies also
depend heavily on the retention time allowed. The formation of
the complexes takes place rather slowly. Depending upon the
excess amount of zinc sulfate or ferrous sulfate added, at least
f30 minute retention time should be allowed for the formation of
the cyanide complex before continuing on to the clarification
stage.
One experiment with an initial concentration of 10 mg/1 of
cyanide showed that 98 percent of the cyanide was completed ten
minutes after the addition of ferrous sulfate at twice the
144
-------
theoretical amount necessary. Interference from other metal
retentioTtimes. CadmiUm' mi9ht reSUlt in the "^d for longer
Vcol! 'mr^KSr YSSiS!
-' *
ti n.
because, (1) theH was usuall
^ was usually well below the optimum level of
fA2;<=i-lnJi S ^storical treatment data were not obtained using
™?«? J 5 I cyanide analysis procedure; and (3) matched input?
output data were not made available by the plant Scannina the
available da^ indicates that the raw waste CN level was In* the
of 25.0; the PH 7.5; and treated CN level was from 0.1 to
range
\J • 4* •
The concentrations are those of the stream entering and leaving
tiL fn.afhen£ sy?tem- ' Plant 1057 allowed a 27 minute retention:
JiS VH f f°rmatlon of the comPlex. The retention time for
SS« J Plants is not known. The data suggest that over a wide
range of cyanide concentration in the raw wastewater, the
CYanide Can be reduced in the effluent stream to
and Performance. Cyanide precipitation can be used
when cyanide destruction is not feasible because of the presence
of cyanide complexes which are difficult to destroy. Effluent
concentrations of cyanide well below 0.15 mg/1 are possible.
Advantages and Limitations. Cyanide precipitation is an
inexpensive method of treating cyanide. Problems may occur %hS
metal ions interfere with the formation of the complexes.
Status; Cyanide precipitation is used in at least
4. Granular Bed Filtration
Filtration occurs in nature as the surface ground waters are
Si™a^ed-iby Sa2d* Silica sand' anthracite coal, and garnet are
common filter media used in water treatment plants. These are
usually supported by gravel. The media may be used singly or in
reTaMv^r^' J^ r}ti"media filters »a* b^ arranged to^aintai'n
nriS^i ?i dlstinct layers by virtue of balancing the forces of
f^™ i • K 2W'u buoyancy on the individual particles. This is
accomplished by selecting appropriate filter flow rates (gpm/sq-
ft)7 media grain size, and density. wt«»/=>4
145
-------
Granular bed filters may be classified in terms of filtration
rail, filter media, flow pattern, or method 6f P?«8«"»tlo!k
Traditional rate classifications are slow sand, rafjj *anf? *£*
high rate mixed media. In the slow sand filter, ""* °r
hvdraulic loading is relatively low, and removal of collected
solids to clean the filter is therefore relatively infrequent
The filter is often cleaned by scraping off the inlet face (top)
of the sand bed. In the* higher rate filters, cleaning is
frequent and is accomplished by a periodic backwash, opposite to
the direction of normal flow.
A filter may use a .single medium such as sand or diatomaceous
ea^ but Sual and mixed Tmultiple) media filters a low higher
flow rates and efficiencies. The dual media filteu usually
consists of a fine bed of sand under a coarser bed of anthracite
e
ST
and 'anthracite coal at the top. Some mixing of these layers
occurs and is, in fact, desirable.
The flow pattern is usually top-to-bottom, but other patterns are
sometimes used. Upflow filters are sometimes used and in a
horizontal filter the flow is horizontal. In a biflow filter,
the influent enters both the top and the bottom and exits
laterally The advantage of an upflow filter is that with an
upflow backwash the particles of a single filter medium are
distributed and maintained in the desired coarse-to-fine (bottom-
to- top) arrangement. The disadvantage is that the bed tends to
become fluidized, which ruins filtration efficiency. The biflow
design is an attempt to overcome this problem.
The classic granular bed filter operates by gravity flow;
however? pressure filters are fairly widely used. They permit
higher solids loadings before cleaning and are advantageous when
thl filter effluent must be pressurized for further downstream
treatment. In addition, pressure filter systems are often less
costly for low to moderate flow rates.
Figure VII-14 (page 250) depicts a high rate, dual media, gravity
downflow granular bed filter, with self-stored backwash. Both
filtratl 9and backwash are piped around the bed in an arrangement
that permits gravity upflow of the backwash, with the stored
filtrate se?ving a! backwash. Addition of the indicated
146
-------
coagulant and polyelectrolyte usually results
improvement in filter performance.
in a substantial
Auxiliary filter cleaning is sometimes employed in the upper few
inches of filter beds. This is conventionally referred to as
surface wash and is accomplished by water jets just below the
surface of the expanded bed during the backwash cycle. These
jets enhance the scouring action in the bed by increasina the
agitation.
An important feature for successful filtration and backwashing is
the underdrain. This is the support structure for the bed. The
underdrain provides an area for collection of the filtered water
without clogging from either the filtered solids or the media
grains. In addition,, the underdrain prevents loss of the media
with the water, and during the backwash cycle it provides even
flow distribution over the bed. Failure to dissipate the
velocity head during the filter or backwash cycle will result in
bed upset and the need for major repairs.
Several standard approaches are employed for filter underdrains.
The simplest one consists of a parallel porous pipe imbedded
under a layer of coarse gravel and manifolded to a header pipe
for effluent removal. Other approaches to the underdrain system
are known as the Leopold and Wheeler filter bottoms. Both of
these incorporate false concrete bottoms with specific porosity
configurations to provide drainage and velocity head dissipation.
Filter system operation may be manual or automatic. The filter
backwash cycle may be on a timed basis, a pressure drop basis
with a terminal value which triggers backwash, or a solids carry-
over basis from turbidity monitoring of the outlet stream. All
of these schemes have been used successfully.
Application and Performance. Wastewater treatment plants often
use granular bed filters for polishing after clarification,
sedimentation, or other similar operations. Granular bed
filtration thus has potential application to nearly all
industrial plants. Chemical additives which enhance the upstream
treatment equipment may or may not be compatible with or enhance
the filtration process. Normal operating flow rates for various
types of filters are as follows:
Slow Sand
Rapid Sand
High Rate Mixed Media
2.04 - 5.30 1/sq m-hr
40.74 - 51.48 1/sq m-hr
81.48 - 122.22 1/sq m-hr
Suspended solids are commonly removed from wastewater streams by
filtering through a deep 0.3-0.9 m (1-3 feet) granular filter
bed. The porous bed formed by the granular media can be designed
147
-------
to remove practically all suspended particles. Even colloidal
suspensions (roughly 1 to 100 microns) are adsorbed on the
surface of the media grains as they pass in close proximity in
the narrow bed passages.
Properly operated filters following some pretreatment to reduce
suspended solids below 200 mg/1 should produce water with less
than 10 mg/1 TSS. For example, multimedia filters produced the
effluent qualities shown in Table VII-9 (page 221).
The principal advantages of granular bed filtration are its
comparatively (to other filters) low initial and operating costs,
reduced land requirements over other methods to achieve the same
level of solids removal, and elimination of chemical additions to
the discharge stream. However, the filter may require
pretreatment if the solids level is high (over 100 mg/1).
Operator training must be somewhat extensive due to the controls
and periodic backwashing involved, and backwash must be stored
and dewatered for economical disposal.
Operational Factors. Reliability: The recent improvements in
filter technology have significantly improved filtration
reliability. Control systems, improved designs, and good
operating procedures have made filtration a highly reliable
method of water treatment.
Maintainability: Deep bed filters may be operated with either
manual or automatic backwash. In either case, they must be
periodically inspected for media attrition, partial plugging, and
leakage. Where backwashing is not used, collected solids must be
removed by shoveling, and filter media must be at least partially
replaced.
Solid Waste Aspects: Filter backwash is generally recycled
within the wastewater treatment system, so that the solids
ultimately appear in the clarifier sludge stream for subsequent
dewatering. Alternatively, the backwash stream may be dewatered
directly or, if there is no backwash, the collected solids may be
disposed of in a suitable landfill. In either of these
situations there is a solids disposal problem similar to that of
clarifiers.
Demonstration Status. Deep bed filters are in common use in
municipal treatment plants. Their use in polishing industrial
clarifier effluent is increasing, and the technology is proven
and conventional. Granular bed filtration is used in many
manufacturing plants. As noted previously, however, little data
is available characterizing the effectiveness of filters
presently in use within the industry. However, 3 canmaking
148
-------
Siv?S- T? 9ra"ujar bed filtration equipment in-place as
polishing filters before discharging treated wastewater.
5. Pressure Filtration
Pressure filtration works by pumping the liquid through a filter
material which is impenetrable to the solid phase. The positive
nrnvf^6 fherted by th!-ffed pumps ^ • other mechanical meanl
provides the pressure differential which is the principal drivinq
1°pfof p£S£.Vi11I&.(p— "" repreSe"tS •"••"oP-r.tion of onS
A typical pressure filtration unit consists of a number of plates
2«* E? K are held rigidly in a frame to ensure alignment
and which are pressed together between a fixed end and a
traveling end. On the surface of each plate- is mounted a filter
made of cloth or a synthetic fiber. The feed stream is pumped
into the unit and passes through holes in the trays along the
length of the press until the cavities or chambers between the
trays are completely filled. The solids are then entrapped, and
a cake begins to form on. the surface of the filter material The
water passes through the fibers, and the solids are retained
A 3- K * 5rayS are draina9€ PPrts. The filtrate is
? ^ discharged to a common drain. As the filter medium
becomes coated with sludge, the flow of filtrate through the
filter drops sharply, indicating that the capacity of the filter
has been exhausted. The unit must then be cleaned of the sludge.
After the cleaning or replacement of the filter media, the unit
is again ready for operation.
Application and Performance. Pressure filtration is used in coil
S™^?i- J°5 SiU(% dewaterin
-------
than that from centrifuge or vacuum filter. Thus, it can be
easily accommodated by materials handling systems.
As a primary solids removal technique, pressure filtration
requires less space than clarification and is well suited to
streams with high solids loadings. The sludge produced may be
disposed without further dewatering, but the amount of sludge is
increased by the use of filter precoat materials (usually
diatomaceous earth). Also, cloth pressure filters often do not
achieve as high a degree of effluent clarification as clarifiers
or granular media filters.
Two disadvantages associated with pressure filtration in the past
have been the short life of the filter cloths and lack of
automation. New synthetic fibers have largely offset the first
of these problems. Also, units with automatic feeding and
pressing cycles are now available.
For larger operations, the relatively high space requirements, as
compared to those of a centrifuge, could be prohibitive in some
situations.
Operational Factors. Reliability: With proper pretreatment,
design, and control, pressure filtration is a highly dependable
system.
Maintainability: Maintenance consists of periodic cleaning or
replacement of the filter media, drainage grids, drainage piping,
filter pans, and other parts of the system. If the removal of
the sludge cake is not automated, additional time is required for
this operation.
Solid Waste Aspects: Because it is generally drier than other
types of sludges, the filter sludge cake can be handled with
relative ease. One of several accepted procedures may be used to
dispose of the accumulated sludge, depending on its chemical
composition. The levels of toxic metals present in sludge from
treating canmaking wastewater necessitate proper disposal.
Demonstration Status. Pressure filtration is a commonly used
technology in a great many commercial applications.
6. Settling
Settling is a process which removes solid particles from a liquid
matrix by gravitational force. This is done by reducing the
velocity of the feed stream in a large volume tank or lagoon so
that gravitational settling can occur. Figure VII-16 (page 252)
shows two typical settling devices.
150
-------
Settling is often preceded by chemical precipitation which
converts dissolved pollutants to solid form and by coagulation
which enhances settling by coagulating suspended precipitates
into larger, faster settling particles. ««.ipitaces
If no chemical pretreatment is used, the wastewater is fed into a
tank or lagoon where it loses velocity and the suspended solids
P
O ^^ ^ut' The "te of settling isdefned by an
equation known as Stokes1 Law. Long retention times
:Tfh2reral-Veq?,ired- Accumulated sludge can be co?lec?Id
either periodically or continuously and either manually or
mechanically. Simple settling, however, may require excessively
alum
?™™a?tiC€' chemical Precipitation often precedes settling, and
inorganic coagulants or polyelectrolytic flocculants are usually
afd™d« fS W6J Common coagulants include sodium sulfate, sodium
aluminate, ferrous or ferric sulfate, and ferric chloride
?a?a^%1P°lyeleCtl;0lyJeS Vary in structure, but all Usually form
larger, floe particles than coagulants used alone.
Following this pretreatment, the wastewater can be fed into a
orf Ia9°on for settling, but is more often piped into
tor^ the same Purpose. A clarifier reduces space
ro • reduce^ ^tention time, and increases solid!
removal efficiency. Conventional clarifiers generally consist of
?«nSi^ 5 .or rectangular tank with a mechanical sludge
for liudn? ™y?CVr Witr a sl°Pin9 funnel-shaped bottom designed
for sludge collection. In advanced settling devices inclined
Sifhfn'th? M^V^63^ ?r a lamellar network may be included
within the clarifier tank in order to increase the effective
a£fa' in?reasing capacity. A fraction of the sludge
' formation of
aS£L.per£?""ance. Settling and clarification are used
- Canma ung industry to remove precipitated metals.
ing can be used to remove most suspended solids in a
particular waste stream; thus it is used extensively by many
SiS61^! ^n<3ust?Jal wastewater treatment facilities. Because
most metal ion pollutants are readily converted to solid metal
hydroxide precipitates, settling is of particular use in those
^d"friee.associated with metal Production, metal finishing?
moJU working, and any other industry with high concentrations of
^?«hii°^ -"-I- ?e^r wastewaters. In addition to toxic metals,
suitably precipitated materials effectively removed by settling
151
-------
include aluminum, iron, manganese, cobalt, antimony, beryllium,
molybdenum, fluoride, phosphate, and many others.
A properly operating settling system can efficiently remove
suspended solids, precipitated metal hydroxides, and .other
impurities from wastewater. The performance of the process
depends on a variety of factors, including the density and
particle size of the solids, the effective charge on the
suspended particles, and the types of chemicals used in
pretreatment. The site of flocculant or coagulant addition also
may significantly influence the effectiveness of clarification.
If the flocculant is subjected to too much mixing before entering
the clarifier, the complexes may be sheared and the settling
effectiveness diminished. At the same time, the flocculant must
have sufficient mixing and reaction time in order for effective
set-up and settling to occur. Plant personnel have observed that
the line or trough leading into the clarifier is often the most
efficient site for flocculant addition. The performance of
simple settling is a function of movement rate, retention time,
particle size and density, and the surface area of the basin.
The data displayed in Table VII-10 (page 221) indicate suspended
solids removal efficiencies in settling systems.
The mean effluent TSS concentration obtained by the plants shown
in Table VII-10 is 10.1 mg/1. Influent concentrations averaged
838 mg/1. The maximum effluent TSS value reported is 23 mg/1.
These plants all use alkaline pH adjustment to precipitate metal
hydroxides, and most add a coagulant or flocculant prior to
settling.
Advantages and Limitations. The major advantage of simple
settling is its simplicity as demonstrated by the gravitational
settling of solid particulate waste in a holding tank or - lagoon.
The major problem with simple settling is the long retention time
necessary to achieve complete settling, especially if the
specific gravity of the suspended matter is close to that of
water. Some materials cannot be practically removed by simple
settling alone.
Settling performed in a clarifier is effective in removing slow-
settling suspended matter in a shorter time and in less space
than a simple settling system. Also, effluent quality is often
better from a clarifier. The cost of installing and maintaining
a clarifier, however, is substantially greater than the costs
associated with simple settling.
Inclined plate, slant tube, and lamella settlers have even higher
removal efficiencies than conventional clarifiers, and greater
capacities per unit area are possible. Installed costs for these
152 --
-------
advanced clarification systems are claimed to be one half the
cost of conventional systems of similar capacity.
Operational Factors Reliability: Settling can be a highly
reliable technology for removing suspended solids. Sufficient
aff-HrJJ!" hime a?d F^?Vjar sludae removal are important factors
affecting the reliability of all settling systems Prooer
control of pH adjustment, chemical precipitltionfand coagulant
or flocculant addition are additional factors affecting settling
J.thodTSrru.id. SySt€mS (fre^uently clarifiers) where thesl
settjers Usin9 slanted tubes, inclined plates, or
. net"ork may require pre-screening of the waste in
M™ fh iminate a"y fibrous materials which could potentially
snock foSdfnS' SomVnSt?llati°nS are especially vulnerable to
design illl pSienfthS. ^ «*«-. "»*>«< but Proper system
Maintainability: When clarifiers or other advanced settlina
devices are used, the associated system utilized for chemical
pretreatment and sludge dragout must be maintained on a rSguJar
n™!; RouTtine maintenance of mechanical parts is also
llttle «^ten.nceP other
Demonstration Status
m rePresents .the typical method of solids removal and is
SSSiSfS extens^yely in industrial wastewater treatment. The
advanced clarifiers are just beginning to appear in significant
ClanSSnra.t-COinmerCJ^- aPPlicati^. Twen?? three ?anmak?ng
?i£t!c S'lli11 °f thSSe u~ ""ling following
7. Skimming
f?bitta.m«lf2J SfeCi5i° aravifcy less tha" water will often
tipat unassisted to the surface of the wastewater Skimminn
removes these floating wastes. Skimming normaf ly ?ak5s pface "%
a tank designed to allow the floating debris to rise and remain
?h* ??«S^na
-------
floating oil which escapes the drum skimmer. The belt type
skimmer is pulled vertically through the water, collecting oil
which is scraped off from the surface and collected in a drum
Gravity separators, such as the API type, utilize overflow and
underflow biffles to skim a floating oil layer from the surface
of the wastewater. An overflow-underflow baffle allows a small
amount of wastewater (the oil portion) to flow over into a trough
for disposition or reuse while the majority of the water flows
underneath the baffle. This is followed by an overflow baffle,
which is set at a height relative to the first baffle such that
only the oil bearing portion will flow over the first baffle
during normal plant operation. A diffusion device, such as a
vertical slot baffle, aids in creating a uniform flow through the
system and increasing oil removal efficiency.
Application and Performance. Lubricants cleaned from most
seamless cans dUFing the canwashing process are the Principal
Source of oil. Skimming is applicable to any wastewater stream
containing pollutants which float to the surface. It is commonly
used to remove free oil and grease. Skimming is often used in
conjunction with air flotation or clarification in order to
increase its effectiveness.
The removal efficiency of a skimmer is partly a function of the
retention time of the water in the tank. Larger, more buoyant
particles require less retention time than smaller particles.
Thus the efficiency also depends on the composition of the waste
stream. The retention time required to allow phase separation
and subsequent skimming varies from 1 to 15 minutes, depending on
the wastewater characteristics.
API or other gravity-type separators tend to be more suitable for
use where the amount of surface oil flowing through the system is
consistently significant. Drum and belt type skimmers are
applicable to wastewater streams which evidence smaller amounts
of floating oil and where surges of floating oil are not a
problem. Using an API separator system in conjunction with a
drum type skimmer would be a very effective^ method of removing
floating contaminants from non-emulsified oily waste streams.
Sampling data illustrate the capabilities of the technology with
both extremely high and moderate oil influent levels.
These data, displayed in Table VII-11 (page 222), are intended to
be illustrative of the very high level of oil and grease removals
attainable in a simple two stage oil removal system. Based on
the performance of installations in a variety of manufacturing
plants and permit requirements that are constantly achieved, it
is determined that effluent oil levels may be reliably reduced
below 10 mg/1 with moderate influent concentrations. Very high
influent concentrations of oil such as the 22 percent shown in
154
-------
the Table for plant 06058 may require two step treatment in order
to achieve 10 mg/1 in the treated effluent.
Skimming which removes oil may also be used to remove base levels
of organics. Plant sampling data show that many organic
compounds tend to be removed in standard wastewater treatment
equipment. Oil separation not only removes oil but also orqanics
that are more soluble in oil than in water. Clarification
removes organic solids directly and probably removes dissolved
organics by adsorption on inorganic solids. .
The source of these organic pollutants is not always known with
certainty, although in metal forming operations they seem to
derive mainly from various process lubricants. They are also
sometimes present in the plant water supply, as additives to
proprietary formulations of cleaners, or due to leaching from
plastic lines and other materials.
High molecular weight organics in particular are much more
soluble in organic solvents than in water. Thus they are much
more concentrated in the oil phase that is skimmed than in the
wastewater. The ratio of solubilities of a compound in oil and
water phases is called the partition coefficient. Table VII-12
(page 223) lists the logarithm of the partition coefficients in
?£xo? water for selected polynuclear aromatic hydrocarbon
(PAH) compounds and for other organic compounds found in
canmaking wastewaters.
A review of toxic organic compounds found in metal forming
wastewater streams indicates that removal of these compounds
often occurs as a result of oil removal or clarification
processes. When all available organics analyses from aluminum
forming, copper forming, and coil coating are considered, removal
of organic compounds appears to be marginal by waste treatment
technologies other than oil removal or clarification^ Organics
removal as a result of oil removal becomes especially apparent
when raw waste concentrations of organics are above 0.05 mg/1
but are also demonstrated when raw waste concentrations are less
than this value. The API oil-water separation system performed
notably in this regard, as shown in Table VII-13 (page 224)
When these factors are taken into account, analysis data indicate
tVat._most clarification and oil removal treatment systems remove
significant amounts of the organic compounds present in the raw
wastewater. .
Data from five plant days demonstrate removal of organics by the
combined oil skimming and settling operations performed on coil
coating wastewaters. Days were chosen where treatment system
influent and effluent analyses provided paired data points for
oil and grease and the organics present. All organics found at
155
-------
quantifiable levels on those days were included. Further, only
those days were chosen where oil and grease concentrations in raw
wastewater exceeded 10 mg/1 and where there was reduction in oil
and grease going through the treatment system. All plant
sampling days which met the above criteria are included below.
The conclusion is that when oil and grease is removed, toxic
organics are removed, also.
Percent Removal
Plant-Day Oil & Grease Orqanics
1054-3
13029-2
13029-3
38053-1
38053-2
Mean 96.9 84.2
For aluminum forming wastewaters, effective oil removal
technology (such as oil skimming or emulsion breaking) is capable
of removing approximately 97 percent of the total toxic organics
(TTO) from the raw waste. As shown in Table VII-29 (page 235),
the achievable TTO concentration is approximately 0.690 mg/1.
The influent and effluent concentrations presented for each
pollutant were taken from the aluminum forming category for
several plants with effective oil removal technologies in place.
In calculating the concentrations, if only one day's sampling
datum was available, that value was used; if two day's sampling
data were available, the higher of the values was used; and, if
three day's sampling data were available, the mean or the median
value was used, whichever was higher. The 0.690 mg/1 value is an
appropriate basis for effluent limitations, since the highest
values were used in the calculation.
The estimated level of oil and grease in raw wastewater at BAT
flow levels for the categories discussed above is:
Untreated
Source Oil Concentration
Aluminum Forming - 17,752 mg/1
(rolling with emulsions)
Coil Coating - 801.5 mg/1
(Steel subcategory)
(Canmaking Subcategory) -19,838 mg/1
156
-------
Advantages and Limitations. Skimming as a pretreatment is
fmnrovir Jh remov^ naturally floating waste material. It also
improves the pertormance of subsequent downstream treatments
Many pollutants, particularly dispersed or emulsified oil, "if i
JSr* 2?t .natu^ally but require additional treatments. There-
fore, skimming alone may not remove all the pollutants capable of
P
or
.I1 Because of its
Maintainability: The skimming mechanism requires periodic
lubrication, adjustment, and replacement of worn parts. pericxnc
Solid Waste Aspects: The collected layer of debris must be
disposed of by contractor removal, landfill, or incineration
elatively large quantities of water are present ?n thi
wastes' incineration is not always a viable disposal
method es' ncneraon is not always a viable disposal
Demonstration Status. Skimming is a common operation utilized
industlFial waste treatment system!. Oil removal
ufo . ova
w??hPnh^- ? skimming as a separate process or in conjunction
with chemical emulsion breaking, or dissolved air flotation
(discussed below) is in. place at 21 canmak ing plants.
MAJOR TECHNOLOGY EFFECTIVENESS
~™~ Pe5fo^mance ^ of individual treatment technologies was
he?e ?wo °dfff^!^0rmanJe °f °Peratin9 ^sterns is discussel
r^™: -i- ^ different systems are considered: L&S (hydroxide
?£JS lpl*5;tlon and sedimentation or lime and settle) and LS&F
set?leX1and fn?i??tati°E' sedimentation and filtration or lime,
settle, and filter . Subsequently, an analysis of effectiveness
*L ^ systems is made to develop one-daj maximum, and ten-day
and thirty-day average concentration levels to be used in
5SI? im "g pollujants- Evaluation of the L&S and the LS&F
systems is carried out on the assumption that chemical reduction
SL 5 o miS*f cyanidePrecipitation, and oil removal are installed
and operating properly where appropriate.
L&S Performance — Combined Metals Data Base
o x,bfse ^nown as the "combined metals data base" (CMDB) was
used to determine treatment effectiveness of lime and settle
treatment for certain pollutants. The CMDB was developed over
several years and has been used in a number of regulations!
^!ing ^r? Development of coil coating and other cltegoricli
effluent limitations and standards, chemical analysis data were
157
-------
collected of raw wastewater (treatment influent) and treated
wastewater (treatment effluent) from 55 plants (126 data days)
sampled by EPA (or its contractor) using EPA sampling and
chemical analysis protocols. These data are the initial data
base for determining the effectiveness of L&S technology in
treating nine pollutants. Each of the plants in the initial data
base belongs to at least one of the following industry
categories: aluminum forming, battery manufacturing, coil coating
(including canmaking), copper forming, electroplating and
porcelain enameling. All of the plants employ pH adjustment and
hydroxide precipitation using lime or caustic, followed by
Stokes1 law settling (tank, lagoon or clarifier) for .solids
removal. An analysis of this data was presented in the
development documents for the proposed regulations for coil
coating and porcelain enameling (January 1981). Prior to
analyzing the data, some values were deleted from the data base.
These deletions were made to ensure that the data reflect
properly operated treatment systems. The following criteria were
used in making these deletions:
Plants where malfunctioning processes or treatment
systems at the time of sampling were identified.
Data days where pH was less than 7.0 for extended
periods of time or TSS was greater than 50 mg/1 (these
are prima facie indications of poor operation).
In response to the coil coating and porcelain enameling
proposals, some commenters claimed that it was inappropriate to
use data from some categories for regulation of other categories.
In response to these comments, the Agency reanalyzed the data.
An analysis of variance was applied to the data for the 126 days
of sampling to test the hypothesis of homogeneous plant mean raw
and treated effluent levels across categories by pollutant. This
analysis is described in the report "A Statistical Analysis of
the Combined Metals Industries Effluent Data" which is in the
administrative record supporting this rulemaking. The main
conclusion drawn from the analysis of variance is that, with the
exception of electroplating, the categories included in the data
base are generally homogeneous with regard to mean pollutant
concentrations in both raw and treated effluent. That is, when
data from electroplating facilities are included in the analysis,
the hypothesis of homogeneity across categories is rejected.
When the electroplating data are removed from the analysis the
conclusion changes substantially and the hypothesis of
homogeneity across categories is not rejected. On the basis of
this analysis, the electroplating data were removed from the data
base used to determine limitations for the final coil coating and
porcelain enameling regulations and proposed regulations for
158
-------
copper forming, aluminum forming, battery manufacturing
nonferrous metals (Phase I) and canmaking. manufacturing,
The statistical analysis provides support for the technical
enaineerina -inrinmenf *-h=4- electroplatina uecnnicai
.^. Purpose of determining treatment effectiveness
additional data were deleted from the data base? ThesJ SllJtiSi
were made, almost exclusively, in cases where eKlueSt dSS
in ?wn J6re a|sociatef with low influent values. This was done
tn^ ?tePfv First, effluent values measured on the same day as
influent values that were less than or equal to 0.1 mg/1 weS
deleted. Second, the remaining data were screened for cases in
values at a plant were low a!?hougS sf?|htly
/J ^aJUe* These data were Deleted not as
data points but as plant clusters of data that were
X?w and thus not relevent to assessing treatment I
points were also deleted where malfunctions not
f ?tified We^ rfc°g"i2^. The data basf c °S ?he
- 248) sPlaved graphically in Figures VII-4 to 12 (Pages 240
a" Deletions', 148 data points from 19 plants remained.
S determine the concentration basis of
regultions. m fc CM°B used for the Proposed canmaking
*
few data
=
CMDB.
"Sed ?S .the basis for limitations in canmaking
>Kmodel treatment technology for canmaking, lime and
m *he same as for the categories represented in the
The selection of lime and settle was basld on the juSgmen?
the Process steps and wastewater characteristics in
for wMhiLeliniir to °ther. categories that process Stale
technology. 1S a" aPPr°Priate and demonstrated
aPPr?fcJ.in Analyzing the combined metals data was to
,statistical homogeneity of the categories with respect
mean Poll"tant concentrations in both raw and treated
tewateJ- l°r the Pr°P°sed canmaking regulation, the
w h ™naw wastewater data from canmaking were analyzed along
with the CMDB raw wastewater data. In the analysis, ' canmaking
additional category in the CMDB and the saml
159
-------
The results indicated substantial homogeneity ampng untreated
wastewater from canmaking and the combined metals categories.
Homogeneity is the absence of statistically discernable
differences among the categories while heterogeneity is the
opposite, i.e., the presence of statistically discernable
differences. The homogeneity found among the canmaking raw waste
data and the combined ,metals raw waste data supported the
hypothesis of similar raw waste characteristics and suggests that
lime and settle treatment will reduce the concentrations of toxic
metal pollutants in canmaking to levels comparable to those
achievable by lime .and settle treatment of wastewater from the
other categories.
The CMDB was reviewed following its use in a number of proposed
regulations (including canmaking). Comments pointed out a few
errors in the data and the Agency's review identified a few
transcription errors and some data points that were appropriate
for inclusion in the data that had not been used previously
because of errors in data record identification numbers.
Documents in the record of this rulemaking identify all the
changes, the reasons for, the changes, and the effect of these'
changes on the data base. Other comments on the CMDB asserted
that the data base was too small and that the statistical methods
used were overly complex. Responses to specific comments are
provided in a document included in the canmaking rulemaking. The
Agency believes that the data base is adequate to determine
effluent concentrations achievable with lime and settle
treatment. The statistical methods employed in the analysis are
well known and appropriate statistical references are provided in
the documents in the record that describe the analysis.
The revised data base was re-examined for homogeneity. The
earlier conclusions were unchanged. The categories show good
overall homogeneity with respect to concentrations of the nine
pollutants in both raw and treated wastewaters with the exception
of electroplating.
The same procedures used in developing proposed limitations from
the combined metals data base were then used on the revised data
base. That is, certain effluent data associated with low
influent values were deleted, and then the remaining data were
fit to a lognormal distribution to determine limitations values.
The deletion of data was done in two steps. First, effluent
values measured on the same day as influent values that were less
than or equal to 0.1 mg/1 were deleted. Second, the remaining
data were screened for cases in which all influent values at a
plant were low although slightly above the 0.1 mg/1 value. These
data were deleted not as individual data points but as plant
clusters of data that were consistently low and thus not relevant
to assessing treatment.
160
-------
c
sa*pled plants had apprSpr^te^e'and Sttl.°t™SKt 'llf o?
23
-------
lime were equally as effective. The Agency had fluoride data
frSm 3 plants that use lime and 8 plants that use caustic.
Statistical analysis of these data show the lime ^o.uP^hie^
significantly lower fluoride concentrations. In addition, the
data show the caustic group exceeded the concentration basis for
the fluoride limitation in over half the samples while the lime
group shows no exceedances of the limitation.
Aluminum was not one of the pollutants included in the CMDB. As
described in Section IX, limitations for aluminum that apply to
canmaking direct dischargers were developed from aluminum
effluent data collected by EPA at 3 aluminum forming plants and
one Suminum coil coating plant. The use of these aluminum datj
in canmaking was supported by comparison with aluminum data
collected by industry at canmaking plants with appropriate lime
Snd settle treatment. Comparison of the industry aluminum
effluent data <3 plants, 8 observations) with the EPA data (4
plantl? IT observations) showed no significant difference between
the two groups. Also, comparison Of influent aluminum data
collected by industry and EPA at canmaking plants and the
Influent aluminum datacorres^ohding to the effluent data used £o
determine the aluminum limitations showed no significant
difference among the two groups. The details of this comparison
are also described in the canmaking record.
One-day Effluent Values
The same procedures used to determine the concentration basis of
the limitations for lime and settle treatment from the CMDB at
proposal were used in the revised CMDB for the final limitations.
The basic assumption underlying the determination of treatment
effectiveness is that the data for a particular Pollutant are
lognormally distributed by plant. The lognormal has been found
to provide a satisfactory fit to plant effluent data in a number
of effluent guidelines categories and there was no evidence that
the logSal was not suitable in the case of the CMDB Thus we
assumed measurements of each pollutant from a particular plant,
denoted by X, were assumed followed a lognormal distribution with
log mean * and log variance **. The mean, variance and 99th
percentile of X are then:
mean of X * E(X) - exp (» + «* /2)
variance of X = V(X) = exp (2 * + «2) [exp( «2 )-l1
99th percentiie = X.9, - exp ( n * 2.33 *)
where exp is e, the base of the natural logarithm. The term
lognormal is used' because the logarithm of X has J. normal
distribution with mean * and variance «*. Using the basic
162
-------
Of.lo9no«|>ality the actual treatment effectiveness was
determined using a lognormal distribution that, in a sense
XK°£mate® the distribution of an average of the p?an?s inlhe
data base i.e. an "average plant" distribution. The notion of
an average plant" distribution is not a strict statistical
concept but is used here to determine limits that would represent
3"06 Capabilifcy of an avera9* of the plants in Ihl
This "average plant" distribution for a particular pollutant was
developed as follows: the log mean was determined bytakinS ?he
average of all the observations for the pollutant across plants
^ Pl
*s tne weighted average of the plant vananrvac
Thus the log mean represents the average of all thS data for the
pollutant and the log variance represents the average of the
pollutant 9 Variances or ^erage plant variability for the
The one day effluent values were determined as follows:
the jth observation on a particular pollutant at
i « 1, ..., I
j = 1, ...f Ji
I = total number of plants
Ji » number of observations at plant i.
Then
where
Then
yij - In Xij
In means the natural logarithm.
y « log mean over all plants
I
where
n = total number of observations
and
V(y) m pooled log variance
163
-------
where Si2 = log variance at plant i
yj = log mean at plant i.
Thus, v and V(y) are the log mean and log variance, respectively,
of the lognormal distribution used to determine the treatment
effectiveness. The estimated mean and 99th percent ile of this
distribution form the basis for the long term average and daily
maximum effluent limitations, respectively. The estimates are
mean - £(X) - exp(y) V n (0;5 V(y»
99th percentile = X.9, = exp [y + 2.33 W(yf ]
where * {.) is a Bessel function and exp is e, the base of the
natural logarithms (See Aitchison, J. and J.A.C. Brow.n, T£e
Loqnormal Distribution, Cambridge University Press, 1963). in
cases where zeros were present in the data, a generalized form of
the lognormal, known as the delta distribution was used (See
Aitchison and Brown, op. cit., Chapter 9).
For certain pollutants, this approach was modified slightly to
ensure that well operated lime and settle plants in all CMDB
categories would achieve the pollutant concentration values
calculated f rom ' the CMDB. For instance, after excluding the
electroplating data and other data that did not reflect pollutant
removal or proper treatment, the effluent copper data from the
copper forming plants were statistically signif icantly greater
than the copper data from the other plants. This indicated that
copper forming plants might have difficulty achieving an effluent
concentration value calculated from copper data from all CMDB
categories. Thus, copper effluent values shown in Jable VII-14
(paqe 224) are based only on the copper effluent data from the
copper forming plants. That is,, the log mean for copper is the
mean of the logs of all copper values from the copper forming
plants only and the log variance is the pooled log variance of
the copper* forming plant data only. In the case of cadmium
after excluding the electroplating data and data that did not
reflect removal or proper treatment, there were insufficient data
to estimate the log variance for cadmium. The variance used to
determine the values shown in Table VII-14 for cadmium was
estimated by pooling the within plant variances for all the other
metals? Thus, the cadmium variability is the average of the
164
-------
Average Effluent Values
was '
Ten-Sample Average
mean of X10 s E(XIO) = E(X)
variance of XIO . V(xld) = V(X)'A 10.
165
-------
Where E(X) and V(X) are the mean and variance of X, resPfc^vely,
defined above. We then assume that X10 follows a lognormal
distribution with log mean ,10 and log standard deviation ,*.
The mean and variance of X10 are then
10 + °-5 *2. o)
(2 v ,0
Now, M 10 and ^to can be derived in terms of M and *« as
. + tfz /2 - 0.5 In [l+(exp(
Therefore P,O and «f210 can be estimated using the above
SStiSXnips'and the estimates of , and *« obtained for the
underlying lognormal distribution. The 10 sample limitation
value was determined by the estimate of the approximate 99th
plrcentile of the distribution of the 10 sample average given by
2-33
where ^ 10 and ^ ,0 are tne estimates of MIO and
respectively.
Thirty Sample Average
Monthly average values based on the average of 30 daily
measurements we?e also calculated. These are included because
monthly limitations based on 30 samples have been used in the
past and for comparison with the 10 sample values. The average
values based on 30 measurements are determined on the basis of a
statistical result known as the Central Limit Theorem. This
Iheorem states that, under general and nonrestrictive
issumDtions the distribution of a sum of a number of random
vtliables? 'say n is approximated by the normal distribution.
The approximation improves as the number of variables, n,
IncrealS? The Theorem is quite general in -that no Particular
distributional form is assumed for the distribution of the
individual variables. In most applications (as in approximating
the distribution of 30-day averages) the Theorem is used to
approximate the distribution of the average of n observations of
a random variable. The result makes it possible to compute
tDpSximate probability statements about the average in a wide
rlSge teasel. For instance, it is possible to compute a value
below which a specified percentage (e.g. , 99 percent) of the
averages of n observations are likely to fall. Mo«ot te^books
state that 25 or 30 observations are sufficient for the
approximation to be valid. In applying the Theorem to the
distribution of the 30 day average effluent values, we
approximaifthe distribution of the average of 30 observations
166
-------
and use the
Thirty Sample Average Calculation
•
approximately normally distributed. The'mean aSd variLce'Sf
mean of X30 ^ E(X30) = E(X)
variance of X30 = V(X30) = v(X) 4 30.
The 30 sample average value
approximate 99th percentile
average given by
X3Q<.99) =
where A
E(X)
Application
167
-------
e
valutsT EencePte tn-ay average provides a reasonable basis
for a 'mon?Sly avJrage limitation and is typical of the sampling
frequency required by existing permits.
The
?equiKIttoabrnana!J^rLSd 'Z^T » "tSe "permit or ' th.
pretreatment authority.
CANMAKING DATA - To determine the applicability of the combined
Statistical'procedures used to assess homogeneity of the combined
SetalS data Sere performed. The results indicate substantial
categories and sugglst that lime and settle treatment would
Sducl concentrations of the CMDB pollutants in canmaking to
iSveS Sparable to those achievable by lime and settle in the
PMDB cateao?ies Additionally, the concentrations of aluminum,
?lSor?df9and phosphorus found in canmaking raw wastewaters are
*. iuv^i. *w»^. v*.«— tr *r . __ *r__ 4-u/^r-/^ r\n I liiranr*; TOlincl
comparable to or lower than values for
l dSiame>a^
mltals dSta base. The analysis of the canmaking wastewater data
and of the combined metals data base is detailed in the
administrative record of this rulemaking.
Additional Pollutants
T*m additional pollutant parameters were evaluated to determine
the perf<£m!nc! of lime and settle treatment systems in removing
them from industrial wastewater. Performance data for these
oarameters is not a part of the CMDB so other data available to
the Agenc? from Sther Categories has been used to determine the
168
-------
values were calculated by multiplying the mean performance from
Table VI1-15 (page 225) by the appropriate variability factor.
(The variability factor is the ratio of the value of concern to
the mean). The pooled variability factors are: one-day maximum -
4.100; ten-day average - 1.821; and 30-day average - 1.618 these
one-, ten- and thirty-day values are tabulated in Table VI1-21
(page 230).
In establishing which data were suitable for use in Table VII-14
two factors were heavily weighed; (1) the nature of the
wastewater; and (2) the range of pollutants or pollutant matrix
in the raw wastewater. These data have been selected from
processes that generate dissolved metals in the wastewater and
which are generally free from complexing agents. The pollutant
matrix was evaluated by comparing the concentrations of
pollutants found in the raw wastewaters with the range of
pollutants in the raw wastewaters of the combined metals data
set. These data are displayed in Tables VII-16 (page 225) and
VI1-17 (page 226) and indicate that there is sufficient
similarity in the raw wastes to logically assume transferability
of the treated pollutant concentrations to the combined metals
data base. Ganmaking wastewaters also were compared to the
wastewaters from plants in categories from which treatment
effectiveness values were calculated. The available data on
these added pollutants do not allow homogeneity analysis as was
performed on the combined metals data base. The data source for
each added pollutant is discussed separately.
Antimony (Sb) - The achievable performance for antimony is based
on data from a battery and secondary lead plant. Both EPA
sampling data and recent permit data (1978-1982) confirm the
achievability of 0.7 mg/1 in the battery manufacturing wastewater
matrix included in the combined data set.
Arsenic (As) - The achievable performance of 0.5 mg/1 for arsenic
is based on permit data from two nonferrpus metals manufacturing
plants. The untreated wastewater matrix shown in Table VII-17
(page 226) is comparable with the combined metals data base
matrix.
Beryllium (Be) - The treatability of beryllium is transferred
from the nonferrous metals manufacturing industry. The 0.3 mg/1
performance is achieved at a beryllium plant with the comparable
untreated wastewater matrix shown in Table VII-17.
Mercury (Hq) - The 0.06 mg/1 treatability of mercury is based on
data from four battery plants. The untreated wastewater matrix
at these plants was considered in the combined metals data base.
169
-------
Selenium (Se) - The 0.30 mg/1 treatability of selenium is based
onrecent permit data from one of the nonferrous metals
manufacturing plants also used for antimony performance. The
untreated wastewater matrix for this plant is shown in Table
VII-17.
Silver - The treatability of silver is based on a 0.1 mg/1
treatability estimate from the inorganic chemicals industry.
Additional data supporting a treatability as stringent or more
stringent than 0.1 mg/1 is also available from seven nonferrous
metals manufacturing plants. The untreated wastewater matrix for
these plants is comparable and summarized in Table VII-17.
Thallium (Tl) - The 0.50 mg/1 treatability for thallium is
transferred from the inorganic chemicals industry. Although no
untreated wastewater data are available to verify comparability
with the combined metals data set plants, no other sources of
data for thallium treatability could be identified.
Aluminum (Al) - The 2.24 mg/1 treatability of aluminum is based
on the mean performance of three aluminum forming plants and one
coil coating plant. These plants are from categories included in
the combined metals data set, assuring untreated wastewater
matrix comparability.
Cobalt (Co) - The 0.05 mg/1 treatability is based on nearly
complete removal of cobalt at a porcelain enameling plant with a
mean untreated wastewater cobalt concentration of 4.31 mg/1. In
this case, the analytical detection using aspiration techniques
for this pollutant is used as the basis of the treatability.
Porcelain enameling was considered in the combined metals data
base, assuring untreated wastewater matrix comparability.
Fluoride (F) - The 14.5 mg/1 treatability of fluoride is based on
the mean performance (216 samples) of an electronics
manufacturing plant. The untreated wastewater matrix for this
plant shown in Table VII-17 is comparable to the combined metals
data set. The fluoride level in the electronics wastewater (760
mg/1) is significantly greater than the fluoride level in raw
canmaking wastewater (16.7 mg/1 - see Table X-l) leading to the
conclusion that the canmaking wastewater should be no more
difficult to treat for fluoride removal than the electronics
wastewater. The fluoride level in the CMDB - electroplating data
ranges from 1.29 to 70.0 mg/1 while the fluoride level in the
canmaking wastewater was lower ranging from <1.0 to 16.5 mg/1 and
leading to the conclusion that the canmaking wastewater should be
no more difficult to treat to remove fluoride than electroplating
wastewater.
170
-------
ta"nc
is a
LS&F Performance
Tables VII-i8 and VII-19 (pages 227 and 228) show long term data
aS^r^f^^-^^^SSHiS!
soiias. Plant A uses a pressure filter, while Plant B
rapid sand filter. »•«**« riant B
W3S collect^ only occasionally at each
a nf . ™t Jf of ' from
P2ints and Discrepancies Tnmethod f
and
data . for .inc and cadmium
.
taken immediately before the smelter was closed It has
arranged similarity to Plants A and B for companion and £st
These data are presented to demonstrate the performance of
coagulation) generally produces better and more
of sacrlf iSST
171
-------
The LS&F performance data presented here are based on systems
that provide polishing filtration after effective L&S treatment.
We have previously shown that L&S treatment is equally applicable
to wastewaters from the five categories because of the
homogeneity of its raw and treated wastewaters, and other
factors. Because of the similarity of the wastewaters after L&S
treatment, the Agency believes these wastewaters are equally
amenable to treatment using polishing filters added to the L&S
treatment system. The Agency concludes that LS&F data based on
porcelain enameling and nonferrous smelting and refining is
directly applicable to the aluminum forming, copper forming,
battery manufacturing, coil coating, and metal molding and
casting categories, and the canmaking subcategory as well as it
is to porcelain enameling and nonferrous melting and refining.
Analysis of Treatment System Effectiveness
Data are presented in Table VII-14 showing the mean, one-day, 10-
day, and 30-day values for nine pollutants examined in the L&S
combined metals data base. The pooled variability factor for
seven metal pollutants (excluding cadmium because of the small
number, of data points) was determined and is used to estimate
one-day, 10-day and 30-day values. (The variability factor is
the ratio of the value of concern to the means the pooled
variability factors are: one-day maximum - 4.100; ten-day average
- 1.821; and 30-day average - 1.618.) For values not calculated
from the common data base as previously discussed, the mean value
for pollutants shown in Table VII-15 were multiplied by the
variability factors to derive the value to obtain the one-, ten-
and 30-day values. These are tabulated in Table VII-21 .
LS&F technology data are presented in Tables VII-18 and VI1-19.
These data represent two operating plants (A and B) in which the
technology has been installed and operated for some years. Plant
A data was received as a statistical summary and is presented
without change. Plant B data was received as raw laboratory
analysis data. Discussions with plant personnel indicated that
operating experiments and changes in materials and reagents and
occasional operating errors had occurred during the data
collection period. No specific information was available on
those variables. To sort out high values probably caused by
methodological factors from random statistical variability, or
data noise, the plant B data were analyzed. For each of four
pollutants (chromium, nickel, zinc, and iron), the mean and
standard deviation (sigma) were calculated for the entire data
set. A data day was removed from the complete data set when any
individual pollutant concentration for that day exceeded the sum
of the mean plus three sigma for that pollutant. Fifty-one data
days (from a total of about 1300) were eliminated by this method.
172
-------
Another approach was also used as a check on the above method of
eliminating certain high values. The minimum values of raw
wastewater concentrations from Plant B for the same four
pollutants were compared to the total set of values for the
corresponding pollutants. Any day on which the treated
wastewater pollutant concentration exceeded the minimum value
selected from raw wastewater concentrations for that pollutant
was discarded. Forty-five days of data were eliminated by that
procedure. Forty-three days of data in common were eliminated by
t1*?^1^6?111^5*^51"06 common engineering practice (mean plus
3 sigma) and logic (treated wastewater concentrations should be
less than raw wastewater concentrations) seem to coincide, the
data base with the 51 spurious data days eliminated is the basis
;ȣ m2 further analysis. Range, mean plus standard deviation
and mean plus two standard deviations are shown in Tables VII-18
and VII-19 for Cr, Cu, Ni, Zn and Fe.
The Plant B data was separated into 1979, 1978, and total data
base (six years) segments. With the statistical analysis from
P £• -K <-u°r 1978 and 1979 this in- effect created five data sets
in which there is some overlap between the individual years and
i 2 Si « ? 4-feJS4.£rom Plant B' By comparing these five parts it
is apparent that they are quite similar and all appear to-be from
the same family of numbers. The largest mean found among the
five data sets for each pollutant was selected as the long term
technology and is used as the LS&F mean in Table
Plant C data was used as a basis for cadmium removal performance
and as a check on the zinc values derived from Plants A and B.
The cadmium data is displayed in Table VII-20 (page 229) and is
incorporated into Table VI 1-21 for LS&F. The zinc data was
analyzed for compliance with the 1-day and 30-day values in Table
Vll-21; no zinc value of the 103 data points exceeded the 1-dav
hi2£irlai!e™ 1:°? mg/i* The 103 data Points were separated into
blocks of 30 points and averaged. Each of the 3 full 30-dav
?^??6S iTaS*.v,lel? Jhan the Table VII~21 value of 0.31 mg/1.
fSJi^i f/Jiyon^e Plant C raw wastewater pollutant concentrations
ronrin^J-~20> f*lt "^i . *ithin the ran9e of ™ wastewater
concentrations of the combined metals data base (Table VI 1-16)
further supporting the conclusion that Plant C wastewater data is
compatible with similar data from Plants A and B.
SoncentraSion values for regulatory use are displayed in Table
VII-21. Mean one-day, ten-day and 30-day values for L&S for nine
pollutants were taken from Table VII-14; the remaining L&S values
®®J°Pl usin9 the mean values in Table VII-15 and the mean
ity factors discussed above.
173
-------
LS&F mean values for Cd, Cr, Ni, Zn and Fe are derived from
plants A, B, and C as discussed above. One-, ten- and thirty-day
values are derived by applying the variability factor developed
from the pooled data base for the specific pollutant to the mean
for that pollutant. Other LS&F values are calculated using the
long term average or mean and the appropriate variability
factors. Mean values for LS&F for pollutants not already
discussed are derived by reducing the L&S mean by one-third. The
one-third reduction was established after examining the percent
reduction in concentrations going from L&S to LS&F data for Cd,
Cr, Ni, Zn, and Fe. The average reduction is 0.3338 or one
third.
Copper levels achieved at Plants A and B may be lower than
generally achievable because of the high iron content and low
copper content of the raw wastewaters. Therefore, the mean
concentration value achieved is not used; LS&F mean used is
derived from the L&S technology.
L&S cyanide mean levels shown in Table VI1-8 are ratioed to one-
day, ten-day and 30-day values using mean variability factors.
LS&F mean cyanide is calculated by applying the ratios of
removals L&S and LS&F as discussed previously for LS&F metals
limitations. The cyanide performance was arrived at by using the
average metal variability factors. The treatment method used
here is cyanide precipitation. Because cyanide precipitation is
limited by the same physical processes as the metal
precipitation, it is expected that the variabilities will be
similar. Therefore, the average of the metal variability factors
has been used as a basis for calculating the cyanide one-day,
ten-day and thirty-day average treatment effectiveness values.
The filter performance for removing TSS as shown in Table VII-9
(page 221) yields a mean effluent concentration of 2.61 mg/1 and
calculates to a 10-day average of 4.33, 30-day average of 3.36
mg/1; a one-day maximum of 8.88. These calculated values more
than amply support the classic thirty-day and one-day values of
10 mg/1 and 15 mg/1, respectively, which are used for LS&F.
Although iron concentrations were decreased in some LS&F
operations, some facilities using that treatment introduce iron
compounds to aid settling. Therefore, the one-day, ten-day and
30-day values for iron at LS&F were held at the L&S level so as
to not unduly penalize the operations which use the relatively
less objectionable iron compounds to enhance removals of toxic
metals.
The removal of additional fluoride by adding polishing filtration
is suspect because of the high solubility of calcium fluoride.
The one available data point appears to question the ability of
174
-------
filters to achieve high removals of additional fluoride Th^
fluoride levels demonstrated for L&S are used as the trJitmlSt
effectiveness for LS&F. treatment
MINOR TECHNOLOGIES
apDaon n technol?9ies were considered for possible
dilcussed Cere thlS subcate^y- These technologies are
8. Flotation
Flotation is the process of causing particles such as metal
hydroxides or oil to float to the surface of a tank where ?hev
can be concentrated and removed. This is accoBDlilhed bX
releasing gas bubbles which attach to the solid particles
increasing their buoyancy and causing them to float
ir flM«n in ?UCh less time *y flotation ssolve
*n* i£1?tatl°^ls,?f greatest interest in removing oil from water
and is less effective in removing heavier precipitates.
This process may be performed in several ways: foam, dispersed
^pi^^r^,^ assy's." i?°
enhance the performance of the flotation process.
^^ Tfhof
imooe thl "ff— ^ S?a11 Particle«. Chemicals mly be used
to improve the efficiency with any of the basic methods.
Froth Flotation - Froth flotation is based on differences in the
physiochemical properties in various particles. Wet tabilitv and
surface properties affect the particles' Ability to attach
themselves to gas bubbles in an aqueous medium. In Iroth
SS.-S
175
-------
Dispersed Air Flotation - In dispersed air flotation, gas bubbles
are generated by introducing the air by means of mechanical
agitation with impellers or by forcing air through porous media.
Dispersed air flotation is used mainly in the metallurgical
industry.
Dissolved Air Flotation - In dissolved air flotation, bubbles are
produced by releasing air from a supersaturated solution under
relatively high pressure. There are two types of contact between
the gas bubbles and particles. The first type is predominant in
the flotation of flocculated materials and involves the
entrapment of rising gas bubbles in the flocculated particles as
they increase in size. The bond between the bubble and particle
is one of physical capture only. The second type of contact is
one of adhesion. Adhesion results from the intermolecular
attraction exerted at the interface between the solid particle
and gaseous bubble.
Vacuum Flotation - This process consists of saturating the
wastewater with air either directly in an aeration tank, or by
permitting air to enter on the suction of a wastewater pump. A
partial vacuum is applied, which causes the dissolved air to come
out of solution as minute bubbles. The bubbles attach to solid
particles and rise to the surface to form a scum blanket, which
is normally removed by a skimming mechanism. Grit and other
heavy solids that settle to the bottom are generally raked to a
central sludge pump for removal. A typical vacuum flotation unit
consists of a covered cylindrical tank in which a partial vacuum
is maintained. The tank is equipped with scum and sludge removal
mechanisms. The floating material is continuously swept to the
tank periphery, automatically discharged into a scum trough, and
removed from the unit by a pump also under partial vacuum.
Auxiliary equipment includes an aeration tank for saturating the
wastewater with air, a tank with a short retention time for
removal of large bubbles, vacuum pumps, and sludge pumps.
Application and Performance. The primary variables for flotation
design are pressure, feed solids concentration, and retention
period. The suspended solids in the effluent decrease, and the
concentration of solids in the float increases with increasing
retention period. When the flotation process is used primarily
for clarification, a retention period of 20 to 30 minutes usually
is adequate for separation and concentration.
Advantages and Limitations. Some advantages of the flotation
process are the high levels of solids separation achieved in many
applications, the relatively low energy requirements,.and the
adaptability to mefet the treatment requirements of different
176
-------
waste types. Limitations of flotation are that it often requires
addition of chemicals to enhance process performance and that it
generates large quantities of solid waste.
Operational Factors. Reliability: Flotation systems normally
are very reliable with proper maintenance of the sludge collecto?
mechanism and the motors and pump_s used for aeration,
andTH maintenance is required on the pumps
and motors. The sludge collector mechanism is subject to
breaka9e and may require periodic
Solid Waste Aspects: Chemicals are commonly used to aid the
flotation process by creating a surface or a structure that can
easily adsorb or entrap air bubbles. Inorganic chemicals, such
as the aluminum ajjd ferric salts, and activated silica, can bind
the particulate matter together and create a structure that can
entrap air bubbles. Various organic chemicals can change the
nature of either the air-liquid interface or the solid-liquid
interface, or both. These compounds usually collect on the
ih«^fa?6 *? bllng about the Desired changes. The added
chemicals plus the particles in solution combine to form a large
disposed which must be further treated or properly
Demonstration Status. Flotation is a fully developed process and
is readily available for the treatment of a wide variety of
industrial waste streams. Dissolved air flotation (DAF)
equipment is installed at 23 canmaking plants. One plant uses
DAF primari-ly— for oil removal. Nineteen plants use DAF primarilv
solids remova>-and secondarily for oil removal. Four plants
UfJLr. o i-S017 Oil remo*^L and solids removal in conjunction with
other solids removal equipment.
9. Chemical Emulsion Breaking
Chemical treatment is often used to brdak stable oil-water (O-W)
5!!i?t0!i2V A? ?~? emul?ion consists of oil dispersed in water,
stablized by electrical charges and emulsifying agents. A stable
emulsion will not separate or break down without some form of
treatment .
Once an emulsion is broken, the difference in specific gravities
allows the oil to float to the surface of the water. Solids
usually form a layer between the oil and water, since some oil is
^moiUff X2 ^ei-SPlid?: The lonaer the retention time, the more
complete and distinct the separation between the oil, solids, and
water will be. Often other methods of gravity differential
separation, such as air flotation or rotational separation (e.g.,
177
-------
centrifugation), are used to enhance and speed separation. A
schematic flow diagram of one type of application is shown in
Figure VII-31 (page 267).
The major equipment required for chemical emulsion breaking
includes: reaction chambers with agitators, chemical storage
tanks, chemical feed systems, pump, and piping.
Emulsifiers may be used in the plant to aid in stabilizing or
forming emulsions. Emulsifiers are surface-active agents which
alter the characteristics of the oil and water interface. These
surfactants have rather long polar molecules. One end of the
molecule is particularly soluble in water (e,.g., carboxyl,
sulfate, hydroxyl, or sulfonate groups) and the other end is
readily soluble in oils (an organic group which varies greatly
with the different surfactant type). Thus, the surfactant
emulsifies or suspends the organic material (oil) in water.
Emulsifiers also lower the surface tension of the O-W emulsion as
a result of solvation and ionic complexing. These emulsions must
be destabilized in the treatment system.
Application and Performance. Emulsion breaking is applicable to
wastestreams containing emulsified oils or lubricants such as
rolling and drawing emulsions.
Treatment of spent O-W emulsions involves the use of chemicals to
break the emulsion followed by gravity differential separation.
Factors to be considered for breaking emulsions are type of
chemicals, dosage and sequence of addition, pH, mechanical shear
and agitation, heat, and retention time.
Chemicals, e.g., polymers, alum, ferric chloride, and organic
emulsion breakers, break emulsions by neutralizing repulsive
charges between particles, precipitating or salting out
emulsifying agents, or altering the interfacial film between the
oil and water so it is readily broken. Reactive cations, e.g.,
H(+l), AK+3), Fe(+3), and cationic polymers, are particularly
effective in breaking dilute O-W emulsions. Once the charges
have been neutralized or the interfacial film broken, the small
oil droplets and suspended solids will be adsorbed on the surface
of the floe that is formed, or break out and float to the top.
Various types of emulsion-breaking chemicals are used for the
various types of oils.
If more than one chemical is required, the sequence of addition
can make quite a difference in both breaking efficiency and
chemical dosages.
pH plays an important role in emulsion breaking, especially if
cationic inorganic chemicals, such as alum, are used as
178
-------
coagulants. A depressed pH in the range of 2 to 4 keeps the
aluminum ion in its most positive state where it can function
most effectively for charge neutralization. After some of the
oil is broken free and skimmed, raising the pH into the 6 to 8
range with lime or caustic will cause the aluminum to hydrolyze
and precipitate as aluminum hydroxide. This floe entraps or
adsorbs destablized oil droplets which can then be separated from
the water phase. Cationic polymers can break emulsions over a
wider pH range and thus avoid acid corrosion and the additional
sludge generated from neutralization; however, an inorganic
flocculant is usually required to supplement the polymer emulsion
breaker's adsorptive properties.
Mixing is important in breaking O-W emulsions. Proper chemical
feed and dispersion is required for effective results. Mixing
also causes collisions which help break the emulsion, and
subsequently helps to agglomerate droplets.
In all emulsions, the mix of two immiscible liquids has a
specific gravity very close to that of water. Heating lowers the
viscosity and increases the apparent specific gravity
differential between oil and water. Heating also increases the
frequency of droplet collisons, which helps to rupture the
interfacial film. Chemical emulsion breaking efficiencies are
shown in Table VII-30 (page 236).
Oil and grease and toxic organics removal performance data are
shown in Tables Vil-11 and VII-13 (pages 222 and 224). Data were
obtained from sampling at operating plants and a review of the
current literature. This type of treatment is proven to be
reliable and is considered the current state-of-the-art for
aluminum forming as well as canmaking emulsified oily
wastewaters.
Advantages and Limitions. Advantages gained from the use of
chemicals for breaking O-W emulsions are the high removal
efficiency potential and the possibility of reclaiming the oily
waste. Disadvantages are corrosion problems associated with
Acid-alum systems, skilled operator requirements for batch
treatment, chemical sludges produced, and poor cost-effectiveness
for low oil concentrations.
Operational Factors. Reliability: Chemical emulsion breaking is
a very reliable process. The main control parameters, pH and
temperature, are fairly easy to control.
Maintainability: Maintenance is required on pumps, motors, and
valves, as well as periodic cleaning of the treatment tank to
remove any accumulated'solids. Energy use is limited to mixers
and pumps.
179
-------
Solid Waste Aspects: The surface oil and oily sludge produced are
usually hauled away by a licensed contractor. If the recovered
oil has a sufficiently low percentage of water, it may be burned
for its fuel value or processed and reused.
Demonstration Status. Chemical emulsion breaking (CEB) is a
fully developed technology widely used in other industry
segments, such as metal forming, that use oil-water emulsions.
CEB is installed at 4 canmaking plants where it is used for oil
removal on the total waste stream; 16 other plants use CEB as
pretreatment for oil removal on the oily waste stream.
10. Carbon Adsorption
The use of activated carbon to remove dissolved organics from
water and wastewater is a well demonstrated technology. It is
one of the most efficient organic removal processes available.
This sorption process is reversible, allowing activated carbon to
be regenerated for reuse by the application of heat and steam or
solvent. Activated carbon has also proved to be an effective
adsorbent for many toxic metals, including mercury. Regeneration
of carbon which has adsorbed significant metals, however, may be
difficult.
The term activated carbon applies to any amorphous form of carbon
that has been specially treated to give high adsorption
capacities. Typical raw materials include coal, wood, coconut
shells, petroleum base residues and char from sewage sludge
pyrolysis. A carefully controlled process of. dehydration,
carbonization, and oxidation yields a product which is called
activated carbon. This material has a high capacity for
adsorption due primarily to the large surface area available for
adsorption, 500-1500 ma/g resulting from a large number of
internal pores. Pore sizes generally range from 10-100 angstroms
in radius.
Activated carbon removes contaminants from water by the process
of adsorption, or the attraction and accumulation of one
substance on the surface of another. Activated carbon
preferentially adsorbs organic compounds and, because of this
selectivity, is particularly effective in removing organic
compounds from aqueous solution.
Carbon adsorption requires pretreatment to remove excess
suspended solids, oils, and greases. Suspended solids in the
influent should be less than 50 mg/1 to minimize backwash
requirements; a downflow carbon bed can handle much higher levels
(up to 2000 mg/1), but requires frequent backwashing.
Backwashing more than two or three times a day is not desirable;
at 50 mg/1 suspended solids one backwash will suffice. Oil and
180
-------
grease should be less than about 10 mg/1. A high level of
dissolved inorganic material in the influent may cause problems
with thermal carbon reactivation (i.e., scaling and loss of
activity) unless appropriate preventive steps are taken. Such
steps might include pH control, softening, or the use of an acid
wash on the carbon prior to reactivation.
Activated carbon is available in both powdered and granular form
An adsorption column packed with granular activated carbon is
shown in Figure VII-17 (page 253). Powdered carbon is less
expensive per unit weight and may have slightly higher adsorption
capacity, but it is more difficult to handle and to regenerate.
Application and Performance. Carbon adsorption is used to remove
mercury from wastewaters. The removal rate is influenced by the
mercury level in the influent to the adsorption unit. Removal
i'?^ei? found at three manufacturing facilities are shown in Table
vll-24 (page 233). In the aggregate these data indicate that
very low effluent levels could be attained from any raw waste by
use of multiple adsorption stages. This is characteristic of
adsorption processes.
Isotherm tests have indicated that activated carbon is very
effective in adsorbing 65 percent of the organic priority
pollutants and is reasonably effective for another 22 percent:
Specifically, for the organics of particular interest, activated
carbon^ was very effective in removing all phthalates. It was
resonably effective on 1,1,1-trichloroethane,
bis(2-chloroethyl)ether, and toluene.
Table VII-22 (page 231) summarizes the treatment effectiveness
for most of the organic priority pollutants by activated carbon
as compiled by EPA. Table VII-23 (page 232) summarizes classes
of organic compounds together with examples of organics that are
readily adsorbed on carbon. Table VI1-24 lists the effectiveness
of activated carbon for the removal of mercury.
Advantages and Limitations. The major -benefits of carbon
treatment include applicability to a wide variety of organics,
and high removal efficiency. Inorganics -such as cyanide,
chromium, and mercury are also removed effectively. Variations
in concentration and flow rate are well tolerated. The system is
compact, and recovery of adsorbed materials is sometimes
practical. However, destruction of adsorbed compounds often
occurs during thermal regeneration. If carbon cannot be
thermally desorbed, it must be disposed of along with any
adsorbed pollutants. The capital and operating costs of thermal
regeneration are relatively high. Cost surveys show that thermal
regeneration is generally economical when carbon usage exceeds
about 1,000 Ib/day. Carbon cannot remove low molecular weight or
181
-------
highly soluble organics. It also has a low tolerance for
suspended solids, which must be removed to at least 50 mg/1 in
the influent water.
Operational Factors. Reliability: This system should be very
reliable with upstream protection and proper operation and
maintenance procedures.
Maintainability: This system requires periodic regeneration or
replacement of spent carbon and is dependent upon raw waste load
and process efficiency.
Solid Waste Aspects: Solid waste from this process is
contaminated activated carbon that requires disposal. Carbon
undergoes regeneration, which reduces the solid waste problem by
reducing the frequency of carbon replacement.
Demonstration Status. Carbon adsorption systems have been
demonstrated to be practical and economical in reducing COD, BOD
and related parameters in secondary municipal and industrial
wastewaters; in removing toxic or refractory organics from
isolated industrial wastewaters; in removing and recovering
certain organics from wastewaters; and in the removing and some
times recovering, of selected inorganic chemicals from aqueous
wastes. Carbon adsorption is a viable and economic process for
organic waste streams containing up to 1 to 5 percent of
refractory or toxic organics. Its applicability for removal of
inorganics such as metals has also been demonstrated.
11. Centrifuqation
Centrifugation is the application of centrifugal force to
separate solids and liquids in a liquid-solid mixture or to
effect concentration of the solids. The application of
centrifugal force is effective because of the density
differential normally found between the insoluble solids and the
liquid in which they are contained. As a waste treatment
procedure, centrifugation is applied to dewatering of sludges.
One type of centrifuge is shown in Figure VI1-18 (page 254).
There are three common types of centrifuges: the disc, basket,
and conveyor type. All three operate by removing solids under
the influence of centrifugal force. The fundamental difference
between the three types is the method by which solids are
collected in and discharged from the bowl.
In the disc centrifuge, the sludge feed is distributed between
narrow channels that are present as spaces between stacked
conical discs. Suspended particles are collected and discharged
182
-------
continuously through small orifices in the bowl wall The
clarified effluent is discharged through an overflow weir.'
?h£d SK ?f centrff"9e which is useful in dewatering sludges
the basket centrifuge. In this type of centrifuge sludae
rS?f JJ Xh^?HCC2 ^ the bottom of ^he basket^ and' soUdl
collect at the bowl wall while clarified effluent overflows the
The third type of centrifuge commonly used in sludge dewaterina
£?«. ^f Conv€T°£. ty£e' Slud9e ^ fed through a stationary feed
pipe into a rotating bowl in which the solids are settled out
against the bowl wall by centrifugal force. From the bowl wall
they are moved by a screw to the end of the machine at which
dischar9ed- ^e liquid effESS is dtschargeS
°f the
Application And Performance. Virtually all industrial
~" can use cen?rifuga?n o
?hJ f J^°^m?nqe^f SlildH? dewatering by centrifugation depends on
the feed rate, the rotational velocity of the drum and the
onedKinnmP°?hti0n ?^ concentration. Assuming proper'des?gn an!
2Q-I? percent S° content of the sludge can be increased to
mn Limitations. Sludge dewatering centrifuges have
minimal space requirements and show a high degree of effluent
clarification. The operation is simple, clean, and relativllv
a cen?rtfugl
•r - - rsic%f°is tSi^tiara-t s
' ^'008 so«^Proofing becaue ? thS
— — —-•*- f*- v w A.VA^\JI «jj.nv,^ J-dLyJc?
difficulty encountered in the
'rf i»£- i £?eu ^as' been the disP°sal of the concentrate
relatively high in suspended, nonsettling solids.
183
-------
Operational Factors. Reliability: Centrif ugation :LS highly
reliable with proper control of factors such as sludge feed,
consistency, and temperature. Pretreatment such as grit removal
and coagulant addition may be necessary, depending on the
composition of the sludge and on the type of centrifuge employed.
Maintainability: Maintenance consists of periodic lubrication,
cleaning, and inspection. The frequency and degree of inspection
required varies depending on th<2 type of sludge solids being
dewatered and the maintenance service conditions. If the sludge
is abrasive, it is recommended that the first inspection of the
rotating assembly be made after approximately 1,000 hours of
operation. If the sludge is not abrasive or corrosive, then the
initial inspection might be delayed. Centrifuges not equipped
with a continuous sludge discharge system require periodic
shutdowns for manual sludge cake removal.
Solid Waste Aspects: Sludge dewatered in the centrifugation
process may be disposed of by landfill. The clarified effluent
(centrate), if high in dissolved or suspended solids, may require
further treatment prior to discharge.
Demonstration Status. Centrifugation is currently used in a
great many commercial applications to dewater sludge. Work is
underway to improve the efficiency, increase the capacity, and
lower the costs associated with centrifugation.
12. Coalescing
The basic principle of coalescence involves the preferential
wetting of a coalescing medium by oil droplets which accumulate
on the medium and then rise to the surface of the solution as
they combine to form larger particles. The most important
requirements for coalescing media are wettability for oil and
large surface area. Monofilament line is sometimes used as a
coalescing medium.
Coalescing stages may be integrated with a wide variety of
gravity oil separation devices, and some systems may incorporate
several coalescing stages. In general a preliminary oil skimming
step is desirable to avoid overloading the coalescer.
One commercially marketed system for oily waste treatment
combines coalescing with inclined plate separation and
filtration. In this system, the oily wastes flow into an
inclined plate settler. This unit consists of a stack of
inclined baffle plates in a cylindrical container with an oil
collection chamber at the top. The oil droplets rise and impinge
upon the undersides of the plates. They then migrate upward to a
184
-------
guide rib which directs the oil to the oil collection chamber
from which oil is discharged for reuse or disposal.
The oily water continues on through another cylinder containing
replaceable filter cartridges, which remove suspended particles
from the waste. From there the wastewater enters a final
cylinder in which the coalescing material is housed. As the oily
water passes through the many small, irregular, continuous
passages in the coalescing material, the oil droplets coalesce
and rise to an oil collection chamber.
Application and Performance. Coalescing is used to treat oily
wastes which do not separate readily in simple gravity systems.
The three stage system described above has achieved effluent
concentrations of 10-15 mg/1 oil and grease from raw waste
concentrations of 1000 mg/1 or more.
Advantages and Limitations. Coalescing allows removal of oil
droplets too finely dispersed for conventional gravity
separation-skimming technology. It also can significantly reduce
the residence times (and therefore separator volumes) required to
achieve separation of oil from some wastes. Because of its
simplicity, coalescing provides generally high reliability and
low capital and operating costs. Coalescing is not generally
effective in removing soluble or chemically stabilized emulsified
oils. To avoid plugging, coalescers must be protected by
pretreatment from very high concentrations of free oil and grease
and suspended solids. Frequent replacement of prefilters may be
necessary when raw waste oil concentrations are high.
Operational Factors. Reliability: Coalescing is inherently
highly reliable since there are no moving parts, and the
coalescing substrate (monofilament, etc.) is inert in the
process and therefore not swbject to frequent regeneration or
replacement requirements. Large loads or inadequate
pretreatment, however, may result in plugging or bypass of
coalescing stages.
Maintainability: Maintenance requirements are generally limited
to replacement of the coalescing medium on an infrequent basis.
Solid Waste Aspects: No appreciable solid waste is generated by
this process.
Demonstration Status. Coalescing has been fully demonstrated in
industries generating oily wastewater.
185
-------
13. Cyanide Oxidation by Chlorine
Cyanide oxidation using chlorine is widely used in industrial
waste treatment to oxidize cyanide. Chlorine can be utilized in
either the elemental or hypochlorite forms. This classic
procedure can be illustrated by the following two step chemical
reaction:
1. C12 + NaCN + 2NaOH —> NaCNO + 2NaCl •*• H20
2. 3C12 + 6NaOH +.2NaCNO —> 2NaHCO3 + N2 + 6NaCl + 2H20
The reaction presented as equation (2) for the oxidation of
cyanate is the final step in the oxidation of cyanide. A
complete system for the alkaline chlorination of cyanide is shown
in Figure VII-19 (page 255).
The alkaline chlorination process oxidizes cyanides to carbon
dioxide and nitrogen. The equipment often consists of an
equalization tank followed by two reaction tanks, although the
reaction can be carried out in a single tank. Each tank has an
electronic recorder-controller to maintain required conditions
with respect to pH and oxidation reduction potential (ORP). In
the first reaction tank, conditions are adjusted to oxidize
cyanides to cyanates. To effect the reaction, chlorine is
metered to the reaction tank as required to maintain the ORP in
the range of 350 to 400 millivolts, and 50 percent aqueous
caustic soda is added to maintain a pH range of 9.5 to 10. In
the second reaction tank, conditions are maintained to oxidize
cyanate to carbon dioxide and nitrogen. The desirable ORP and pH
for this reaction are 600 millivolts and a pH of 8.0. Each of
the reaction tanks is equipped with a propeller agitator designed
to provide approximately one turnover per minute. Treatment by
the batch process is accomplished by using two tanks, one for
collection of water over a specified time period, and one tank
for the treatment of an accumulated batch. If dumps of
concentrated wastes are frequent, another tank may be required to
equalize the flow to the treatment tank. When the holding tank
is full, the liquid is transferred to the reaction tank for
treatment. After treatment, the supernatant is discharged and
the sludges are collected for removal and ultimate disposal.
Application and Performance. The oxidation of cyanide waste by
chlorine is a classic process and is found in most industrial
plants using cyanide. This process is capable of achieving
effluent levels that are nondetectable. The process is
potentially applicable to canmaking facilities where cyanide is a
component in conversion coating formulations.
186
-------
Advantages and Limitations. Some advantages of chlorine
oxidation for handling process effluents are operation at ambient
temperature, suitability for automatic control, and low cost
Disadvantages include the need for careful pH control, possible
chemical interference in the treatment of mixed wastes, and the
potential hazard of storing and handling chlorine gas.
Operational Factors. Reliability: Chlorine oxidation is highly
reliable with proper monitoring and control, and proper
pretreatment to control interfering substances.
Maintainability: Maintenance consists of periodic removal of
sludge and recalibration of instruments.
Solid Waste Aspects: There is no solid waste problem associated
with chlorine oxidation.
Demonstration Status. The oxidation of cyanide wastes by
chlorine is a widely used process in plants using cyanide in
cleaning and metal processing baths.
14. Cyanide Oxidation by_ Ozone
Ozone is a highly reactive oxidizing agent which is approximately
ten times more soluble than oxygen on a weight basis in water.
Ozone may be produced by several methods, but the silent
electrical discharge method is predominant in the field. The
silent electrical discharge process produces ozone by passing
oxygen, or air between electrodes separated by an insulating
material. A complete ozonation system is represented in Figure
VII-20 (page 256).
Application and Performance. Ozonation has been applied
commercially to oxidize cyanides, phenolic chemicals, and organo-
metal complexes. Its applicability to photographic wastewaters
has been studied in the laboratory with good results. Ozone is
used in industrial waste treatment primarily to oxidize cyanide
to cyanate and to oxidize phenols and dyes to a variety of
colorless nontoxic products.
Oxidation of cyanide to cyanate is illustrated below:
CN- + 03 •— > CNO- + Og
Continued exposure to ozone will convert the cyanate formed to
carbon dioxide and ammonia; however, this is not economically
practical.
Ozone oxidation of cyanide to cyanate requires 1.8 to 2.0 pounds
ozone per pound of CN-; complete oxidation requires 4.6 to 5.0
187
-------
pounds ozone per pound of CN-. Zinc, copper, and nickel cyanides
are easily destroyed to a nondetectable level, but cobalt and
iron cyanides are more resistant to ozone treatment.
Advantages and Limitations. Some advantages of ozone oxidation
for handling process effluents are its suitability to automatic
control and on-site generation and the fact that reaction
products are not chlorinated organics and no dissolved solids are
added in the treatment step. Ozone in the presence of activated
carbon, ultraviolet, and other promoters shows promise of
reducing reaction time and improving ozone utilization, but the
process at present is limited by high capital expense, possible
chemical interference in the treatment of mixed wastes, and an
energy requirement of 25 kwh/kg of ozone generated. Cyanide is
not economically oxidized beyond the cyanate form.
Operational Factors. Reliability: Ozone oxidation is highly
reliable with proper monitoring and control, and proper
pretreatment to control interfering substances.
Maintainability: Maintenance consists of periodic removal of
sludge, and periodic renewal of filters and desiccators required
for the input of clean dry air; filter life is a function of
input concentrations of detrimental constituents.
Solid Waste Aspects: Pretreatment to eliminate substances which
will interfere with the process may be necessary. Dewatering of
sludge generated in the ozone oxidation process or in an "in
line" process may be desirable prior to disposal.
15. Cyanide Oxidation by. Ozone and UV Radiation
One of the modifications of the ozonation process is the
simultaneous application of ultraviolet light and ozone for the
treatment of wastewater, including treatment of halogenated
organics. The combined action of these two forms produces
reactions by photolysis, photosensitization, hydroxylation,
oxygenation and oxidation. The process is unique because several
reactions and reaction species are active simultaneously.
Ozonation is facilitated by ultraviolet absorption because both
the ozone and the reactant molecules are raised to a higher
energy state so that they react more rapidly. In addition, free
radicals for use in the reaction are readily hydrolyzed by the
water present. The energy and reaction intermediates created by
the introduction of both ultraviolet and ozone greatly reduce the
amount of ozone required compared with a system using ozone
alone. Figure VI1-21 (page 257) shows a three-stage UV-ozone
system. A system to treat mixed cyanides requires pretreatment
188
-------
that involves chemical coagulation, sedimentation, clarification
equalization, and pH adjustment. meat ion,
Application and Performance. The ozone-UV radiation process was
developed primarily for cyanide treatment in the elecrtroplatina
UnniifS ?r Ph°t°-Pr°?essing areas. It has been successfully
applied to mixed cyanides and organics from organic chemicals
manufacturing processes. The process is particularly useful for
treatment of complexed cyanides such as ferricyanide, copper
cyanide and nickel cyanide, which are resistant to ozone aloneT
Ozone combined with UV radiation is a relatively new technology
Four units are currently in operation and all four treat cyanide
oear ing waste.
Ozone-UV treatment could be used in canmaking plants to destroy
cyanide present in waste streams from some conversion coating
op© r 3. u XODS •
16. Cyanide Oxidation by_ Hydrogen Peroxide
Hydrogen peroxide oxidation removes both cyanide and metals in
cyanide containing wastewaters. In this process, cyanide bearing
waters are heated to 49 - 54<>c (120 - 130oF) and the pH
a^?edh*? ^A5 I "•?• Formalin (37 percent forma Idehye)
added while the tank is vigorously agitated. After 2-5 minutes,
a proprietary peroxygen compound (41 percent hydrogen peroxide
with a catalyst and additives) is added. After an hour of
* reaction is complete. The cyanide is converted to
and the metals are precipitated as oxides or hydroxides.
" removed fron> solution by either settling or
The main equipment required for this process is two holding tanks
?h^?P?f»tW u€aters, -and air sPar9ers or mechanical stirr-ers.
These tanks may be used in a batch or continuous fashion, with
???i Jj * i??. USfd ufor treatment while the other is being
filled. A settling tank or a filter is needed to concentrate the
precipitate.
Application and Performance. The hydrogen peroxide oxidation
process is applicable to cyanide bearing wastewaters, especially
those containing metal-cyanide complexes. In terms of waste
reduction performance, this process can reduce total cyanide to
less than 0.1 mg/1 and the zinc or cadmium to less than 1.0 mg/1.
Advantages and Limitations. Chemical costs are similar to those
for alkaline chlorination using chlorine and lower than those for
roSSS?1^ Wlfch .hypochlorite. All free cyanide reacts and is
completely oxidized to the less toxic cyanate state. in
189
-------
addition, the metals precipitate and settle quickly, and they may
be recoverable in many instances. However, the process requires
energy expenditures to heat the wastewater prior to treatment.
Demonstration Status. This treatment process was introduced in
1971 and is used in several facilities.
17. Evaporation
Evaporation is a concentration process. Water is evaporated from
a solution, increasing the concentration of solute in the
remaining solution. If the resulting water vapor is condensed
back to liquid water, the evaporation-condensation process is
called distillation. However, to be consistent with industry
terminology, evaporation is used in this report to describe both
processes. Both atmospheric and vacuum evaporation are commonly
used in industry today. Specific evaporation techniques are
shown in Figure VII-22 (page 258) and discussed below.
Atmospheric evaporation could be accomplished simply by boiling
the liquid. However, to aid evaporation, heated liquid is
sprayed on an evaporation surface, and air is blown over the
surface and subsequently released to the atmosphere. Thus,
evaporation occurs by humidification of the air stream, similar
to a drying process. Equipment for carrying out atmospheric
evaporation is quite similar for most applications. The major
element is generally a packed column with an accumulator bottom.
Accumulated wastewater is pumped from the base of the column,
through a heat exchanger, and back into the top of the column,
where it is sprayed into the packing. At the same time, air
drawn upward through the packing by a fan is heated as it
contacts the hot liquid. The liquid partially vaporizes and
humidifies the air stream. The fan then blows the hot, humid air
to the outside atmosphere. A scrubber is often unnecessary
because the packed column itself acts as a scrubber.
Another form of atmospheric evaporator also works on the air
humidification principle, but the evaporated water is recovered
for reuse by condensation. These air humidification techniques
operate well below the boiling point of water and can utilize
waste process heat to supply the energy required.
In vacuum evaporation, the evaporation pressure is lowered to
cause the liquid to boil at reduced temperature. All of the
water vapor is condensed and, to maintain the vacuum condition,
noncondensible gases (air in particular) are removed by a vacuum
pump. Vacuum evaporation may be either single or double effect.
In double effect evaporation, two evaporators are used, and the
water vapor from the first evaporator (which may be heated by
steam) is used to supply heat to the second evaporator. As it
190
-------
supplies heat, the water vapor from the first evaporator
condenses. Approximately equal quantities of wastewater are
evaporated in each unit; thus, the double effect system
evaporates twice the amount of water that a single effect system
does, at nearly the same cost in energy but with added capital
cost and complexity. The double effect technique is
thermodynamically possible because the second evaporator is
maintained at lower pressure (higher vacuum) and, therefore
lower evaporation temperature. Another means of increasing
energy efficiency is vapor recompression (thermal or mechanical);
which enables heat to be transferred from the condensing water
vapor to the evaporating wastewater. Vacuum evaporation
equipment may be classified as submerged tube or climbing film
In the most commonly used submerged tube evaporator, the heating
and condensing coil are contained in a single vessel to reduce
capital cost. The vacuum in the vessel is maintained by an
eductor-type pump, which creates the required vacuum by the flow
of the condenser cooling water through a venturi. Waste water
accumulates in the bottom of the vessel, and it is evaporated by
means of submerged steam coils. The resulting water vapor
condenses as it contacts the condensing coils in the top of the
vessel. The condensate then drips off the condensing coils into
a collection trough that carries it out of the vessel.
Concentrate is removed from the bottom of the vessel.
The major elements of the climbing film evaporator are the
evaporator, separator, condenser, and vacuum pump. Wastewater is
drawn into the system by the vacuum so that a constant liquid
level is maintained in the separator. Liquid enters the steam-
jacketed evaporator tubes, and part of it evaporates so that a
mixture of vapor and liquid enters the separator. The design of
the separator is such that the liquid is continuously circulated
from the ^separator to the evaporator. The vapor entering the
separator flows out through a mesh entrainment separator to the
condenser, where it is condensed as it flows down through the
condenser tubes. The condensate, along with any entrained air,
is pumped out of the bottom of the condenser by a liquid rina
vacuum pump. The liquid seal provided by the condensate keeps
the vacuum in the system from being broken.
Application and Performance. Both atmospheric and vacuum
evaporation are used in many industrial plants, mainly for the
concentration and recovery of process solutions. Many of these
evaporators also recover water for rinsing. Evaporation has also
been applied to recovery of phosphate metal cleaning solutions.
In theory, evaporation should yield a concentrate and a deionized
condensate. Actually, carry-over has resulted in condensate
191
-------
metal concentrations as high as 10 mg/1, although the usual level
is less than 3 mg/1, pure enough for most final rinses. The
condensate may also contain organic brighteners and antifoaming
agents. These can be removed with an activated carbon bed, if
necessary. Samples from one plant showed 1,900 rog/1 zinc in the
feed, 4,570 mg/1 in the concentrate, and 0.4 mg/1 in the
condensate. Another plant had 416 mg/1 copper in the feed and
21f800 mg/1 in the concentrate. Chromium analysis for that plant
indicated 5,060 mg/1 in the feed and 27,500 mg/1 in the
concentrate. Evaporators are available in a range of capacities,
typically from 15 to 75 gph, and may be, used: in parallel
arrangements for processing of higher flow rates.
Advantages and Limitations. Advantages of the evaporation
process are that it permits recovery of a wide variety of process
chemicals, and it is often applicable to concentration or removal
of compounds which cannot be accomplished by any other means.
The major disadvantage is that the evaporation process consumes
relatively large amounts of energy for the evaporation of water.
However, the recovery of waste heat from many industrial
processes (e.g., diesel generators, incinerators, boilers and
furnaces) should be considered as a source of this heat for a
totally integrated evaporation system. Also, in some cases solar
heating could be inexpensively and effectively applied to
evaporation units. For some applications, pretreatment may be
required to remove solids or bacteria which tend to cause fouling
in the condenser or evaporator. .The buildup of scale on the
evaporator surfaces reduces the heat transfer efficiency and may
present a maintenance problem, or imeeease operating cost.
However, it has been demonstrated that fouling of the heat
transfer surfaces can be avoided or minimized for certain
dissolved solids by maintaining a seed slurry which provides
preferential sites for precipitate deposition. In addition, low
temperature differences in the evaporator will eliminate nucleate
boiling and supersaturation effects. Steam distillable
impurities in the process stream are carried over with the
product water and must be handled by pre or post treatment.
Operational Factors. Reliability: Proper maintenance will
ensure a high degree of reliability for the system. Without such
attention, rapid—fouling or deterioration of vacuum seals may
occur, especially when handling corrosive liquids.
Maintainability: Operating parameters can be automatically
controlled. Pretreatment may be required, as well as periodic
cleaning of the system. Regular replacement of seals, especially
ia a corrosive environment, may be necessary.
Solid Waste Aspects: With only a few exceptions, the process
does not generate appreciable quantities of solid waste.
192
-------
Demonstration Status. Evaporation is a fully develooed
commercially available wastewater treatment system. ll is used
extensively to recover plating chemicals in the electroplating
i±X3- ^F8 f11^ SCale Unit has been used i" connection wi?h
?«S?S? tin9 of aluminum. Proven performance in silver recovery
i"^x^tesh ^at evaporation could be a useful treatment operation
for the photographic industry, as well as for metal finishing?
18. Gravity Sludge Thickening
- dilute sludge is fed from a
? /"* S? clarifier to a thickening tank where
rn- V ?ludge 9fntly to density it and to push it to a
central collection well. The supernatant is returned to the
KJTE* ?CJhU;9 £a"k' The thickened sludge that collets on the
bottom of the tank is pumped to dewatering equipment or hauled
gravity tifcklner?1'24 (pafl* ^ S^S ^ construct ionof f
Application and Performance. Thickeners are generally used in
facilities where the sludge is to be further dewatered bv a
compact mechanical device such as a vacuum filter or centrifuge
?2SSJi;9r.?J iSOlJdS C°^ent in the thi^ener subS?antial!y
reduces capital and operating cost of the subsequent dewatering
device and also reduces cost for hauling? The proceX il
potentially applicable to almost any industrial plant. process 1S
Organic sludges from sedimentation units of one to two percent
solids concentration can usually be gravity thickened to six to
ten percent; chemical sludges can be thickened to four tl six
percent.
Advantages and- Limitations. The principal advantage of a gravity
diw^-n Ckn?ing Process is that it facilitates further sludge
•**•
°£ ^f slud^e thickening process are its sensitivity
rate through the thickener and the sludge removal
to ^isturb°the
Operational Factors . Reliability: Reliability is high with
acors . eaiity: Reliability is high with
KPhL • f €Sign and 0Peration. A gravity thickener is designed on
the basis of square feet per pound of solids per day, in which
the required surface area is related to the solids entering and
lnafirLth? Unit', Thfckener area requirements are also expressed
day ?lEs/sq 9/ 9ramS °f 8plld8' P6r Square meter
193
-------
Maintainability: Twice a year, a thickener must be shut down for
lubrication of the drive mechanisms. Occasionally, water must be
pumped back through the system in order to clear sludge pipes.
Solid Waste Aspects: Thickened sludge from a gravity thickening
process will usually require further dewatering prior to
disposal, incineration, or drying. The clear effluent may be
recirculated in part, or it may be subjected to further treatment
prior to discharge.
Demonstration Status. Gravity sludge thickeners are used
throughoutIndustry to reduce water content to a level where the
sludge may be efficiently handled. Further dewatering is usually
practiced to minimize costs of hauling the sludge to approved
landfill areas. Sludge thickening is used in seven coil coating
plants.
19. Insoluble Starch Xanthate
Insoluble starch xanthate is essentially an ion exchange medium
used to remove dissolved heavy metals from wastewater. The water
may then either be reused (recovery application) or discharged
(end-of-pipe application). In a commercial electroplating oper-
ation, starch xanthate is coated on a filter medium. Rinse water
containing dragged out heavy metals is circulated through the
filters and then reused for rinsing. The starch-heavy metal
complex is disposed of and replaced periodically. Laboratory
tests indicate that recovery of metals from the complex is
feasible, with regeneration of the starch xanthate. Besides
electroplating, starch xanthate is potentially applicable to coil
coating, porcelain enameling, copper fabrication, and any other
industrial plants where dilute metal wastewater streams are
generated. Its present use is' limited to one electroplating
plant.
20. Ion Exchange .
Ion exchange is a process in which ions, held by electrostatic
forces to charged functional groups on the surface of the ion
exchange resin, are exchanged for ions of similar charge from the
solution in which the resin is immersed. This is classified as a
sorption process because the exchange occurs on the surface of
the resin, and the exchanging ion must undergo a phase transfer
from solution phase to solid phase. Thus, ionic contaminants in
a waste stream can be exchanged for the harmless ions of the
resin.
Although the precise technique may vary slightly according to the
application involved, a generalized process description follows.
The wastewater stream being treated passes through a filter to
194
-------
remove any solids, then flows through a cation exchanger which
contains the ion exchange resin. Here, metallic impurities such
as copper, iron, and trivalent chromium are retained. The stream
then passes through the anion exchanger and its associated resin
Hexavalent chromium, for example, is retained in this stage if
«?K-=faSS does.5ot reduce the contaminant levels sufficiently, the
stream may then enter another series of exchangers. Many ion
9 y
.
than one set
The other major portion of the ion exchange process concerns the
rStSSS "f^ Se resjn' which now holdl those ^urftiel
retained from the waste stream. An ion exchange unit with in-
place regeneration is shown in Figure VII-25 (page 261). Metal
rSSfn SUh-ha? "iCkel are removed by an acid? catLn exchange
resin, which is regenerated with hydrochloric or sulfuric acid
S£ia!;i"S- Khe ?etal ion with one or more hydrogen ions. Anion4
^ as _ dichromate ?r! removfed by a* basic, anion exchange resin?
which is regenerated with sodium hydroxide, replacing the anion
with one or more hydroxyl ions. The three principal methods
employed by industry for regenerating the spent resin are:
A) Replacement Service: A regeneration service replaces the
SXnf ™?in <-W-£h re9fne^ted resin, and regenerates the
S£M JS J \ i^S own *acillty- The service then has the
problem of treating and disposing of the spent regenerant.
B) In-Place Regeneration: Some establishments may find it less
expensive to do their own regeneration. The spent resin
column is shut down for perhaps an hour, and the spent resin
whi^he9»n;?a K* Jhis results in one or more waste streams
which must be treated in an appropriate manner.
re(5tuire ifc' usually
C) Cyclic Regeneration: In this process, the regeneration of
the spent resins takes place within the ion exchange unit
itself in alternating cycles with the ion removal process.
A regeneration frequency of twice an hour is typical. This
very short cycle time permits operation with a very small
quantity of resin and with fairly concentrated solutions,
resulting in a very compact system. Again, this process
varies^ according to application, but the regeneration cycle
generally begins with caustic being pumped through the anion
exchanger-, carrying out hexavalent chromium, for example, as
sodium dichromate. The sodium dichromate stream then passes
through a cation exchanger, converting the 'sodium dichromate
to chromic acid. After concentration by evaporation or
Other means, the chromic acid can be returned to the process
line. Meanwhile, the cation exchanger is regenerated with
195
-------
sulfuric acid, resulting in a waste acid stream containing
the metallic impurities removed earlier. Flushing the
exchangers with water completes the cycle. Thus, the
wastewater is purified and, in this example, chromic acid is
recovered. The ion exchangers, with newly regenerated
resin, then enter the ion removal cycle again.
Application and Performance. The list of pollutants for which
the—ion—exchange system has proven effective includes aluminum,
arsenic, cadmium, chromium (hexavalent and trivalent), copper,
cvanide, gold, iron, lead, manganese, nickel, selenium, silver,
tin, zinc, and more. Thus, it can be applied to a wide variety
of industrial concerns. Because of the heavy concentrations of
metals in their wastewater, the metal finishing industries uti-
lize ion exchange in several ways. As an end-of-pipe treatment,
ion exchange is certainly feasible, but its greatest value is in
recovery applications. It is commonly used as an integrated
treatment to recover rinse water and process chemicals. Some
electroplating facilities use ion exchange to concentrate and
purify plating baths. Also, many industrial concerns, including
a number of coil coating plants, use ion exchange to reduce salt
concentrations in incoming water sources.
Ion exchange is highly efficient at recovering metal bearing
solutions. Recovery of chromium, nickel, phosphate solution, and
sulfuric acid from anodizing is commercial. A chromic acid
recovery efficiency of 99.5 percent has been demonstrated.
Typical data for purification of rinse water have been reported
and are displayed in Table VII-25 (page 233).
Ion exchange is a versatile technology applicable to a great many
situations. This flexibility, along with its compact nature and
performance, makes ion exchange a very effective method of waste
water treatment. However, the resins in these systems can prove
to be a limiting factor. The thermal limits of the anion resins,
generally in the vicinity of 60°C, could prevent its use in
certain situations. Similarly, nitric acid, chromic acid, and
hydrogen peroxide can all damage the resins, as will iron,
manganese, and copper when present with sufficient concentrations
of dissolved oxygen. Removal of a particular trace contaminant
may be uneconomical because of the presence of other ionic
species that are preferentially removed. The regeneration of the
resins presents its own problems. The cost of the regenerative
chemicals can be high. In addition, the waste streams
originating from the regeneration process are extremely high in
pollutant concentrations, although low in volume. These must be
further processed for proper disposal.
196
-------
Operational Factors. Reliability: With the
1
utransrusiDie oelt. The belt passes through a compartmented tank
with ion exchange, washing, and regeneration sect ionS ?h£
'
21. Membrane Filtration
aissoxvea salts permeate the membrane. When the recirrniafino
l
197
-------
Application and Performance. Membrane filtration appears to be
applicable to any wastewater or process water containing metal
ions which can be precipitated using hydroxide, sulfide or
carbonate precipitation. It could function as the primary
treatment system, but also might find application as a polishing
treatment (after precipitation and settling) to ensure continued
compliance with metals limitations. Membrane filtration systems
are being used in a number of industrial applications,
particularly in the metal finishing area. They have also been
used for heavy metals removal in the metal fabrication industry
and the paper industry.
In the performance predictions for this technology, pollutant
concentrations are reduced to the levels shown in Table VI1-26
(page 234) unless lower levels are present in the influent
stream.
A major advantage of the membrane filtration system is that
installations can use most of the conventional end-of-pipe
systems that may already be in place. Removal efficiencies are
claimed to be excellent, even with sudden variation of pollutant
input rates; however, the effectiveness of the membrane
filtration system can be limited by clogging of the filters.
Because pH changes in the waste stream greatly intensify clogging
problems, the pH must be carefully monitored and controlled.
Clogging can force the shutdown of the system and may interfere
with production. In addition, relatively high capital cost of
this system may limit its use.
Operational Factors. Reliability! Membrane filtration has been
showntobe a very reliable system, provided that the pH is
strictly controlled. Improper pH can result in the clogging of
the membrane. Also, surges in the flow rate of the waste stream
must be controlled in order to prevent solids from passing
through the filter and into the effluent.
Maintainability: The membrane filters must be regularly
monitored, and cleaned or replaced as necessary. Depending on
the composition of the waste stream and its flow rate, frequent
cleaning of the filters may be required. Flushing with
hydrochloric acid for 6-24 hours will usually suffice. In
addition, the routine maintenance of pumps, valves, and other
plumbing is required.
Solid Waste Aspects: When the recirculating reagent-precipitate
slurry reaches 10 to 15 percent solids, it is pumped out of the
system. It can then be disposed of directly or it can undergo a
dewatering process. Because this sludge contains toxic metals,
it requires proper disposal.
198
-------
inulear|nm°"tt?a%?5^rbrane ""ration
coating" plaS?.b2S2CoS?hesete2te.haS ^ installed * one coil
22. Peat Adsorption
precipitation and sub2eq52?^ifrif icatfon ad3£stment for metals
required for chromium Ses Ssfng ?err?c' chlor?df TS iB ^S°
sulfide. The wastewater i= IK« terric cnioride and sodium
chamber called a kfer which conta?nS PUI"?ed int° alar9« metal
Application and Performance.
PPeOon
detergents, and dy4s PeaJ aSL °^?anic. «ltt«' such as oil
comme?claliy at a fextill ol»S P °n ls currently used
metal reclamation Operation P ' 3 newsPrlnt facility, and a
s
adjustment fo pcciutli anby "ar^icaUon8 ^^ **
199
-------
Advantages and Limitations. The major advantages of the system
include it~ability to yield low pollutant concentrations, its
broad scope in terms of the pollutants eliminated, and its
capacity to accept wide variations of waste water composition.
Limitations include the cost of purchasing, storing, and
disposing of the peat moss; the necessity for regular replacement
of the peat may lead to high operation and maintenance costs.
Also, the pH adjustment must be altered according to the
composition of the waste stream.
Operational Factors. Reliability: The question of long term
reliability is not yet fully answered. Although the manufacturer
reports it to be a highly reliable system, operating experience
is needed to verify the claim.
Maintainability: The peat moss used in this Process soon
Exhausts its capacity to adsorb pollutants. At that time, the
kie?s must bl opined, the peat removed, and fresh peat placed
inKIe Although thiSP procedure is easily and quickly
accomplished, it must be done at regular intervals, or the
system's efficiency drops drastically.
Solid Waste Aspects: After removal from the kier, the spent peat
mSst be eliminated. If incineration is used, precautions should
be taken to insure that those pollutants removed from the water
are not released again in the combustion process. Presence of
Sulfides in the spent^eat, for example, will give rise to sulfur
dioxide in the fumes from burning. The presence of significant
quantities of toxic heavy metals in canmaking wastewater will in
general preclude incineration of peat used in treating these
wastes.
Demonstration Status. Only three facilities currently .use
commercial adsorption systems in the United States -a textile
manufacturer, a newsprint facility, and a metal reclamation firm.
23. Reverse Osmosis
The process of osmosis involves the passage of a liquid through a
semipermeable membrane from a dilute to a more Concentrated
Solution. Reverse osmosis (RO).is an operation in which Pressure
is applied to the more concentrated solution, forcing the per-
meate to diffuse through the membrane and into the more Dilute
solution. This filtering action produces a concentrate and a
permeate on opposite sides of the membrane. The concentrate can
then be furthe? treated or returned to the original operation for
continued use, while the permeate water can be recycled for use
as clean water! Figure VII-26 (page 262) depicts a reverse
osmosis system.
200
-------
im
diameter and 0
.d
outside
201
-------
to clean and regenerate than either the spiral-wound or hoi loj
f^r-SSi caT DrpUh ES55 ^^.SiS
poroul polyurethane plug under pressure through the module
g.-«s?s-?i.
^
tank or sent on for further treatment.
202
-------
?«?°.2?,<. ° i ?2 ^ents, strongly acidic or basic solutions,
solvents, and other organic compounds can cause dissolution of
*X membrane. Poor rejection of some compounds such as borates
and low molecular -weight organics is another problem. Fouling of
inembranes by slightly soluble components in solution or colllids
iras caused failures, and fouling of membranes by feed waters with
high levels of suspended solids can be a problem. A f inal limi-
s™i°£olVnabili£y t0 treat °r a^ieve hifh coScentraUon w?£h
some solutions. Some concentrated solutions may have initial os-
motic pressures which are so high that they either exceed avail-
able operating pressures or are uneconomical to treat.
Operational Factors. Reliability: Very good reliability is
Jh^XSrW* «f°y9 ?S the Pr°per Pre^utions are taken to. iihimiJ!
>*Ii-?»a 5 ' ?£ foulina or degrading the membrane. Sufficient
5il 1 2?oviL theh:as^ stream prior to application of an RO system
application information needed to insure a successful
Membrane life is estimated to range from six
years' d«P««ling on the use of the system. Down
f lushing or cleaning is on the order of 2 hours as often
' nf* Wee?'" a.substantial Portion of maintenance time must
versePIsmos?s 3"1"9 *"* Pr«**lt«- installed -~* of the re-
WaSn^ASFeCtSS ?" a ilosed lo°P system utilizing RO there
wasSf lref?fyr«?-o»f , ?o«centrate a»d a "Animal amount of
?S j ? efiltration eliminates many solids before they
- p the bulldup ^° a minlm>""- The-
Demonstration Status. There are presently at least one hundred
iSSKi. osmosis^f?te water applications in a vlrie?y of
iJf?I i?' addljion to these, there are thirty to forty
units being used to provide pure process water for several
industries. Despite the many types^ and coSiSrationT of
membranes, only the spiral-wound cellulose acetate membrane has
n?dn?ldeS!Pread success in commercial applications. One calking
plant has reverse osmosis equipment in-place. t-nmaxing
24 . Sludge Bed Drying
h treatment procedure, sludge bed drying is employed to
th«- water content of a variety of sludges to the point
ndiney a5Lamen^bie to mechanical collection S* removal ?o
in f «J « These beds usually consist of 15 to 45 cm (6 to 18
in.) of sand^over a 30 cm (12 in.) deep gravel drain system made
up of 3 to 6 mm (1/8 to 1/4 in. ) graded gravel overlying drain
203
-------
tiles. Figure VII-28 (page 264) shows the construction of a
drying bed.
Drying beds are usually divided into sectional areas
approximately 7.5 meters (25 ft) wide x 30 to 60 meters (100 to
200 ft) long. The partitions may be earth embankments, but more
often are made of planks and supporting grooved posts.
To apply liquid sludge to the sand bed, a closed conduit or a
pressure pipeline with valved outlets at each sand bed section is
often employed. Another method of application is by means of an
open channel with appropriately placed side openings which are
controlled by slide gates. With either type of delivery system,
a concrete splash slab should be provided to receive the falling
sludge and prevent erosion of the sand surface.
Where it is necessary to dewater sludge continuously throughout
the year regardless of the weather, sludge beds may be covered
with a fiberglass reinforced plastic or other roof. Covered
drying beds permit a greater volume of sludge drying per year in
most climates because of the protection afforded from rain or
snow and because of more efficient control of temperature.
Depending on the climate, a combination of open and enclosed beds
will provide maximum utilization of the sludge bed drying
facilities. ;
Application and Performance. Sludge drying beds are a means of
dewatering sludge from clarifiers and thickeners. They are
widely used both in municipal and industrial treatment
facilities. .
Dewatering of sludge on sand beds occurs by two mechanisms:
filtration of water through the bed and evaporation of water as a
result of radiation and convection. Filtration is generally
complete in one to two days and may result in solids
concentrations as high as 15 to 20 percent. The rate of
filtration depends on the drainability of the sludge.
The rate of air drying of sludge is related to temperature,
relative humidity, and air velocity. Evaporation will proceed at
a constant rate to a critical moisture content, then at a falling
rate to an equilibrium moisture content. The average evaporation
rate for a sludge is about 75 percent of that from a free water
surface.
Advantages and Limitations. The main advantage of sludge drying
beds over other types of sludge dewatering is the relatively low
cost of construction, operation, and maintenance.
204
-------
Its disadvantages are the large area of land required and long
drying times that depend, to a great extent, on climate and
weather.
• Operational Factors. Reliability: Reliability is high with
favorable climactic conditions, proper bed design and care to
avoid excessive or unequal sludge application. If climatic
conditions in a given area are not favorable for adequate dryinq,
a cover may be necessary.
Maintainability: Maintenance consists basically of periodic
removal of the dried sludge. Sand removed from the drying bed
with the sludge must be replaced and the sand layer resurfaced.
The resurfacing of sludge beds is the major expense item in
sludge bed maintenance, but there are other areas which may
require attention. Underdrains occasionally become clogged and
have to be cleaned. Valves or sludge gates that control the flow
Of sludge to the beds must be kept watertight. Provision for
drainage of lines in winter should be provided to prevent damage
from freezing. The partitions between beds should be tight so
that sludge will not flow from one compartment to another. The
outer walls or banks around the beds should also be watertight,.
Solid Waste Aspects: The full sludge drying bed must either be
abandoned or the collected solids must be removed to a landfill.
These solids contain whatever metals or other materials were
settled in the clarifier. Metals will be present as hydroxides,
oxides, sulfides, or other salts. They have the potential for
leaching and contaminating ground water, whatever the location of
the semidried solids. Thus the abandoned bed or landfill should
include provision for runoff control and leachate monitoring.
Demonstration Status. Sludge beds have been in common use in
both municipal and industrial facilities for many years.
However, protection of ground water from contamination is not
always adequate.
25. Ultrafiltration
Ultrafiltration (UF) is a process which uses semipermeable
polymeric membranes to separate emulsified or colloidal materials
suspended in a liquid phase by pressurizing the liquid so that it
permeates the membrane. The membrane of an ultrafilter forms a
molecular screen which retains molecular particles based on their
differences in size, shape, and chemical structure. The membrane
permits passage of solvents and lower molecular weight molecules.
At present, an ultrafilter is capable of removing materials with
molecular weights in the range of 1,000 to 100,000 and particles
of comparable or larger sizes.
205
-------
In an ultrafiltration process, the feed solution is pumped
through a tubular membrane unit. Water and some low molecular
weight materials pass through the membrane under the applied
pressure of 10 to 100 psig. Emulsified oil droplets and
suspended particles are retained, concentrated, and removed
continuously. In contrast to ordinary filtration, retained
materials are washed off the membrane filter rather than held by
it. Figure VII-29 (page 265) represents the ultrafiltration
process.
Application and Performance. Ultrafiltration has potential
application to~~ canmaking plants for separation of oils and
residual solids from a variety of waste streams. In treating
canmaking wastewater its greatest applicability would be as a
polishing treatment to remove residual precipitated metals alter
chemical precipitation and clarification. Successful commercial
use, however, has been primarily for separation of emulsified
oils from wastewater. Hundreds of such units now operate in the
United States, treating emulsified oils from a variety of
industrial processes. Capacities of currently operating units
range from a few hundred gallons a week to 50,000 gallons per
day7 Concentration of oily emulsions to 60 percent oil or more
are possible. Oil concentrates of 40 percent or more are
generally suitable for incineration, and the permeate can be
treated further and in some cases recycled back to the process.
In this way, it is possible to eliminate contractor removal costs
for oil from some oily waste streams.
Table VII-28 (page 234) indicates ultrafiltration performance
(note that UF is not intended to remove dissolved solids). The
removal percentages shown are typical, but they can be influenced
by pH and other conditions. The high TSS level is unusual for
this technology and ultrafiltration is assumed to reduce the TSS
level by one-third after mixed media filtration.
The permeate or effluent from the ultrafiltration unit is
frequently of a quality that can be reused in industrial
applications or discharged directly. The concentrate or brine
from the ultrafiltration unit can be disposed of as any oily or
solid waste.
Advantages and Limitations. Ultrafiltration is sometimes an
attractive alternative to chemical treatment because of lower
capital equipment, installation, and operating costs, when
treating very high concentrations of oil or where suspended
solids removal to a very low concentration is required. It
places a positive barrier between pollutants and effluent which
reduces the possibility of extensive pollutant discharge due to
operator error or upset as may sometimes occur in settling and
206
-------
skimming systems. Alkaline values in alkaline cleaning solutions
can be recovered and reused in process. =>wiuuions
A. ,.limitation of ultrafiltration for treatment of process
effluents is its narrow temperature range (18<> to 30°C) for
satisfactory operation. Membrane life decreases with hiaher
ThSKf.tUreS' *? flUX incre*ses at elevated temperature^
Therefore, surface area requirements are a function of
temperature and become a tradeoff between initial costs and
^nia?em6K S?StS f°f *he membrane- In addition, ultrafiltration
cannot handle certain solutions. Strong oxidizing agents
KiT?nnSUa«d °^er °rgani£ ,«»»ounds can dissolve the Membrane!
Fouling is sometimes a problem, although the high velocity of the
wastewater normally creates enough turbulence to keep foulinq at
a minimum. Large solids particles can sometimes puncture the
Factors Reliability: The reliability of an
ultrafiltration system is dependent on the proper filtration,
settling or^other treatment of incoming waste streams to prevent
fa^h9® 1° the*e*brane- Careful pilot studies should be done in
each instance to determine necessary pretreatment steps and the
exact membrane type to be used.
Maintainability: A limited amount of regular maintenance is re-
S£T«*- M the Piping system. In addition, membranes must be
periodically changed. Maintenance associated with membrane plug-
ging can be reduced by selection of a membrane with optimum phy-
sical characteristics and sufficient velocity of the waste
loluJTAn i-hL.ih ?hten necesfarv to occasionally pass a detergent
solution through the system to remove an oil and grease film
m«i^ accumulates on the membrane. With proper maintenance
membrane life can be greater than twelve monthsT
Solid Waste Aspects: Ultrafiltration is used primarily to
o?«hT!L s°lidfv.and liquids. It therefore eliminates solid waste
problems when the solids (e.g., paint solids) can be recycled to
£™<-PS°C!:SS' Otherwise, the stream containing solids must be
treated by end-of-pipe treatment. In the most probable
applications within the coil coating category, the ultrafilter
would remove hydroxides or sulf ides of metals which have recovery
Demonstration Status. The ultrafiltration process is well
developed and commercially available for treatment of wastewater
or recovery of certain high molecular weight liquid and solid
contaminants. One canmaking plant has ultrafiltration equipment
in-place treating the entire plant wastewater flow and three or
207
-------
more have ultraf iltration as a pretreatment for small volume high
oil waste streams.
26. Vacuum Filtration
shown in Figure VII-30 (page 266).
clarifier sludge before vacuum filtering.
ssir n ™r s^* ssrs
percent to about 30 percent.
and can be conveniently handled.
some units with similar or greater serv^e life.
number
208
-------
S?Son f approximately 5 to 15 percent of the total time, if
carbonate buildup or other problems are unusually severe
maintenance ^time may be as high as 20 percent. For this relso*'
it is desirable to maintain one or more spare units. reason,
Demonstration Status. Vacuum filtration has been widely used for
e *
IK-PLANT TECHNOLOGIES
?oe^intSnt °f in-Pfant technology for the canmaking subcategory
is to reduce or eliminate the waterborne waste loads which
require end-of-pipe treatment and thereby improve the overall
r^nri1V?heSS °f • *" existin9 wastewater treatment systIS or
Kr-hnli the- re?uireme"ts of a new treatment system. In-plant
™«S??-°gy inY°lves ?Ptiraum machine configuration and operating
pr'actices8 9 ^proved rinsing and water conservation^
h Of the volume of wastewater which must be
discharged^ from a canmaking facility is of highest importance to
reducing the total discharge of pollutants from the* facility
DoJfufSni-^Tn"10^1 tr|ftment produces a constant concentration of
pollutants in the effluent, a major part of the oollutant
redSr?^6 J^K**10? re^uired in this subcategory is achfeved^y
reduction of the volume of water discharged.
Canwasher Configuration
,« °* a can^asher and the conditions under which
Jn • rJP ted ?ay ?ave f substantial impact on a plant's ability
to reduce wastewater flow to meet discharge requirements. The
factors discussed in the following paragraphs may havl
substantial impact in this area and should bCorSidered in any
to. ^"ce. ^astewater generation and discharged
h ^,thf?e internal ^ter reuse practices can
the introduction of new water into the canwasher at any
point except as feed water to the stage 5 rinse.
The basis configuration of a canwasher is established when it is
S«?^UCt?d ?r Kdurin9 a major modification. The classic
Si *S™i2E:18 HWn in FigUCe ni"6 (page 29> although almost
all canwashers have some modifications to this basic
dVring 5r.after installation. The arrangement
and flow is of primary importance. Minor
to
209
-------
Introduction of water in the last riser of a stage (shown in
FigEre III-4, page 27) can substantially reduce the water
required to achieve a given level of can cleanliness. This
technique applies the cleanest water to the can after it has been
wlshedwith less clean water. This process has some similarities
to countercurrent cascade rinsing and is estimated to be about
one-half as efficient resulting in a water use turn down ratio of
about 4.
The number, type and location of spray nozzles and risers is an
important consideration in canwasher effectiveness. Equilibrium
between the concentration of pollutants on the can surface and
the water in the recirculation sumps must be approached to attain
effective rinsing with a minimum of water use.
Oil removal (shown in Figure III-8, page 31) from the system is
desirable to promote the effectiveness of each succeeding stage
of the canwasher. A preliminary - or vesitbule - rinse as the
can enters the washer removes a substantial amount of oil in a
form that it may be recovered for reuse in bodymaker fluid. Oil
removal by skimming in a discharge or recirculation sump at each
stage can also remove oil from the system.
Recovery and reuse of oil from the bodymaker sumps and some
canwasher discharge points is sometimes feasible. This
possibility should not be overlooked both from the stand point of
reduced wastewater flow and the economics of oil use.
The internal reuse of_ water within the canwasher is the most
commonly practiced method of reducing water use and wastewater
discharge in canmaking. There are many ways in which water can
be reused in a canwasher.
Counterflow rinsing, (depicted in Figure III-7, page 30) for the
purpose of this document has been defined as the use of water
from the stage 5 rinse in the stage 3 rinse with no other water
used in the stage 3 rinse. This can completely eliminate the
requirement for new water at the stage 3 rinse.
In some cases, there may be a pH barrier to the reuse of water
from stage 5 to 3. This can be easily overcome by acidifying the
water between stage 5 and stage 3.
Water reuse at stage 1 uses wastewater from stage 3 for all of
the water requirement for this stage.
A vestibule rinse or prerinse added before the entrance to stage
1 can provide some advantage by reducing the amount ot oil to be
removed later in the canwasher. Water for this prerinse may be
210
-------
drawn from the stage 3 discharge. The heavy oil removed from the
can may usually be recovered for reuse in bodymaker fluid.
|fiM|ion ffi|k|U£ water. This water may be drawn from the stage 5
™i?nf .dlschar9e and. "sed as a feed into stage 4 and stage 2 to
maintain a proper fluid level and provide a slight overflow for
IKnSVh0* °n and dj?solved salt in each of these stlgls Even
discharge^ ar€ ' these flows contribute to pollutant
Process wfstewater may be regulated and used as part of
*. wter8UpplX is a demonstrated mechanism for
redunn h . , e mecansm or
reducing the total volume of water which must be discharged from
the canmaking operation. Because the wastewater treatment
™^e^mUCh °f ^he f>llutant introduced in the canwasher it can
At ?i^f€ V "^oV™?"0"' of the water flow to the canwasher
At least two plants in the subcategory use this water
conservation practice. water
Countercurrent Cascade Rinsing
cascade rinsing is a form of canwasher
warrants separate discussion because of the
yi. of water use' Rinse "ater requirements and
™, f?untf ^current rinsing may be influenced by the
volume of solution dragout carried into each rinse stage by the
material being rinsed, by the number of rinse stages used, by thJ
initial concentrations of impurities being removed, and by the
rUi™ o£r?dVCt ^ieaniiness required (See Figures III-3, 4 and 5,
of9countlrr,; . ?° °aSe5 are. Considered: first is the application
of countercurrent cascade rinsing to a simple water circuit
wh"?aS?Hr and P6' «PP"«tion to a more complicated circuit in
which the new water is introduced into the last riser of the
rinse stage. The influence of these factors is expressed in the
rinsing equation which is stated simply below:
A. Simple Water Circuit Canwasher
Vr is the flow through each rinse stage.
Co is the concentration of the contaminant (s) in the
initial process bath
Cf is the concentration of the contaminant (s) in the final
rinse to give acceptable product cleanliness.
n is the number of rinse stages employed
and . .
Vd is the drag-out carried into each rinse stage, expressed
211
-------
as a flow.
For convenience we cari set r = Co/Cf because for any calculation
about flow reduction, the cleanliness ratio Co/Cf_ is maintained
as a constant.. For a multi-stage rinse, the total volume of
rinse wastewater is equal to n times Vr, while for a
countercurrent rinse the total volume of wastewater discharge
equals Vr_.
Drag-out is solution which remains on the surface of material
when it is removed from process baths or rinses.
The potential flow reduction possible with countercurrent cascade
rinse is illustrated by the following analysis. To calculate the
cleanliness ratio, r, we start with an assumed water use of 215
1/1000 cans (the median plant water use of plants in the data
base) and subtract a 10 percent allowance for wastewater
generated from oil sump discharge, ion exchange regeneration,
fume scrubber discharge, and batch dumps of process tanks (i.e.
acid cleaner and conversion coating solution). Thus, 215 - 21.5
« 193.5 1/1000 cans represents the rinse water use for single
stage rinses. .
Without specific data available to determine drag-out we can
assume a dragout film thickness of 0.075 mm (2.9 mils) which is
equivalent to a poorly drained vertical surface film thickness;
and a surface area of 555 sq. cm for a standard 12-ounce can body
(can diameter is 6.5 cm and can height is 12.0 cm). The volume
of dragout or carryover is:
Vd » 555 sq cm/can x .0075 cm • 4.16 cu cm/can (ml/can) or 4.16
1/1000 cans
Given the configuration, of the inverted seamless can body as it
passes through the washer with a dished impression in the bottom,
4.16 ml per can carryover from one stage to the next by an
inverted can which has little time to drain, seems reasonable
especially when an air knife is used. Substituting in the
rinsing equation for a single stage rinse, Vr « r x Vd, and
solving for r, we get
r * 193.5 - 46.51
4.16
If a two stage countercurrent cascade rinse is substituted for
the single stage rinse, we get the following rinse water volumes
Vr « (46.51)1/2 (4.16)
" * 6.82 x 4.16
« 28.4 1/1000 cans
212
-------
If a three stage countercurrent cascade rinse is substituted for
the single stage rinse, we get for a rinse water volume:
Vr = (46.51) V* (4.16)
* 3 . 5 9 • x 4.1 6'
= 15.0 1/1000 cans
Similarly, the introduction of new water to the rinse at the
first riser will reduce the water required to achieve the
constant cleanliness ratio to 48.4 1/1000 cans. Addition of
first riser introduction of water to the first cascade of a 2
stage countercurrent cascade rinse will reduce the water
requirement to 9,4 1/TOOO cans. '
The application of countercurrent cascade rinse technology in the
DL.!l*nse, should also be considered. This would provide an
additional process station where surface contaminates can be
removed from the can surface and provide added insurance of can
cleanliness. ,
Equipment Maintenance
A canwasher is a unified sequence of process operations which
must be operationally coordinated to function optimally. Even
small maintenance omissions or failures can have a substantial
impact on water use and pollutant discharge. The failure or
reduced effectiveness of many functions may be compensated by
increasing the water flow and compensating the fault in can
rinsing rather than correcting the problem. Some examples are:
The failure of an air knife because of plugged jets, low air
pressure or other failure allows additional carryover of
pollutants into the stages that follow the failed air knife.
- The failure or decreased efficiency of a belt wiper between
stages can increase drag out into the following stages.
Decreased efficiency of circulating pumps can reduce the
rinsing effectiveness of rinse stages.
Cleaning and replacement of spray nozzles to ensure proper
effectiveness.
In-process Control
The conversion coating function is a key step of the canmaking
operation. This is one of the steps in which material is added
to the can. The two principal types of conversion coating used
on cans are chromating and phosphating.
213
-------
A number of parameters require monitoring and control to maximize
coating formation rate and minimize the amount of material
discarded.
All types of conversion coating operations require careful
monitoring and control of pH. If the pH is not kept at the
optimum level, either the chemical reaction proceeds too slowly
or the surface of the can is excessively etched. The pH of the
system can be sensed electronically and automatic make-up of
specific chemicals performed in accordance with manufacturers
specifications. Chemical suppliers provide a series of chemicals
for each type of conversion coating. The series includes a new
bath formulation and one or two replenishment chemicals depending
upon the constituent that has been depleted. This system
maximizes use of all chemicals and provides for a continued high
quality product.
Conversion coating temperature must be constantly monitored and
kept within an acceptable range. Low temperatures may slow film
formation and excessively high temperatures will degrade the
freshly formed film. For a given line speed, there should be
adequate spray nozzle coverage and pressure. This assures that
all areas of each can have sufficient reaction time to allow
buildup of a specified film thickness.
The chemicals used in chromate conversion coatings contain
significant quantities of hexavalent chromium. The hexavalent
chromium eventually becomes reduced to the trivalent state,
precluding its use as part of the film. Certain chromate
conversion coating systems are designed to regenerate chromium.
These systems pump the chromate conversion coating solution out
of the process tank to another tank where it is electrolytically
regenerated. This application of electrical current to the
solution increases the valence of the trivalent chromium to
hexavalent chromium. The solution is then returned to the
process tank.
In-Process Substitutions
The in-process substitutions for this subcategory involve only
the conversion coating phases of the total operation. The
cleaning, rinsing, and painting remain virtually unchanged.
These in-process substitutions eliminate the discharge ot a
significant pollutant from the conversion coating operation.
Certain chromating solutions contain cyanide ions to promote
faster reaction of the solution. Cyanide is a priority pollutant
which requires separate treatment to remove it once it is in
solution. Chromating conversion coatings are no longer widely
used in the canmaking subcategory, although it continues to be
214
-------
used in some plants. Where chromating systems are used chemical
formulations which x3o not contain cyanide are available Snd
efforts should be made to eliminate cyanide use where possible.
215
-------
TABLE VII-1
pH CONTROL EFFECT ON METALS REMOVAL
Day 1 Day 2 Day 3
In Out In Out In Out
pH Range 2.4-3.4 8.5-8.7 1.0-3.0 5.0-6.0 2.0-5.0 6.5-8.1
(mg/l)
TSS 39 8 16 19 16 7
Copper 312 0.22 120 5.12 107 0.66
Zinc 250 0.31 32.5 25.0 43.8 0.66
TABLE VI1-2
EFFECTIVENESS OF SODIUM HYDROXIDE FOR METALS REMOVAL
Day 1 Day 2 Day 3
In Out In Out In Out
pH Range 2.1-2.9 9.0-9.3 2.0-2.4 8.7-9.1 2.0-2.4 8.6-9.1
(mg/1)
Cr
Cu
Fe
Pb
Mn
Ni
Zn
TSS
0.097
0.063
9.24
1.0
0.11
0.077
.054
0.0
0.018
0.76
0.11
0^06
0.011
0.0
13
0.057
0.078
. 15.5
1.36
0.12
0.036
0.12
0.005
0.014
0.92
0.13
0.044
0.009
0.0
11
0.068
0.053
9.41
1 .45
0.11
0.069
0.19
0.005
0.019
. 0.95
0.11
0.044
0.011
0.037
11
216
-------
TABLE VI 1-3
EFFECTIVENESS OF LIME AND SODIUM HYDROXIDE FOR METALS REMOVAL
Day 1 Day 2 Day 3
In Out In Out In put
pH Range 9.2-9.6 8.3-9.8 9.2 7.6-8.1 9.6 7.8-8.2
(mg/1)
Al
Co
Cu
Fe
Mn
Ni
Se
Ti
Zn
TSS
37.3
3.92
0.65
137
175
6.86
28.6
143
18.5
4390
0.35
0.0
0.003
0.49
0.12
0.0
o.d
0.0
0.027
9
38.1
4.65
0.63
110
205
5.84
30.2
125
16.2
3595
0.35
0.0
0.003
0.57
0.012
0.0
0.0
0.0
0.044
13
29.9
4.37
0.72
208
245
5.63
27.4
115
17.0
2805
0.35
0.0
0.003
0.58
0.12
0.0
0.0
0.0
0.01
13
TABLE VII-4
THEORETICAL SOLUBILITIES OF HYDROXIDES AND «SULFIDES
OF SELECTED METALS IN PURE WATER
Solubility of metal ion, mq/1
Metal As Hydroxide As Carbonate As Sulfide
Cadmium (Cd++) 2.3 x 10-* 1.0 x 10~* 6 7 x 10-*o
Chromium (Cr+++) 8.4 x 10-* - No nrecioitate
Cobalt
-------
TABLE VI1-5
SAMPLING DATA FROM SULFIDE
PRECIPITATION-SEDIMENTATION SYSTEMS
Treatment
PH
(mg/1)
Cr+6
Cr
Cu
Fe'
Ni
Zn
Lime, FeS, Poly-
electrolyte,
Settle, Filter
In
Out
5.0-6.8 8-9
25.6
32.3
0.52
39.5
<0.014
<0.04
0.10
<0.07
Lime, FeS, Poly-
electrolyte,
Settle, Filter
In
7.7
Out
7.38
0.022 <0.020
2.4 <0.1
108 0.6
0.68 <0.1
33.9, <0.1
NaOH, Ferric
Chloride, Na2S
Clarify (1 stage)
In
Out
11.45 <.005
18.35 <.005
0.029 0.003
0.060 0.009
These data were obtained from three sources:
, ' '' " ' * . *
Summary Report, Control and Treatment Technology for the
Metal Finishing Industry; Sulfide Precipitation, USEPA, EPA
No. 625/8/80-003, 1979.
Industrial Finishing, Vol. 35, No. 11, November, 1979.
Electroplating sampling data from plant 27045.
218
-------
TABLE VI1-6
SULFIOE PRECIPITATION-SEDIMENTATION PERFORMANCE
Parameter
Cd
Cr
Cu
Pb
Hg
Ni
Ag
Zn
Treated Effluent
(mg/1)
0.01
0.05
0.05
0.01
0.03
0.05
0.05
0.01
Table VI1-6 is based on two reports:
Summary Report, Control and Treatment Technology for
iFOi^Msfmi?"- uui&^ButUad.lUt
o Development Document for Effluent Limitations
s and New Source Performance Standards. M«W
inorganic Products sSSmelTt of Inorganics Point '
/ EPA Contract No. EPA-68-01-328T-(TairfT7
219
-------
Table VI1-7
FERRITE CO-PRECIPITATION PERFORMANCE
Metal Influent(mg/1) Effluent(mg/1)
Mercury 7.4 0.001
Cadmium 240 0.008
Copper 10 0.010
Zinc 18 0.016
Chromium 10 <0/°i2
Manganese 12 0.007
Nickel 1,000 0.200
Iron 600 0.06
Bismuth 240 0.100
Lead 475 0.010
NOTE: These data are from:
Sources and Treatment of Wastewater in the Nonferrous
Metals Industry, USEPA, EPA No. 600/2-80-074, 1980.
TABLE VII-8
CONCENTRATION OF TOTAL CYANIDE
Plant
1057
33056
12052
Mean
Method
FeSO*
FeS04
ZnSO4
(mg/1)
In
2.57
2.42
3.28
0.14
0.16
0.46
0.12
Out
0.024
0.015
0.032
0.09
0.09
0.14
0.06
0.07
220
-------
Table VII-9
Plant ID t
06097
' 13924 .
18538
30172
36048
mean
MULTIMEDIA FILTER PERFORMANCE
TSS Effluent Concentration, ma/1
0.0, 0.0, 0.5
1.8,
3.0,
1.0
1 .4, 7
2.61
2.2,
2
.0,
• 0,
5.6,
5.6,
2.6,
0
5
4.0, 4.0, 3.0,
3.6, 2.4, 3.4
2.2, 2.8
TABLE VII-10
PERFORMANCE OF SELECTED SETTLING SYSTEMS
PLANT ID
01057
09025
1105.8
12075
19019
33617
40063
44062
46050
( SETTLING
DEVICE
Lagoon
Clarifier
Settling
Ponds
Clarifier
Settling
Pond
Settling
Tank
Clarifier
Lagoon
Clarifier
Clarifier
Settling
Tank
SUSPENDED SOLIDS CONCENTRATION (mg/1)
Day 1 Day 2 Day 3
In
54
1100
451
284
170
& -
4390
182
295
Out
6
9
-
17
6
1
_
9
13
10
In
56
1900
_
242
50
1662
3595
118
42
Out
6
12
10
1
16
12
14
10
In
50
1620
502
_
1298
2805
174
153
Out
5
5
14
4
1 3
" 23
8
221
-------
Table VI 1-1.1,
SKIMMING PERFORMANCE
Oil & Grease
mg/1
Plant Skimmer Type In Out
06058 API 224,669 17.9
06058 Belt 19.4 8.3
222
-------
TABLE VII-12
SELECTED PAFJITZON COEFFICIENTS
A
_..-„,, Log Octanol/Water
Priority Pollutant Partition Coefficient
Acenaphthene 4.33
1,1,1-Trichloroethane 2.17
1/1-Dichloroethane 1.79
1,1,2,2-Tetrachloroethane 2.56
Bis(2-chloroethyl)ether 1.58
Chloroform 1.97
1,1-Dichloroethylene 1.48
Fluoranthene 5.33
Methylene chloride 1.25
Pentachlorophenol 5.01
Bis(2-ethylhexyl)
phthalate 8.73
Butyl benzyl phthalate 5.80
Di-n-butyl phthalate 5.20
Benzo(a)anthracene 5.61
Benzo(a)pyrene 6.04
3,4-benzofiuoranthene 6.57
Benzo(k)f1uoranthene 6.84
Chrysene 5.61
Acenaphthylene 4.07
Anthracene 4.45
Benzo(ghi)perylene 7.23
Fluorene 4.. 18
Phenanthrene .4.46
Dibenzo(a,h)anthracene 5.97
IndenoO^^cdJpyrene 7.66
Pyrene 5.32
Tetrachloroethylene 2.88
Toluene .2.69
223
-------
TABLE VI1-13
TRACE ORGANIC REMOVAU BY SKIMMING
API SEPARATOR PLUS BELT SKIMMERS
(From Plant 06058)
Oil & Grease 225,000 14
-------
TABLE VII-15
L&S PERFORMANCE
ADDITIONAL POLLUTANTS
Pollutant Average Performance (ma/1)
Sb 0.7
0.51
0.30
g 0.06
fe 0.30
*% 0.10
Jh 0.50
A1 1.11
Co
0.05
F 14.5
TABLE VII-16
COMBINED METALS DATA SET - UNTREATED WASTEWATER
Pollutant Min. Cone (ma/1) Max. Cone, (ma/1)
<0.1 3.83
1 116
1 108
1 29.2
0.1 27.5
°«1• 337.
0.1 263
0.1 5.98
4.6 4390
225
-------
TABLE VI1-17
MAXIMUM POLLUTANT LEVEL ifo UNTREATED WASTEWATER
ADDITIONAL POLLUTANTS
Pollutant As & Se . Be..... Aq
As 4*2 •• ; - *
Be - 10.24
Cd
-------
TAfifcE VII-18
.''."'-', .'"'
PRECIPITATION-SETTLlNG-FIl.TRAi'ldN
Plant A
PERFORMANCE
Pari
For
For
Raw
ameters H
1979-Treated
Cr
Cu
Ni
Zn
Fe
1978-Treated
Cr
Cu
Ni
Zn
Fe
Waste
Cr
Cu
Ni
Zn
Fe
• • i
lo Pts
1 Wast
47
12
47
47
i. • F
.ewater
o.
0.
0.
0.
ianae i
015 -
01 -
08 ~
08 ~
Mean +
i»a/l std.
-------
TABLE VII-19
PRECIPITATION-SETTLING-FILTRATION (LS&F) PERFORMANCE
Plant B
Para
For
For
meters
No Pts.
Range mq/1
Mean + Mean + 2
std. dev. std. dev.
1979-Treated Wastewater
Cr
Cu
Ni
Zn
Fe
TSS
175
176
175
175
174
2
0.0
0.0
0.01
0.01
0.01
1.00
- 0.
- 0.
- 1.
- 0.
- 2.
- 1.
40
22
49
66
40
00
0.
0.
0.
0.
0.
068
024
219
054
303
+0.075
To,
+0
+0
+ 0
.021
.234
.064
.398
0.22
0,
.07
0.69
0.18
1
.10
1 978-Treated Wastewater
Cr
wv
Cu
Ni
Zn
Fe
Total 1974-1
Raw
Cr
Cu
Ni
Zn
Fe
Waste
Cr
Cu
Ni
Zn
Fe
TSS
144
143
143
131
144
979-Treated
1288
1290
1287
1273
1287
3
3
3
2
3
2
0.0
0.0
0.0
0.0
0.0
- 0.
,70
- 0.23
• - 1 .
- 0.
- 1 ,
.03
.24
.76
0.059
0.017
0.147
0.037
0.200
+0
+0
+0
+0
+ 0
.088
.020
.142
.034
.223
0
0
0
0
0
.24
.06
.43
.11
.47
Wastewater
0.0
0.0
0.0
0.0
0.0
2.80
0.09
1.61
2.35
3.13
177
- 0.56
- 0
- 1
' - 0
- 3
- 9
- 0
- 4
- 3
-35
-466
.23
.88
.66
.15
.15
.27
.89
.39
.9
.
0
0
0
0
0
5
0
3
22
.038
.011
.184
.035
.402
.90
.17
.33
.4
+0
+ 0
+ 0
+0
+0
.055
.016
.211
.045
.509
0
0
0
0
1
.15
.04
.60
.13
.42
228
-------
TABLE VI1-20
PRECIPITATION-SETTLING-FILTRATJQN (LS&F) PERFORMANCE
Plant C .
For Treated Wastewater Mean
Rane
103 0.010.- 0.50Q 0.049 +0.049 0.147
ol S'?nn " °'8" 0-29070.131 0.552
IS' ?:!°°:7::?° J:JI4 ?.;:043 3-as
For Untreated Wastewater
.£? - JJJ 0.0?9 - 2.319 0.542 £0.381 -1.304
Z« 103 0.949-29.8 11.009 +6.933 24.956
J,® 3 0.107 - 0.46 0.255 ~
T^ J03 0.80 -19.6 5.616 ±2.896 11.408
PH 103 6.8 - 8.2 7.6*
* pH value is median of 1Q3 values.
229
-------
TABLE VII-21
Pollutant
Parameter
SUMMARY OF TREATMENT EFFECTIVENESS
- : ( mg/1 ) ~"
L&S
Technology
System
114 Sb
115 AS
117 Be
118 ca
119 Cr
120 CU
121 CN
122 Pb
123 Hg
124 Ni
1 25 Se
126 Ag
127 Tl
128 Zn
Al
Co
F
Fe
Mn
P
Mean
0.70
0.51
0.079
0.084
0.58
0.07
0.12
0.06
0.74
0.30
0.10/
0.50
0.33
2.24
0.05
14.5
o!41
0.16
A!OB
One
Day
Max.
2.87
2.09
1.23
0.34
0.44
1.90
0.29
0.42
0.25
1.92
1.23
0.41
2.05
1.46
6.43
0.21
59. 5
1.20
0.68
16.7
Ten
Day
Avq.
1.28
0.86
0.51
0.15
0.18
1 .00
0 . 1 2
0.20
0.10
1,27
0.55
0.17
0.84
0.61
3.20
0.09
26. 4
0.61
0.29
6.83
Thirty
Day
Avq.
1.14
0.83
0.49
0.13
0.12
0.73
0. 1 1
0.16
0.10
1 .00
0.49
0.16
0.81
0.45
2.52
0.08
23.5
0.50
0.21
6.60
™
System
Mean
0.47
0.34
0.20
0.049
0.07
0.39
0.047
0.08
0.036
0.22
0.20
0.07
0.34
0.23
1 .49
One
Day
Max.
1.93
1.39
0.82
0.20
0.37
1 .28
0.20
0.28
0.15
0.55
0.82
0.29
1.40
1.02
6.11
Ten
Day
Avq.
0.86
0.57
0.34
0.08
0.15
0.61
0.08
0.13
0.06
0.37
0.37
0.12
0.57
0.42
2.71
Thirty
Day
Avq.
0.76
0.55
0.32
0.08
0.10
0.49
0.08
o.n
0.06
0.29
0.33
0.10
0.55
0.31
0.034 0.14
0.07
12.0
.!S:8
0.06
230
-------
TABtl VII-22
THBTABILITY mSXHG Of PRXOXXn MtEOTMITS
OTIIIZZWS CAKBON AOSORPTZON
Priority Pollutant:
•MBDTOl
Hating
priority Pollutant
1,
2.
3.
4.
5.
6.
7.
. 8.
9.
10.
11:
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
26.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38..
39.
40.
41.
42.
43.
45. Bethyl chloride
46. Methyl broad.de
47. bronoforn (tribroaaawthane)
48.
acanaphthana
acrolein
acryloixitrile
heniana
benzidlne
cmrbon tetrachlorida
(tetrachlorcBethane)
chlorobenzene
1,2,3-Urichlorobenzene
hexachlorobenzene
1,2-dichloroethane
1, 1* 1-txiehloroethane
haocachloroethane
1 , 1-diehloroethane
1,1,2-trichloroethana
1,1,2,2-tetrachlorathana
ehloroethaae
biiXchloroawtbyl) ether
bin(2-chloroethyl) ether
2-ehloroethylvinyl ether
(aixed)
2-ehloronaphthalene
2 ,4,6-trichlorophenol
parachloroetata cresol
chloroform (fcrichloroaethane)
2-chlorophenol
1,2-dieiU.orpbencene »
l,3-dichlorob«nrene
1,4-dieulorobenBene
3,3 '-dichlorobenridine
1 , 1-dichloroethylene
1 r 2-trans-dichloroethylene
2 , 4-dichloroph*nol
1 • 2 -diehloropropanc
l«2<-* )
2 ,4-dia»thylph«nol
2,4-dinitrotolncn*
2 ,6-dinitro«oliun«
1.2«diph*nylhydr>zina
luting
fluorantbon*
4-chloroph«nyl phcnyl «thor
4-bronoph«nyl ph«nyl «th«j,-
bi« ( 2 -chloroiiopropy 1 ) «thor
bla(2-chloro*«hoxy)n«thmno
n»fchyl«n« ehlorid*
(chloro«feh«n«)
M
a
H
H
n
n
D
H
II
H
I,
K
L
K
H
a
L
H
H
H
H
H
L
L
H
N
M
H
H
H
a
M
'H
H
H
H
M
I,
L
H
M
triehlorefluerontthaa*
dichlorodiflnoroaathan*
ehlorodibroaoaaithuM
hMudslerebotadica*
h«xachlorocyclop«ntadi«n«
isophoron*
naphthalcn*
nitrob«nzan«
2-nitroph«nol
4-nitroph«nol
2,4-dinitrophenol
4,«-dinitro-o-cr«Bol
M-nitrosodl»«thyl««in«
H-nitrosodiph«nyl««in«
H-nitrosodi-n-proprlaBin*
p«ntachloroph*nol
phanol
bis(2-«thylh«xyl)phthal«t«
batyl b«n«yl phthalat*
di-n-butyl phthalat*
di-n-octyl phthalata
diathyl phthalat*
di»«thyl phthalat*
1,2-banxanthracan*
(b«nKo(a)anthraesn«)
buizo(a)pyr«n« (3,4-bmm-
pyr«n«)
3,4-banzofluoranthan*
(twnzo(b) f luoranthan* )
ll,12-b«n«ofluoranth«n«
(b«nco( k) f Ivioran th«n« )
chry»«n«
acanaphthylana
anthracan*
l,12-b«uop«rylea« (tanso
•not* Explanation of Ranoval Ratings
Catagory H (high removal)
adsorbs
adsorb*
Category M
adsorbs
adsorbs
Cntaqory I,
49.
SO.
51.
52.
53.
54 .~
55.
56.
57.
58.
59.
60,
61.
62.
63.
64.
65.
66.
67.
63.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
(dibanro(a.h) antthzacww)
83. indano (1,2.3-cdJ pyrana
(2»3-o-ph«nyl«n« pyrana)
84. pyrana
85. tatrachloroathylenti
86. toluana
87. trichloroathylen*
88. vinyl, cblorida .
( chlozoathy lana )
105. SCS-1242 (Aroclor 1242)
107. PCB-1254 (Aroclor 1254)
108. PCB-1221 (Aroclor 1221)
109. PCB-1332 (Arocler 1232)
110. PCS-1248 (Aroclor 1248)
111. PCB-1260 (Aroclor 1260)
112. PCB-1016 (Aroclor 1016)
fluorana
phananthran*
a
a
a
a
R
a
a
a
a
N
M
a
a
a
a
a
a
at lavals i 100 mg/g carbon at C? - 10 ag/1
at lavals >100 «g/g carbon at C. < 1.0 «g/l
(•odarata raaoral)
at levels ilOO mg/g carbon at C- «• 10 »g/l
adsorbo
edsorbn
C£ - final
at l«vals £100 »g/g carbon at
(low raaoval)
** lavals < 100 ag/g carbon at C
-
< 1.0 ne/1
. 10 .ag/1
concentrations of priority pollutant at equilibria
at lavals < 10 mg/g carbon at C < 1.0 ' ag/1
231
-------
, • . TABLE VH -23
QF ORGANIC OO^OtMOS ADSORBED ON CARBON
Organic Ch*""1*"*1 Class
Arcnatic ^drocarbons
Etolynuclear Aronatics
Chlorinated Aronatics
Phenolics
Chorinatad Phcnolics
Molecular Weight Aliphatic and
Branch Chain hydrocarbons
Chlorinated Aliphatic hydrocarbons
Molecular Weicjht Aliphatic
Acids and Arooatic Acids
*High Molecular Weight Aliphatic
Aninea and Aromatic Amnea
*Hi^i Molecular Weight Ketones.
Esters, Ethers and Alcohols
Surfactants
Soluble Cfcganic Dyes
of Chemical Class
benzene, toluene, xylene
naphthalene, anthracanu
biphenyls
chlorobenzene, polychlorimted
biphenyls, aldrin* endrin,
toxaphene, DDT
phenol, cresol, resorcanol
and polyphenyls
trichlorophenoi, pentachloro-
phenol
gasoline, kerosine
carbon tetrachloride,
perchloroethylene
tar acids, benzoic acid
aniline, toluene diamine
hydroquinona, polyethylene
glycol
alkyl benzene sul£onates
nethylene blue, indigo carmine
* High Molecular Weight includes compounds in
4 to 20 carbon atoass
broad range o£ from
232
-------
Plant
A
B
C
Table VI1-24
ACTIVATED CARBON PERFORMANCE (MERCURY)
Mercury levels - ma/1
In
28.0
0.36
0.008
Out
0.9
0.015
0.0005
Table VI1-25
ION EXCHANGE PERFORMANCE
Parameter
All Values mg/1
Al
Cd
Cr+3
Cr+6
Cu
CN
Au
Fe
Pb
Mn
Ni
Ag
S04
Sn
Zn
Plant
Prior To
Purifi-
cation
5.6
5.7
3.1
7.1
4.5
9*8
—
7.4
. -
4.4
6.2
1.5
—
1.7
14.8
A
After
Purifi-
cation
0.20
0.00
0.01
0.01
0.09
0.04
^*
0.01
0.00
0.00
0.00
—
0.00
0.40
Plant
Prior To
Purifi-
cation
_
<•»
^^
43.0
3.40
2.30
_
V1.70
^'
1.60
9.10
210.00
1.10
_
B
After
Purifi-
cation
•"_ .
0.10
0.09
0.10
0.01
0.01
0.01
2.00
0.10
233
-------
Specific
Metal
Al
Cr, (+6)
Cr (T)
Cu
Fe
Pb
IN
Mi
Zn
TSS
Table VI1-26
MEMBRANE FILTRATION SYSTEM EFFLUENT
Manufacturers
Guarantee
6
0
0
. 0
0
0
0
0
0
0
.5
.02
.03
.1
.1
.05
.02
.1
.1
— f
Plant 19066
in Out
__
0.
4.
18.
288
0.
<0.
9.
2.
632
_
46
13
8
652
005
56
09
— .
0.
0.
0.
0.
0.
<0.
0.
0.
0.
_
01
018
043
3
01
005
017
046
1
Plant
In
__
5.
98.
8.
21.
0.
<0.
194
5.
13.
—
25
4
00
1
288
005
00
0
31022
Out
__
<0.
0.
0.
0.
0.
<0.
0.
0.
8.
«T
005
057
222
263
01
005
352
051
0
Predicted
Performance
0.05
0.20
0.30
0.05
0.02
0.40
0.10
1.0
Pollutant
(mg/l)
Cr+6
Cu
CN
Pb
Hg
Nl
Ag
Sb
Zn
Table VI1-27
PEAT ADSORPTION PERFORMANCE
In
35,000
250
36.0
20.0
.1.0
2.5
1.0
2.5
1.5
Table VI1-28
ULTRAFILTRATION PERFORMANCE
Out
0.04
0.24
0.7
0.025
0.02
0.07
0.05
0.9
0.25
Parameter
Oil (freon extractable)
COD
TSS
Total Solids
Feed (mo/1)
1230
8920
1380
2900
Permeate (mq/1)
4
148
13
296
234
-------
TABLE VI1-29
REMOVAL OF TOXIC ORGANICS BY OIL REMOVAL
Pollutant Parameter
001
038
055
062
065
066
068
078/081
080
084
085
086
087
097
098
107
110
a:
b:
acenaphthene
ethylbenzene
naphthalene
N-nitrosodiphenylamine
phenol
bis(2-ethylhexyl)phthalate
di-n-butyl phthalate
anthracene/phenanthrene
fluorene
pyrene
tetrachloroethylene
toluene
trichloroethylene
endosulfan sulfate
endrin
PCB-1254 (a)
PCB-1248 (b)
(mg/1)
Influent
Concentration
(ma/1)
5
0,
7
089
0.75
1 .5
0.18
1 .25
1 .27
2.0
0.76
0.075
4.2
0.16
4.8
0.012
0.066
1.1
1 .8
25.7
Effluent
Concentration
(ma/1)
ND
0.01
0.23
0.091
0,04
0.01
0.019
0.1
0.035
0.01
0.1
0.02
0.01
ND
0.005
0.005
0.005
0.690
™ PCB~1254, PCB-1221, PCB-1232 reported together.
PCB-1248, PCB-1260, PCB-1016 reported together. t09etner'
235
-------
TABLE VI1-30
CHEMICAL EMULSION BREAKING EFFICIENCIES
Concentrat ion (mg/1)
Parameter
O&G
TSS
O&G
TSS
O&G
TSS
O&G
Influent
6,060
2,612
13,000
18,400
21,300
540
680
1,060
2,300
12,500
13,800
1,650
2,200
3,470
7,200
Effluent
98
46
277
189
121
59
. 140
52
27
18
187
153
63
80
Reference
Sampling data*
Sampling data*
Sampling data**
Katnick and Pavilcius, 1978
*0il and grease and total suspended solids were taken as grab
samples before and after batch emulsion breaking treatment which
used alumn and polymer on emulsified rolling oil wastewater.
+011 and grease (grab) and total suspended solids<9"b) samples
were taken on three consecutive days from emulsified rolling
oil wastewater. A commercial demulsifier was used in this batch
treatment.
**0il and grease (grab) and total suspended solids (composite)
samples were taken on three consecutive days from emulsified
rolling oil wastewater. A commercial demulsifier (polymer)
was used in this batch treatment.
•n-This result is from a full-scale batch chemical treatment system
for emulsified oils from a steel rolling mill.
236
-------
10'
10 It 12 13
FIGURE VIM. COMPARATIVE SOLUBILITIES OF METAL HYDROXIDES
AND SULFIDE AS A FUNCTION OF pH
237
-------
n/ow '
238
-------
o
0
0
o
O G
O
°o c
coo
--O-
1
-------
(I/Bui) uoiinnuaauoo waniua paiBajj. uimuipeo
!<=«
e
240
-------
Ul
Ul
Ul
u.
u.
Ul
in
i
00
X
o
(|/fitu) uoiienuaauoQ luanm^ pajeajj. uimtuojqa
241
-------
.2
to
9
UJ
i
i
i
5
i
e
o
co ^
i *•
-L iu
5 S oe
co
u
ui
oe
(I/Bui) uoiienuaauoo
pajeaii Jaddoo
242
-------
=
0
-------
X
1
< :
•^ x —
X
X
%
@_JaLj:...
W
>
<5
i I! 8
UJ
iu
>
&"1*
^ iu
«35 U.
sr E "•
^ ~ ui
II 1,
i S S =»
•S = S z
a s « z s
I § =|i
« 3 « > S -J
«s
-------
3
^
.,
'
- , — , ,., „,
p
3 '
( )
3
— — ~
g
©
)
0
©
— ,,
<5V
HI
<5)
— T
>
.
'
5
—
_
—
©
•'•
,
—
_
IU
u.
— . u.
1 g
•I .1
g J. u
5 I Ms
5 IgB
1 « 0
i "• P
1 g
* E
g3
K3 |
IU
a
X
e
ce
a o
=
(I/Bui) uo!iBJjuaauo3 juanjyg
245
-------
-
0
©
s
•
•« i
j)
5
r»
©
(s>
S>
S
©
© <
- —
•
A—
"S
~1~~
4=
._
c
IQ iss IOC
, Concentration (mg/l) (Number of observations = 28)
IRE VII -10
1 SEDIMENTATION EFFECTIVENESS
1.0
Iron Raw Wasti
FIGl
HYDROXIDE PRECIPITATION
-
S
(l/Oui) UOIJUJ.U83U03 )uanm3
246
-------
•uu
i ^
vations *
ber of o
00
IVENESS
Ul
u.
u.
Ul
z
o
78*
1 SE ui
i 55
5* uj *C
U4 W«
« as 5
» e S
£2 f= s
-g
5:
u
ui
ee
Q.
Ul
• O
X
• e .
ce
o
z
(I/Bui) uoi)ej}U83uo3
, , o
paji&ij. asaue&uayy
247
-------
VEN
12
ENTATION EFFE
-
*sg
EC eo »-
is
1
tS
UJ
oc
IU
s
s
e
cc
o
(I/Bui) uoiienuasuoo tuaniyg paieajj. SSI
248
-------
e ._.
u u
E 1 s
5 2 x
282
U K Q
U
Id
S
li
D
U
DC
0
249
-------
EFFLUENT
INFLUENT
ALUM
SB
k.
WATER
LEVEL
STORED
BACKWASH
WATER
&
FILTER
COMPARTMENT
I •••"»— FILTER I
—K BACKWASH*- V
FILTER
MEDIA
SAND-^£g^£j
^?Wi:
u u u u
F&
[V U U U U
COLLECTION CHAMBER
COAL
U U U U U
DRAIN
FIGURE VI1-14. GRANULAR BED FILTRATION
250
-------
PERFORATED
BACKING PLATE
FABRIC
FILTER MEDIUM
SOLID
RECTANGULAR
END PLATS
INLET
SLUDGE
FABRIC
FILTKR MEDIUM
ENTRAPPED SOLIDS
FILTERED LIQUID OUTLET
PLATES AND FRAMES ARE
PRESSED TOGETHER DURING
FILTRATION CYCLE
RECTANGULAR
METAL PLATE
RECTANGULAR FRAME
FIGURE VI1-15. PRESSURE FILTRATION
251
-------
SEDIMENTATION BASIN
INLET ZONE
INLET LIQUID
BAFFLES TO MAINTAIN
QUIESCENT CONDITIONS
^•"•""•V^* * SETTLING PARTIJLf
• • • "*"*'*-«.l <• TRAJECTORY , •
r'X
OUTLET ZONE
OUTLET LIQUID
WELT-TYPE SOLIDS COLLECTION
MECHANISM
SETTLED PARTICLES COLLECTED
AND PERIODICALLY REMOVED
CIRCULAR CLARIFIER
SETTLING ZONE.
INLET LIQUID
.CIRCULAR BAFFLE
. ANNULAR OVERFLOW WEIR
INLET ZONE —l
• • . - • •
• • V • •
• • . * V • . *.
• • • x • •
I "_ • V • 4
• •
« •
•
-
/• LIQUID
*••**•-,,.... •
f » FLOW •
OUTLET LIQUID
•SETTLING PARTICLES
REVOLVING COLLECTION
MECHANISM
I
SETTLED PARTICLES
COLLECTED AND PERIODICALLY
REMOVED
SLUDGE DRAWOFF
FIGURE VI1-16. REPRESENTATIVE TYPES OF SEDIMENTATION
252
-------
FLANGE
WASTE WATER
INFLUENT
DISTRIBUTOR
WASH WATER
SUMP ACE WASH
MANIFOLD
•ACKWASH
BACKWASH
MBPbACEMENT CAMBON
CARSON REMOVAL PORT
TREATED WATER
SUPPORT PLATE
FIGURE VI.I-17. ACTIVATED CARBON ADSORPTION COLUMN
253
-------
LIQUID
OUTLET
SLUDGE
INLET
I""~"ZONE
—•OWL DRIVE I
[\M\JVl\JvrJV
CYCLOGEAR
BOWL REGULATING IMPELLER
RING
FIGURE VII-18. CENTRIFUGATION
254
-------
ill
K
§
255
-------
CONTROLS
OZONE
GENERATOR
DRY AIR
D
^ " '
li
OZONE
REACTION
TANK
TREATED
WASTE
X
RAW WASTE-
FIGURE Vll-20. TYPICAL OZONE PLANT FOR WASTE TREATMENT
256
-------
MIXER
0
WASTEWATER
PEED TANK
1
Pll
ST
SE
ST
Tl
SI
\,
t
CO
tST §
AGE j
3
»-
:OND §
AGE '5
*•;
-------
258
-------
OILY WATER
INFLUENT
WATER
DISCHARGE
OVERFLOW
SHUTOFF
VALVE
AIR IN
BACK PRESS
VALVE
TO SLUDGE
TANK •*
EXCESS
AIR OUT
LEVEL
CONTROLLER
FIGURE VI1-23. DISSOLVED AIR FLOTATION
259
-------
CONDUIT
TO MOTOR
INFLUENT
CONDUIT TO
OVERLOAD
ALARM
EFFLUENT FIFE
EFFLUENT CHANNEL
PLAN
TURNTABLE
BASE
HANDRAIL
INFLUENT
CENTER COLUMN
CENTER CAGE
WEIR
STILTS
CENTER SCRAFER
SQUEEGEE
SLUDGE FIFE
FIGURE VII-24. GRAVITY THICKENING
260
-------
WASTE WATER CONTAINING
DISSOLVED METALS OR
OTHER IONS
DIVERTER VALVE
REGENERANT
SOLUTION
DISTRIBUTOR
REGENERANT TO REUSE.
TREATMENT. OR DISPOSAL '
UPPORT
-DIVERTER VALVE
METAL-FREE WATER
FOR REUSE OR DISCHARGE
F8GURE VII-25. ION EXCHANGE WITH REGENERATION
261
-------
MACROMOLECULES
AND SOLIDS
MEMBRANE
AP»430PSI
WATER
PERMEATE (WATER)
-MEMBRANE CROSS SECTION.
IN TUBULAR, HOLLOW FIBER.
OR SPIRAL-WOUND CONFIGURATION
"(•-" -*t (r-*t •*.
™-^°ovi;v-A.:"-:-c
• • . ° »0 O/ 0*/%
Oo*o °°y0oo- o
• 1 •• *>
CONCENTRATE
(SALTS)
• 1 •' ^J! •
•I- - '•
O SALTS OR SOLIDS
• WATER MOLECULES
FIGURE VII-26. SIMPLIFIED REVERSE OSMOSIS SCHEMATIC
-------
PERMEATE
TUBE
FEED
ADHESIVE BOUND
SPIRAL MODULE
CONCENTRATE
PERMEATE
FLOW
BACKING MATERIAL
•MESH SPACER
•MEMBRANE \
SPIRAL MEMBRANE MODULE
SNAP
RING
"O" RING
SEAL
BRACKISH
WATER
FEED FLOW
BRINE
CONCENTRATE
FLOW
PRODUCT WATER
TUBULAR REVERSE: OSMOSIS MODULE
OPEN ENDS
OF FIBERS
• EPOXY
TUBE SHEET
POROUS
BACK-UP DISC
SNAP
RING
CONCENTRATE
OUTLET
END PLATE
POROUS FEED
DISTRIBUTOR TUBE-
PERMEATE
END PLATE
HOLLOW FIBER MODULE
FIGURE VII-27. REVERSE OSMOSIS MEMBRANE CONFIGURATIONS
263
-------
•-IN. Cl PIPE
PLAN
6-IN. FINE SAND
3-IN. COARSE SAND
3-IN. FINE GRAVEL
3-IN. MEDIUM GRAVEL
S TO « IN. COARSE GRAVEL
3-IN. MEDIUM GRAVEL
2-IN. PLANK
WALK
PIPE COLUMN FOR
GLASS-OVER
6-IN. UNDERDRAIN LAI1
WITH OPEN JOINTS
SECTION A-A
FIGURE Vll-28. SLUDGE DRYING BED
264
-------
ULTRAFILTRATION
• •
MACROMOLECULeS
* *
WATER SALTS
MEMBRANE
O OIL PARTICLES
• DISSOLVED SALTS AND LOW-MOLECULAR-WEIGHT ORGANIC*
FIGURE VII-29. SIMPLIFIED ULTRAFILTRATION FLOW SCHEMATIC
265
-------
FABRIC OR WIRE
FILTER MEDIA
STRETCHED OVER
REVOLVING DRUM
ROLLER
DIRECTION OF ROTATION
STEEL
CYLINDRICAL
FRAME
VACUUM
SOURCE
LIQUID FORCE
THROUGH
MEDIA BY
MEANS OF
VACUUM
SOLIDS SCR APED
OFF FILTER MEDIA
SOLIDS COLLECTION
HOPPER
V
INLET LIQUID
TO BE
FILTERED
FILTERED LIQUID
FIGURE VI1-30. 'VACUUM FILTRATION
266
-------
w Ul
o ui
I- CO
_l
u.
oc
p
03
2
Cd
<*>
M
IH
0)
U
a
Ed
a!
«
a
O
M
CO
irl
W
ei
3
M
-Q.
O
CO
_J
a si
UJO i
267
-------
-------
SECTION VIII
COST OF WASTEWATER CONTROL AND TREATMENT
This section presents estimates of the costs of implementing the
major wastewater treatment and control technologies described in
Section VII. These cost estimates, together with the estimated
pollutant reduction performance.for each treatment and control
option presented in Sections IX, X, XI, and XII provide a basis
for evaluating the options presented and identification of the
best practicable control technology currently available (BPT),
best available technology economically achievable (BAT), best
demonstrated technology (BDT), and the appropriate technology for
pretreatment. The cost estimates also provide the basis for the
determination of the probable economic impact on the canmaking
subcategory of regulation at different pollutant discharge
levels. In addition, this section addresses nonwater quality
environmental impacts of wastewater treatment and control alter-
natives, including air pollution, noise pollution, solid wastes,
and energy requirements.
Briefly, the approach taken to estimate capital and annual costs
was as follows: first, for each regulatory option, several flow
rates were selected that covered the expected range in siee of
can manufacturing plants. Next, the characteristics of the
influent to wastewater treatment were specified based on
analytical data collected by the Agency from sampled plants (see
Section V). These flow rates and compositions were used as input
to a computer cost estimation model. Next, the cost data
estimated by the model were tabulated and plotted as cost curves.
Finally, the costs for each plant in the canmaking subcategory
were estimated by applying for each regulatory option a specific
plant's wastewater flow to the cost curves. These costs are the
cost basis for the Agency's economic impact analysis for this
subcategory.
CHANGES IN COSTS BETWEEN PROPOSAL AND PROMULGATION
Several substantive differences occurred in the cost assumptions
used to develop costs for promulgation from those used at
proposal. First, the raw wastewater characteristics used at
proposal were based in most cases on maximum values or raw waste
concentrations of the analytical data in the subcategory at a
flow of 27,100 liters per hour. For promulgated costs, after
reevaluating the data base and correcting errors, influent
concentrations were based on the mean values of sampling data at
a mean flow of 9,000 liters per hour. This revised base had a
tendency to lower costs compared to those calculated at proposal,
269
-------
due primarily to the decreased pollutant loading on the vacuum
filter.
Second, oil removal costs at promulgation were based on an
integrated technology set instead of a combination of independent
technologies as used at proposal. The integrated set, which
consisted of chemical emulsion breaking, dissolved air flotation
(DAF) and oil skimming, tended to result in lower costs compared
to the independent case since redundant equipment costs were
excluded (e.g. tanks, pumps). Also, oil skimming, when
integrated with DAF, was based on a belt skimmer instead of a
more costly continuous oil skimmer. In addition, a comparative
analysis was performed between proposal and promulgation to
examine the cost tradeoff between ultrafiltration and the
integrated technology set described above to accomplish oil
removal. The results showed that the integrated technology costs
were lower and were thus retained as the oil removal costs at
promulgation.
Third, the "six-tenths" rule was used to extrapolate cost data to
different size flows for proposed costs, while final costs were
developed and plotted for seven separate model flow rates and
characteristics yielding a more accurate estimate of compliance
costs. This revised approach generally tended to lower final
costs across the range of flows examined.
Fourth, costs for contract hauling of wastewater treatment sludge
were not included at proposal. They are included in costs at
promulgation. This tended .to increase the final costs over the
proposal costs.
Finally, several specific changes were made in many of the
modules; these are addressed in the discussion of each module
later in this section.
COST ESTIMATION METHODOLOGY
For the canmaking subcategory, cost estimation is accomplished
using a computer model which accepts inputs specifying the
treatment system to be estimated, chemical characteristics of the
raw waste streams, flow rates and operating schedules. This
model utilizes a computer-aided design of a wastewater treatment
system containing modules that are configured to reflect the
model wastewater treatment equipment at an individual plant. The
model designs each module and then executes a costing routine
that contains the cost data for each module. The capital and
annual costs from the costing routine are combined with capital
and annual costs for the other modules to yield the total costs
for that regulatory option. The process is then repeated for
each regulatory option.
270
-------
Each module was developed by coupling theoretical design informa-
tion from the technical literature with actual design data from
operating plants. These data are used to design the component
pieces of equipment in each module. Designing and estimating
costs for each piece of equipment separately permits greater
accuracy in the total estimated costs than if modules that
Included several pieces of equipment were the fundamental unit of
costing. This approach closely reflects the way a plant would
actually design and purchase its equipment. The resulting costs
are thus more closely tied to the actual costs that would be
incurred by the facility.
Overall Structure
The cost estimation model comprises two main parts: a material
design portion and a costing portion. The material design por-
tion uses input provided by the user to calculate design param-
eters for each module included in the treatment system. The
design parameters are then used as input to the costing routine,
which contains cost equations for each discrete component in the
system. The structure of the program is such that the entire
system is designed before any costs are estimated.
Throughout the
are tracked:
program, the following pollutants or parameters
Flow
Total suspended solids
pH
Temperature
Acidity
Aluminum
Ammonia
Antimony
Arsenic
Cadmium
Chromium (trivalent)
Chromium (hexavalent)
Cobalt
Copper
Cyanide (amenable
Cyanide (total)
Fluoride
Iron
Lead
Manganese
Nickel
Oil & Grease
Phosphorus
Selenium
Silver
Thallium
Zinc
to chlorination)
The overall logic flow of the computer programs is depicted in
Figure VI11-1 (page 291). First, constants are initialized and
certain variables such as the modules to be included, the system
configuration, plant and wastewater flows, compositions, and
entry points are specified by the user. Each module is designed
utilizing the appropriate flow and composition data for influent
streams. The design values are transmitted to the cost routine.
The appropriate cost equations are applied, and the module costs
and system costs are computed. Figures VII1-2 and VII1-3 (pages
271
-------
292 and 293) depict the logic flow diagrams in more detail for
the two major segments of the program.
System Input Data
Several data inputs are required to run the computer model.
First, the treatment modules to be costed and their sequence must
be specified. The sequence for each regulatory option is
determined from the treatment technology diagrams shown in
Section X. The hours of operation per day and number of days of
operation per year is required. The flow values and
characteristics must be specified for each wastewater stream
entering the treatment system. These values will dictate the
size and other parameters of components to be included.
These inputs are derived from actual data if costs are sought for
actual plants. Where costs are developed for representative
plants, flows and concentrations are derived from aggregated
data. For development of costs for the canmaking subcategory,
data from Section V were used.
Model Results
For a given plant, the model will generate comprehensive material
balances for each parameter tracked in the system. It will also
summarize design values for key equipment in each-treatment
module, and provide a tabulation of costs for each element in
each module, module summaries, total equipment costs, and system
capital and annual costs.
GENERAL COST FACTORS
Dollar Base - All costs are adjusted to first quarter 1982
dollars.
Cost Update Factors
Investment - Investment costs were updated using the EPA-Sewage
Treatment Plant Construction Cost Index. The value of this index
for the first quarter of 1982 is 414.0.
Operation and Maintenance Labor -^The ENB_ Skilled Labor Wage
Index—Is Used to update the portion of O&M costs attributable to
labor. The March 1982 value is 325.
Maintenance Materials - The producer price index published by the
Department of Labor, Bureau of Statistics is used. The March
1982 value of this index is 276.5.
272
-------
?»S £ * i "u T1?e ,Cnemical Engineering Producer Price Index for
industrial chemicals is used. This index is published biweekly
in Chemical Engineering magazine. The March 1982 value of this
index is 362.
" Updating power costs is accomplished by using the price
for the desired date for electricity and multiplying it by the
energy requirements for the module in kwhr equivalents;
Annual Costs
L|bor -A base labor rate for skilled labor of $9.00 per hour was
'¥X!' To*CCOUnt for supervisory personnel, 15 percent of ?he
i2£T«Jia£e **** includ?d- Pla"t overhead at 100 percent of the
combined base and supervisory labor charges is also included.
The resulting composite labor rate used in this study is $21 .00
per nour .
Operating Schedule - Two hundred and fifty days per year 24
hours per day was assumed. * y««, *•
Energy - An electrical cost of 4.83 cents/kwhr (March. 1982) was
Energy Review? °D "*" industrial Cost derived from DOE's Monthly
System Costs
Engineering - This was assumed to be 15 percent of the total
module cost.
Hit.' ThlS "aS 8SSUmed to be 10 percent of the
Contingency - This was assumed at 10 percent of the summed module
COS t •
Taxef §M Insurance - This was assumed at 1 percent of the total
capital cost.
Monitoring - These costs are estimated at $120 per sample, which
are in turn estimated according to the breakdown shown in Table
vm-i (page 288).
Capital Recovery - These costs for recovery of committed capital
followiS eUiatid "Sin9 a capital recovery factor, given by the
CRP = i +
273
-------
where CRF s capital recovery factor
i » interest rate (%), and
n * period (in years) of amortization
For this analysis, an interest rate of 12 percent and a period of
10 years were used. This.yields a CRF of 0.17698. This value is
multiplied by the total capital investment to give the annual
amortization charge.
TECHNOLOGY BASIS FOR COST ESTIMATION
Treatment technologies have been selected from among the larger
set of available alternatives discussed in Section VII after-
considering such factors as raw waste characteristics, typical
plant characteristics (e.g., location, production schedules,
product mix, and land availability), and present treatment
practices. Specific rationale for selection is addressed in
Sections IX, X, XI, and XII. Cost estimates for each technology
addressed in this section include investment costs and annual
costs for depreciation, capital, operation and maintenance, and
energy.
Options for existing sources and new sources were identified as
the treatment alternatives for the canmaking subcategory. The
technologies used, which were described in detail in Sections III
and VII; include:
Counterflow rinsing,
- Countercurrent cascade rinsing,
- Equalization,
- Chromium reduction,
- Chemical emulsion breaking,
Dissolved air flotation,
Oil skimming,
Chemical precipitation-sedimentation,
Vacuum filtration,
Multimedia filtration,
Contract hauling,
Ultrafiltration, and
Electrodialysis
The specific assumptions for each wastewater treatment module are
listed under the subheadings to follow. Costs are presented as a
function of influent wastewater flow rate except where noted in
the unit process assumptions.
New source costs are based on the characteristics of a "normal"
plant. The normal plant "is a concept developed to aid in the
estimation of new source costs and average plant characteristics.
The production size of the normal was determined by summing the
274
-------
production of all plants in the subcategory and dividing by the
total number of plants (696 million cans per year). Wastewater
flow for the normal plant was assumed equal to the average
production normalized flow for the subcategory and the raw waste
characteristics equal to the average pollutant concentrations
shown in Table V-l1. This normal plant was also used for
estimating pollutant reduction,benefits and other factors in the
following Sections.
Counterflow Rinsing
This technology is applied to product rinsing operations. It
involves a number of spray rinse stages, with product and rinse
water moving in opposite directions (more detail may be found in
Sections III and VII). This allows for significantly reduced
flow over single stage rinsing by reusing the rinse water from
the stage 5 rinse as the stage 3 rinse.
The counterflow rinsing equipment and costs were evaluated
against the modified countercurrent cascade rinsing costed at
proposal and found to have nearly identical costs except for the
$1000 allowance for installing a baffle. The previously
developed cost module for countercurrent cascade rinsing was thus
used to estimate the cost of counterflow rinsing.
Countercurrent Cascade Rinsing
The countercurrent cascade rinsing system used for estimating
costs for existing plants in this subcategory at proposal was
designed assuming that a tank for single stage rinse was already
installed. The tank was converted to a two stage countercurrent
operation by installing a baffle in the tank, recycle piping, an
additional spray rinsing system, artd em additional pump. The
cost of -the baffle was assumed to be constant at $1,000. A
centrifugal pump, rated for the influent flow rate was assumed to
be required. The spray rinsing system included additional spray
nozzles, valves, and instrumentation (conductivity monitor,
probe, controller, etc.). Installation costs were assumed to be
50 percent of the total equipment cost. Recycle piping costs at
20 percent and a retrofit allowance at 15 percent of the total
installed equipment cost were also added.
The countercurrent cascade rinsing design used as a basis for new
sources differs from the technology as applied in existing
sources. An extended stage canwasher operation was used as an
alternate basis for flow reduction since this represents for many
plants a suitable tradeoff between achievable water conservation
and the cost of additional equipment. Costs were developed for
this technology by adding additional spray rinsing units.
Additional piping, tankage, nozzles, and pumps were included to
275
-------
add three additional countercurrent cascade rinse stages to a
conventional six stage canwasher.
Operation and maintenance costs were calculated assuming 5
percent of the plant annual operating hours as operating and
maintenance labor and 2 percent of the capital cost as
maintenance materials costs. The capital and annual costs for
additional spray rinsing are presented in Figure VII1-4 (page
294) for existing sources. Costs for new source spray rinsing
for countercurrent cascade rinsing are also shown in Figure
VIII-4 (page 294).
Chromium Reduction
This technology can be applied to waste streams containing signi-
ficant concentrations of hexavalent chromium. Chromium in this
form will not precipitate until it has been reduced to the tri-
valent form. The waste stream is treated by addition of acid and
gaseous S02 dissolved in water in an agitated reaction vessel.
The SO2 is oxidized to sulfate while reducing the chromium. The
equipment required for this continuous stream includes an SO2
feed system (sulfonator), an H2SO4 feed system, a reactor vessel
and agitator, and a pump. The reaction pH is 2.5 and the SO2
dosage is a function of the influent loading of hexavalent
chromium. A conventional sulfonator is used to meter SO2 to the
reaction vessel. The mixer velocity gradient is 100 cm/sec/cm.
Annual costs are as follows:
(1) SOj. feed system
—SO2 cost at $0.11/kg ($0.25/lb)
—operation and maintenance labor requirements vary
from 437 hrs/yr at 4.5 kg SO2/day (10 Ib SO2/day)
to 5,440 hrs/yr at 4,540 kg S02/day (10,000 Ib SO2/day)
—energy requirements may vary from 570 kwh/yr at 4.5 kg SO2/day
{10 Ib SO2/day) to 31,000 kwh/yr at 4,540 kg
S02/day (10,000 Ib S02/day)
(2) H2S04 feed system
—operating and maintenance labor varies from 72 hrs/yr at
37.8 I/day (10 gpd) of 93 percent H2SO4 to 200 hrs/yr at
3,780 I/day (1,000 gpd)
—maintenance materials at 3 percent of the equipment
cost
—energy requirements for metering pump and storage
heating and lighting
276
-------
(3) Reactor vessel and agitator
—operation and maintenance labor at 120 hrs/yr
—electrical requirements for agitator
The capital and annual costs for this technology are shown in
Figure VI11-5 (page 294).
Equalization
Equalization tanks are of the vertical steel type with capacities
which vary as a function of flow rate. The detention time is
eight hours and the excess capacity is 20 percent. The tanks are
fitted with agitators with a horsepower requirement of 0.006
kw/T,000 liters (.03 hp/1,000 gallons) of capacity to prevent
sedimentation. A control system, valves, a pump, arid piping are
also included. The capital and annual costs are presented in
Figure VIIi-6 (page 296).
Chemical Emulsion.Breaking
Chemical emulsion breaking involves the separation of relatively
stable oil-water mixtures by addition of certain chemicals, in
this case, alum and polymer. To determine the capital and annual
costs, 400 mg/1 of alum and 2 mg/1 of polymer are assumed to be
added to waste streams containing emulsified oils. The equipment
included in the capital and annual costs for continuous operation
are as follows:
- Chemical feed system
V. Storage units
2. Dilution tanks
3. Conveyors and chemical feed lines
4. Chemical feed pumps
- Rapid mix tank
1. Tank
2. Mixer
3. Motor drive unit
- Flocculator Tank (retention time of 45 minutes)
- Pump
The stabilized oil-water mixture is then pumped to a flotation tank,
which is discussed under dissolved air flotation below.
277
-------
For the batch emulsion breaking unit, the following items are
included:
- Sulfuric acid feed system
1. Storage tanks or drums
2. Chemical feed pumps
- Tank (retention time of 8 hours)
- Agitator
- Effluent water pump
in either mode, alum, polymer, and sulfuric acid (93 percent)
costs wire assumed to be $0.257/kg ($0.118/lb) $4.95/kg
($2.25/lb) and $0.08/kg ($0.037/lb), respectively. The
breakpoint between batch and continuous modes is approximately
5,000 1/hr.
The capital and annual costs are presented in Figure VII1-7 (page
297).
Dissolved Air Flotation
Dissolved air flotation (DAF) is an oil removal method. It is
designed to function as a stand-alone device, but may also be
used in combination with emulsion breaking equipment to increase
oil removal efficiency. The DAF system costs include a slop tank
to allow for separation of the oil-water-air mixture leaving the
DAF unit. The DAF system is typically followed by oil skimming
to remove the oil-rich phase for disposal based on a continuous
oil-water separator. However, when the two technologies are used
in conjunction, oil skimming may be accomplished with a belt
skimmer for relatively low oil removal rates (less than 50 gal/hr
of oil), provided the oil-rich phase has formed a surface layer.
The belt skimmer is located in the slop tank, whose retention
time (4 hours) is assumed to be sufficient to allow the oily
surface layer to form.
Capital costs were obtained from various vendors for package DAF
units consisting of the following equipment:
dissolved air flotation unit
o rectangular tank
o sludge auger and drive
o float skimmer and drive
o distributors
- recycle-pressurization pump
air dissolution tank
- electrical equipment and instrumentation.
278
-------
Costs for the slop tank, an influent pump, a sludge pump, a
concrete slab and installation of the unit are also included.
Assumptions made in the design of the DAF system include:
hydraulic loading = 1 gpm/ft2
oil concentration in effluent = 10 mg/1
float composition: 10 wt percent oil and solids, 40 wt percent water
50 wt percent air
25 percent of influent TSS settles in the unit; 65 percent
emerges in the float
- installation time = 16 manhours
Operation and maintenance labor and process energy costs dominate
annual costs, according to the vendors contacted. Therefore,
material costs are assumed to be negligible. Operation of the
DAF unit requires approximately 200 hr/yr labor regardless of
unit size. Maintenance labor requirements are also assumed
constant at 20 hr/yr. Energy requirements range from 15,700
kwhr/yr for a 10 gpm unit to 75,300 kwhr/yr for a 500 gpm unit.
The capital and annual costs for dissolved air flotation used in
conjunction with oil skimming are shown in Figure VIII-8 (page
*• y * )• *
Oil Skimming
Oil skimming, when used in conjunction with DAF, includes the
following equipment:
belt skimmer
Oil storage tank (sized for 2 weeks of storage)
Recycle pump
Oil discharge pump
The capital and annual costs of oil skimming for this subcategory
are included with dissolved air flotation in Figure VIII-8 (page
298). The cost of oil skimming is estimated at approximately
$18,500 capital cost and $7,500 total annual cost.
Chemical Precipitation
Quicklime (CaO) or hydrated lime [Ca(OH)2] can be used to
precipitate toxic and other metals. Hydrated lime is commonly
used for wastewaters with low lime requirements since the use of
slakers, required for quicklime usage, is practical only for
large-volume application of lime. Due to the low lime dosage
requirements in this subcategory, hydrated lime is used for
costing. The lime dosage requirements were determined by the
279
-------
model using specific influent characteristics and flow derived
from wastewater data for representative canmaking operations.
The following equipment was included in the determination of
capital and annual costs based on continuous operation:
- Lime feed system
1. Storage units (sized for 30-day storage)
2. Dilution tanks (five minutes average retention)
3. Feed pumps
- Rapid mix tank (detention time of five minutes; mixer
velocity gradient is 300/sec)
- Clarifier (overflow rate is 7.3 Iph/m* (20.8 gph/ft2);
underflow solids is 3 percent)
1. Sludge rakes
2. Skimmer
3. Weirs
- Sludge pump
The model assumes that a 10 percent excess of lime is used, that
the final pH is 9.0, and the effluent pollutant concentrations
are based on the CMDB L&S treatment effectiveness values.
Batch operation assumes a two fiberglass or steel tank system (if
additional capacity is required, tanks are added in pairs) with
one lime feed system (includes one agitated mixing tank with
hydrated lime added manually in 22.7 kg (50 Ib) bags for every
two tanks), a sludge pump for up to four tanks, and a simple
control system. A lime storage shed is included for lime
addition rates > 90.7 kg/batch (200 Ib/batch).
O&M costs for the,continuous system are for operating and mainte-
nance labor for the clarifier and lime feed system, and the cost
for chemicals, maintenance materials, and energy. For the batch
mode, operational labor is assumed at one-half hour per batch for
lime addition up to 90.7 kg/batch (200 Ib/batch) and one hour per
batch for additional rates above 90.7 kg/batch (200 Ib/batch).
Maintenance labor is constant for the batch system at 52 hours
per year (one hour/week). Lime is $47.30/kkg ($43/toh) in 22.7
kg (50 Ib) bags and energy requirements and maintenance materials
are negligible.
The operating mode is selected based on an annualized cost com-
parison assuming a 1,200 mg/1 lime dosage. Three minor changes
were made to this module between proposal and promulgation.
First, the maximum volume for a single batch reactor tank was
280
-------
increased from 10,000 gallons to 25,000 gallons. Second, the
single batch duration was reduced from 12 hours to 8 hours.
Third, the minimum cost for a batch lime feed system was reduced
to $2,500 from $16,000. These changes were made to more
accurately reflect actual practice at plants. The net effect of
each is to decrease capital costs. The capital and annual costs
are presented in Figure VII-9 (page 299).
Multimedia Filtration
Multimedia filtration is used as a wastewater treatment polishing
device to remove suspended solids not removed in previous treat-
ment processes. The filter beds consist of graded layers of
gravel, coarse anthracite coal, and fine sand. The equipment
used to determine capital and annual costs are as follows:
- Influent storage tank sized for one backwash volume
- Gravity flow, vertical steel cylindrical filters
with media (anthracite, sand, and garnet)
- Backwash tank sized for one backwash volume
-Backwash pump to provide necessary flow and head for
backwash operations
- Piping, valves, and a control system
The hydraulic loading rate is 63.2 Iph/m* (180 gph/ft2) and the
backwash loading is 252.8 lph/ms (720 gph/ft«). The filter is
backwashed once per 24 hours for 10 minutes. The backwash volume
is provided from the stored filtrate. The backwash stream is
recycled to the clarifier. The capital and annual costs are
shown in Figure VIII-10 (page 300).
Effluent pollutant concentrations are based on the LS&F treatment
effectiveness data in Table VII-21.
Ultrafiltration
The Ultrafiltration process employs a semipermeable polymeric
membrane to remove colloidal material from a wastewater. In
contrast to multimedia filtration, Ultrafiltration does not
operate intermittently, i.e., retained materials are continuously
rather than periodically removed.
The equipment costed for this process includes:
Membrane modules
Equalization tank
Process tank
- Feed pump
Recirculation pump
Piping
281
-------
- Electrical and instrumentation
A flux rate of 0.51 lph/m« (1.46 gph/ft*) is applied in the
tubular module. . .'.•'..
Operation and maintenance labor is assumed to be negligible for
this module. Chemical costs include cleaning solution, caustic,
and acid for pH control. Maintenance materials primarily include
replacement of filter membranes, which are estimated to have a
two year life. The capital and annual costs for this technology
are presented in Figure VIII-11 (page 301).
Vacuum Filtration
The underflow from the clarifier is routed to a rotary precoat
vacuum filter, which dewaters the mostly hydroxide sludge (it
also includes calcium fluoride precipitate) to a cake of 20 per-
cent dry solids. The filtrate is recycled to the rapid mix tank
as seed material for sludge formation.
The capital costs, for the vacuum filter include the following:
- Vacuum filter with precoat but no sludge conditioning
- Housing
- Pump
The yield from the filter is assumed at 0.126 kg/hr/m2 (3
lb/hr/ft2) with a solids.capture of 95 percent. Housing for the
filter is required. Two changes were made to this module after
proposal. First, the housing costs were modified to account only
for the area required by the vacuum filter and peripheral
equipment. Second, the operating schedule was reduced to 8 hours
per day. At proposal, this schedule was equivalent to the number
of hours the plant operated. Costs are presented in Figure VIII-
12 (page 302). .
Electrodialvsis
Water to be used in rinsing operations in a canwasher may require
treatment prior to use to remove dissolved solids. One process
currently in use at a can manufacturing facility to reduce
dissolved solids levels is electrodialysis.
As shown in Figure VIII-13 (page 303), electrodialysis units
consist of alternating cationic and anionic membranes arranged
between two electrically charged plates. Due to the different
charges on the plates, cations and anions will tend to migrate in
opposite directions. Each alternating membrane allows passage of
only one type of ion. Thus, a solution concentrated with ions
will accumulate in every other chamber. The result is an ion
282
-------
concentrated stream (brine) and an ion depleted stream suitable
for use in a canwasher.
The amount of electricity required, which accounts for a
significant portion of the annual costs, is a strong function of
the ion concentration. Thus electrodialysis is most suited for
dilute solutions.
The electrodialysis process can be operated either continuously
or on a batch basis (which involves recirculation of the product
stream). Pretreatment of the incoming water (e.g., filtration,
aeration) may be required to minimize membrane fouling, depending
on the feed characteristics. However, it is unlikely to be
necessary for the application discussed here since the source
water should be relatively pure.
The required capacity of an electrodialysis plant can be
expressed as the number of stages and the membrane area per
Stage. The number of stages is determined from the desired
reduction in dissolved solids and the area required is determined
by the influent flow rate.
Direct capital costs include the costs for purchase and
installation of the electrodialysis equipment and storage for the
feed and prbduct streams. System capital costs include
engineering, contingency and contractor's fee, which are
estimated at 37.5 percent of the total direct capital costs.
Total capital costs are presented in Figure VIII-14 (page 304) as
a function of flow rate. These costs are based on one plant's
reported investment cost for installation of a 46,000 gallon per
day electrodialysis unit reducing solids from 700 mg/1 to 120
mg/1. .The unit included necessary pretreatment, storage of feed
and product, and pumping. The curve was developed for other flow
rates from the "six tenths" rule, where
/Installed) - /Installed) x /Flow rate A\o-«
V, Cost / Plant B V Cost /Plant A \Flow rate B/
Direct annual costs are derived from an EPA electrodialysis
demonstration unit. Based on a flow of 216,000 gpd, these costs
include:
283
-------
S/l,000 gal
Power
Operating Labor
Maintenance labor
Membrane Replacement
Filter Replacement
Total
At different flows these costs (except for power costs) were
adjusted downward slightly to reflect economies of scale. The
power cost/1000 gal remained the same since this requirement
should be directly proportional to the flow. To calculate total
annualized costs, amortization at 17.7 percent and taxes and
insurance at 1 percent of the total capital investment were added
to the direct annual costs. The total annualized costs are showri
as a function of flow rate in Figure VIII-15 (page 305).
Contract Hauling!
This module, which was not included at proposal, provides for
removal of sludges and oils to a nonhazardous disposal site. The
cost is a strong function of the distance to the disposal site.
A 50-mile round trip was assumed. This results in a disposal
cost of $0.40 per gallon and is shown in Figure VIII-16 (page
306).
SYSTEM COST DEVELOPMENT
Options considered for existing and new sources were costed as
follows:
Option A. This option includes chromium reduction, equali-zation,
chemical emulsion breaking, dissolved air flotation, oil
skimming, lime precipitation and sedimentation, vacuum
filtration, and contract hauling. A production normalized flow
of 215 1/1000 cans and individual plant data along with the costs
displayed in Figures VIII-17 and VIII-18 (pages 307 and 308) were
used to estimate compliance costs for BPT and PSES-0.
Option B. This option for end-of-pipe treatment is the same as
for option A. In addition costs for counterflow rinsing (from
Figure VIII-4) were combined with the end-of-pipe costs, and are
displayed in Figures VII1-19 and VII1-20 (pages 309 and 310). A
production normalized flow of 83.9 1/1000 cans and individual
plant data along with the costs displayed in Figures VIII-19 and
VII1-20 were used to estimate compliance costs for the
promulgated BAT and PSES.
284
-------
The normal plant characteristics were used to evaluate additional
cost options. Compliance costs for these options are displayed
in Table X-5 (page 335) and were based on the unit cost curves
displayed in this Section.
Option C. This option includes option B in^process costs and
adds polishing filtration to the end-of-pipe treatment.
Option D. This option included option B in-process costs and
added ultrafiltration to the end-of-pipe treatment. This option
was not re-evaluated for costing after proposal.
Option E. This option includes additional flow reduction
achieved by including additional spray rinse units to option B
in-process and end-of-pipe costs. A production normalized flow
of 63.6 1/1000 cans along with the unit costs were used to
estimate compliance costs for the promulgated new source
standards. They overstate the costs for a new source plant
because alternatively a plant can redesign a six stage
conventional canwasher to achieve adequate flow reduction.
Option F. This option includes option E costs and adds polishing
filtration to the end-of-pipe treatment.
Treatment Jjn Place
The costs shown on the figures are greenfield costs that do not
account for equipment that plants may already have in place.
When costs are computed for an actual plant that has some of the
equipment already installed, that cost component must be sub-
tracted from the total module cost before adding subsidiary costs
(costs such as engineering or contingency added at the system
level as a percentage of the installed equipment cost).
Following proposal, treatment in place at canmaking plants was
reevaluated. This information along with the costs presented in
this section were used for•calculating compliance costs for each
plant for each selected treatment option and summed. Results of
these calculations are presented in Table X-5 (page 335). These
costs were then used for the economic impact analysis.
NONWATER QUALITY ENVIRONMENTAL ASPECTS
Nonwater quality aspects including energy requirements of all of
the wastewater treatment technologies described in Section VII
are summarized in Tables VIII-2 and VIII-3 (pages 289 and 290).
General energy requirements are listed, the impact on
environmental air and noise pollution is noted, and solid waste
generation characteristics are summarized. The treatment
285
-------
processes are
processes in
processes in Table VII1-3.
Energy Aspects
Energy aspects of the wastewater treatment processes are
imoortant because of the impact of energy on natural resources
and the economy. Based on dcp information, the EPA determined a
current energy consumption of 4,051 million kwhr/yr for canmaking
operations in the subcategory, and 3.21 million kwhr/yr for
wlstewater treatment system operation. The energy requirements
?or the Option A (BPT) technology for direct dischargers is
approximately 0.76 million kwhr/yr. Due to the reduction in
wStewater flow, the BAT technology for direct dischargers should
oily ?eq£ire Approximately 0.30 million kwhr/yr. The energy
Requirements for PSES technology for indirect dischargers is
eltimated to be 7.92 million kwhr/yr. A new source normal plant
wlstewater treatment system would add 0.075 million kwhr/yr to
the energy requirement.
The energy requirements for the wastewater treatment options for
the subcategory are generally low. When compared to the total
pllnt energy usage, the wastewater treatment processes contribute
less than 1.0 percent to the overall energy usage.
Other Environmental Aspects
It is important to consider the impact of each treatment process
on water scarcity; air, noise, and radiation; and solid waste
pollution of the environment to preclude the development of an
adverse environmental impact.
Consumptive Water Loss. Where evaporative cooling mechanisms are
used, water TSii^may result and contribute to water scarcity
problems, a concern primarily in arid and semi-arid regions.
These treatment options do not require substantial evaporative
cooling and recycling which would cause a significant consumptive
water loss.
Air Pollution. In general, none of the wastewater handling and
treatment processes considered for this subcategory cause air
pollu?iSn problems. For the precipitation of hexavalent chromium
using SO, as a reducing agent, the potential exists for tne
evolution of S02 as a gas. However, proper design of the
treatment tank! and proper PH control eliminates this problem
incineration of waste oil lubricants could cause air pollution
problems which need to be controlled by suitable scrubbers or
Kecipitators, as well as proper incinerator operation and
Maintenance. The wastewater treatment sludges are not generally
286
-------
amenable to incineration because of their high noncombustible
solids content.
Noise and Radiation. None of the wastewater treatment processes
cause objectionable noise levels and none of the treatment
processes has any potential for radiation hazards.
Solid Waste. Costs for wastewater treatment sludge handling were
included in the costing analysis performed for the subcategory.
To estimate the amount of wastewater treatment sludge produced as
a result of the treatment technologies, a computer program is
used. This program takes into account the amount of each
pollutant element in the sludge at each treatment level given in
Tables- X-l and XI-1 (pages 331 and 347). A 20 percent solids
content of the sludge and a 10 percent excess of lime are the
essential calculation parameters. For new sources a normal plant
is used as the basis for cost estimating.
The lime precipitation and settling technology produces a sludge
with a high solids content, consisting of calcium salts, toxic
metals (chromium, copper, nickel and zinc), and other metals
(aluminum and manganese) and a high pH. When this waste stream
is subjected to the RCRA hazardous waste criteria, it is judged
to be nonhazardous and therefore no hazardous waste disposal
costs are attributed to disposal of the sludge.
Spent lubricating oil waste is also generated by canmaking plants
and is generally disposed of in a landfill or reclaimed by
contract waste haulers. Based upon dcp data, the quantity of
this spent lubricant is estimated to be 595,000 kg/yr (270,000
Ibs/yr) for a normal plant. Since the spent lubricant is
considered to be nonhazardous under RCRA criteria, there are no
RCRA related costs attributed to the disposal of this material.
287
-------
5 I * * &
0}
n . >H
8 g 8
0090
in in o in
oo CN in - ^*
r-t en en
288
-------
S 3 I S a I
! ! 2
i 1
i I
i
ll III I I I I I -I «!f I iljfilJ I
a* lii i i I i
i j
1.5
I
\ti
j
III
I II tiff
III I
I I I I I I % I I I I I
m «
Si 2
T T 5
i ^ * S
m r4 .4 e
i .
* I
ill
289
-------
ii
«M
! J
i x
•8 H t '!.
I i I I
I I I I
it I I i'litf 1 £
II i i 11 i i i
I al
B .* ..
I I 3 t I
I
i
X j
s I a = i I
I
1
!i
»^»
!
P
!
290
-------
INPUT USER-
SPECIFIED
VARIABLES
C START J
i-HZ
INITIALIZE
CONSTANTS
•±
EXECUTIVE
ROUTINE TO
CALL REQUIRED
MODULES
INPUT
DESIRED
MODULES
So
oo
II
O
5
x u
0 GC
ss
DESIGN
PARAMETERS
CALL COST
ROUTINE
1
CALL COST
EQUATIONS
FOR EACH
MODULE
COMPUTE
SYSTEM
COSTS
I
OUTPUT
COSTS
FIGURE VIII-1. GENERAL LOGIC DIAGRAM OF COMPUTER CO$T MODEL
291
-------
DATA FROM
PREVIOUSLY
UNTREATED
WASTEWATER
FLOW AND
CONCENTRATIONS
FROM PREVIOUS
MODULE
SPECIFY
CONSTANTS.
DESIGN VALUES
ASSUME INITIAL
VALUES FOR
RECYCLE
STREAMS
COMPUTE
MATERIAL
BALANCES
CALCULATE
DESIGN
VALUES
HAS
TEST PARAM-
ATER(S) CON-
VERGE?
YO
WANT DESIGN
VALUES TO
PRINT?
PRINT MATERIAL
BALANCES AND
DESIGN VALUES
GO TO NEXT
MODULE
FIGURE VIII-2. LOGIC DIAGRAM OF MODULE DESIGN PROCEDURE
292
-------
DESIGN VALUES
AND CONFIGURATION
FROM MATERIAL
BALANCE PROGRAM
I
CALLMODULE
SUBROUTINES
MODULE 1
COMPONENTS
MODULE 2
COMPONENTS
MODULE N
COMPONENTS
CALL COST
EQUATIONS
CALL COST
EQUATIONS
CALL COST
EQUATIONS
COST
EQUATIONS
i
COMPUTE
SUMMED
MODULE
COSTS
COMPUTE
SYSTEM
COSTS
i
OUTPUT
COSTS
i
CRETURN FOff\
NEXT PLANT}
FIGURE Vlll-3. LOGIC DIAGRAM OF THE COSTING ROUTINE
• •. " " 293
-------
(Z8.UVV\!-$) S1SOO
294
-------
in
©
O
m
o
O
O
D
Q
ui
DC
5
O
DC
z
u
K
(A
O
O
UI
DC
CM
O
in
o
(28,UVIAI-$) SJ.SOO
295
-------
S
N
Ul
cc
o
u.
(0
^
o
U
to
Ul
K
0
S1SOO
296
-------
DC
X
O
CD
ill
cc
CO
O
I
i
s
UJ
,X
O
K
O
LL.
p
CO
O
O
til
IE
O
sisoo
297
-------
cc
£
_j_
1
IU
Z
O
I
Q
LU
oc
O
u.
P
(A
O
O
CO
UJ
e
D
O
(Z8,UVV\l-$) S1SOO
298
-------
o
I
CL
O
UJ
cc
O.
e
z
^ Oul
25
UJ
02
o<
0>
UJ
DC
3
o
SJLSOO
299
-------
cc
X
Ul
s
3
K
O
I -
—1 {A
UL. r-
O
O
LU
CC
S1SOO
300
-------
(2
to
• o.
..
.3.
£
0»
O
O
I
<
tr
e
z
j_
O
co
o
e
2
o
o
Ul
DC
O
UL
(O
o
to
o
(zs ,avwi-$) sisoo
301
-------
DC
X
K
HI
2
u.
z
o
§
oc
u.
D
O
5
DC
O
LL
(A
O
U
ui
DC
S.LSOO
302
-------
§
£
ui
Z
'£
ea
o w
3l
ge
as
80 SO
U UJ
S 2
cc cc
UI uj
a. a.
o 2
— uj
tu >
gSES
22i2
UJ UI i g
f $ If
2
3
(0
I
5
o
e
UJ
UJ
Z
u.
O
5
DC
UI
§
UJ
CC
303
-------
CO
CO
o
o
DC
O
S
LU
U.
O
p
CO
Q.
O
•
<*
Ul
K
3
C9
(28, UVW-000'l$) S1SOO IVlldVO
304
-------
(zs, avw-ooo'u)
305
-------
<
O
O
cc
o
u.
P
(A
O
U
to
UJ
ff
) S1SOO
306
-------
\
n
o
n
10
M
o
(M
Q.
UJ
Q
O
oc
o
X
V)
E
IU
to n
f O
o
to
o
o
<0
s
10
O
Q.
<
O
e
Ul
oc
to
o
00
o
-------
10
n
5
QL
Ml
z
g
_J
UJ
Q
O
K
O
x
«o
K
O
in
1 s *
vt
(38. avw-*eoi) sisoo i viid vo i VJ.QI
8
too
OfZ
oE
z
g
06
UJ
EC
D
(9
308
-------
10
9
O
3
oo
w
oc
w
flC
§
CM
en
o
(O
01
(28, avwi- £oi) sisoo iviidva ivioi
5
0.
CD
Z
-I —
UJO)
QUJ
Om
"
2'
I
o»
UJ
OC
3
O
309
-------
1
\
\
\
\
z
<
o
Ul5)
QUJ
QC
8 1
CO
QC
0(0
Sz
<2
Dr
(fi*
CO
§
O
W
S
CM
g
O
CN
>
UJ
cc
o
CM
CM
(Z8, UVW- EOl) S1SOO IVlldVO 1V1OJL
310
-------
SECTION IX
BEST PRACTICABLE CONTROL TECHNOLOGY
CURRENTLY AVAILABLE
This section defines the effluent characteristics attainable
through the application of best practicable control technoloqv
currently available (BPT). BPT reflects the performance by
plants of various sizes, ages/ and manufacturing processes within
the canmaking subcategory.
The factors considered in defining BPT include"the total cost of
applying the technology in relation to the effluent reduction
benefits from such application, the age of equipment and
^facilities involved, the process employed, nonwater quality
environmental impacts (including energy requirements) and other
factors the Administrator considers appropriate. In general, the
'BPT level represents the average of the best existing
performances of plants of Various ages, sizes, processes or other
common characteristics. : Where existing performance is
inadequate, BPT may be transferred from a different subcategory
or category. Limitations based on transfer technology must be
supported by a conclusion that the technology is, indeed,
transferable and a reasonable prediction that it will be capable
6f achieving the prescribed effluent limits. See Tanners'
Council of America v_._ Train. BPT focuses on end-of-pipe
treatment rather than process changes or internal controls
except where such are common industry practice.
TECHNICAL APPROACH TO BPT
EPA first studied canmaking operations to identify the processes
used and the wastewaters generated during the canmaking process.
The information collected by EPA during the development of this
regulation is described in detail in Sections III and V. This
information includes complete and updated data -colLection
portfolios (dcp), data from engineering visits to seven plants
prior to proposal, data from engineering visits to seventeen
plants following proposal, and plant sampling and analysis data.
In addition, industry provided information following proposal,
including sampling and analysis data at fourteen canmaking
plants. The Agency -evaluated these data to determine what
constituted an appropriate BPT.
Canmaking consists of cupping, drawing and ironing, and washing,
where the cans are cleaned and prepared for the decoration
process. These process steps generate different wastewater
streams. In all wastestreams, as discussed in Sections III and
311
-------
IV, the volume of wastewater is related to the number of cans
processed. As discussed in detail in Section IV, canmaking is
regulated as a single subcategory. In this regulation, only
seamless cans made from uncoated stock are regulated, since no
process water is generated from the manufacture of seamed cans or
seamless cans made from coated stock.
BPT limitations are generally based on the average of the best
existing performance by plants of various ages, sizes, and unit
processes within the subcategory for control of familiar (i.e.
classical) pollutants. This document has already discussed some
of the factors which must be considered in establishing effluent
limitations based on BPT. The age of equipment and facilities
and the processes employed were taken into account and are
discussed fully in Section IV. Nonwater quality impacts
including energy requirements are considered in Section VIII.
The general approach to BPT for this subcategory is to treat all
canmaking wastewaters in a single (combined) treatment system.
Many plants combine wastewater for treatment because it is less
expensive than treating wastestreams separately. Oil, which is
used as a lubricant and coolant during the formation of the
seamless can body, and is removed during washing, must be removed
from the wastewater; and hexavalent chromium, where present, must
be reduced to the trivalent state so that it can be precipitated
and removed along with other metals. The dissolved metals,
phosphorus and fluoride must be precipitated and suspended
solids, including the precipitate, removed.
The final model end-of-pipe treatment technology for BPT is .oil
removal by skimming, dissolved air flotation, or emulsion
breaking or a combination of these technologies; chromium
reduction when necessary; lime precipitation of other pollutants;
and removal of precipitated solids by Stokes1 law sedimentation
("lime and settle" technology), (Figure IX-1, page 323). The
proposed model end-of-pipe treatment technology also included
cyanide precipitation where necessary, but this element was
deleted since cyanide was not found in canmaking wastewaters in
treatable quantities and was thus not regulated. Nonetheless,
cyanide compounds may be used in some conversion coatings so that
cyanide precipitation may be necessary in individual cases if
these coatings are used.
The strategy for BPT also includes flow normalization through
water flow reduction and water reuse practices. These practices
are commonly practiced in the subcategory and are described more
fully in Sections III and VII. The proposed BPT flow reduction
strategy was based on the average production normalized
wastewater flow among the 32 plants in the subcategory which EPA
believed practiced reuse of process wastewater within the
312
-------
canwasher. This proposed strategy was modified when additional
data was received which verified that 14 plants practice reuse
using counterflow technology within the canwasher. The final BPT
flow is based on the performance of the median plant among the 62
plants in the data base for which we have complete data (Figure
IX-2, page 324). Average production normalized data for each
plant was displayed in Table V-2 (page 54).
The final BPT limitations are mass-based since
concentration-based standards do not limit the quantity of
pollutants discharged. The BPT limitations were derived as the
product of the BPT flow and the overall effectiveness of the
model end-of-pipe treatment technology.
SELECTION OF POLLUTANT PARAMETERS FOR REGULATION
The pollutant parameters selected for BPT limitations in the
canmaking subcategory were frequently found at treatable
concentrations in wastewaters from some plants. Chromium, zinc,
aluminum, fluoride, oil and grease and TSS were frequently found
at treatable concentrations in the raw wastewaters of canmaking
plants. Chromium appears in wastewaters in treatable
concentrations as a result of its continued use in chromating
surface treatment in a few instances in the subcategory and as an
apparent result of dissolution of chrome-containing steel alloys
in canwashers by acid baths. Zinc appears in wastewaters as a
consequence of its use as an alloying agent in the aluminum strip
used for forming cans, and aluminum appears since it is the
principal raw material used. Fluoride is a constituent of
hydrofluoric acid, a common process chemical used in canmaking.
In addition, phosphorus was found in treatable concentrations in
the wastewaters of several canmaking plants, as a result of its
use in zirconium phosphate conversion coatings. Oil and grease
appears in wastewaters as a result of lubricants used in
canmaking cupping and ironing machines. See Section V for
details.
The pollutant parameters selected for BPT regulation are
chromium, zinc, aluminum, fluoride, phosphorus, oil and grease,
TSS and pH. These parameters are the same as proposed. pH is
regulated to assure the proper operation of the model end-of-pipe
treatment technology for solids removal (lime and settle) and to
assure optimum removal of all regulated pollutants except oil and
grease. Cyanide is not regulated since it was not found in
treatable concentrations in sampled canmaking wastewaters.
CANMAKING SUBCATEGORY BPT
BPT Flow Calculation
313
-------
The BPT limitations include reductions in flow since the best
performing plants in the subcategory achieve significant flow
reductions, as presented in Table V-7 (page 59). Most aluminum
canmaking plants provided sufficient information in their dcp to
calculate the production normalized process water use at plants
in the subcategory, which was used to establish BPT regulatory
flows.
The flow basis fpr BPT is the performance of the median plant
among the 62 plants in the subcategory for which we had complete
data. The median plant was defined as the plant in an even
numbered population of plants that will include one-half of the
population. The median plant was chosen in preference to the
average because the industry provides a skewed distribution of
flow values, as illustrated in Figure IX-2; five percent of the
62 plants for which we have complete data account for 16 percent
of the total flow. The production normalized water use for the
canmaking subcategory at BPT is 215.0 1/1000 cans.
Plants with production normalized flows significantly above the
flow used in calculating the BPT limitations will need to reduce
flows to meet the BPT limitations. Generally this reduction can
be made by using a number of commonly used techniques. These
techniques are related to the optimal operation of canwashers,
including reduction in the flow to the canwasher (.water
conservation); maintaining adequate recirculation within each
stage of the canwasher until equilibrium is achieved; turning off
the water supply to the canwasher when production is stopped;
cleaning or replacing plugged spray nozzles; and proper operation
and maintenance of the canwasher. These techniques, which are
described in more detail in Sections III and VII, are commonly
used and can be implemented at all canmaking plants in the
subcategory to achieve the BPT flow.
Prior to establishing the BPT flow, the Agency evaluated thirteen
specific factors which commenters identified following proposal
as possible barriers to the achievement of flow reductions.
These factors are:
o Customer requirements for end use .
o Quality of incoming fresh water
o Can bottom geometry with respect to drag-in and
drag-out
o Can geometry (height/diameter ratio)
o Washer age and design
o Customer can quality requirements
o Type of organic coating to be applied
o Type of lubricants to be washed off
o Surface finish on can forming tooling
o Type of label used
314
-------
o Insensitivity of water use to variations in number of
cans washed
o Size of canwasher
o Location of plant in arid or wet regions of the
country.
These factors were evaluated using data provided by commenters,
data contained in the data collection portfolios for the
industry, and data received on plant visits and in response to
Agency requests for further information after proposal. EPA
concluded that none of these thirteen factors will prevent the
achievement of the estimated flow reductions for this regulation
by any plant.
One factor examined is whether the taste of beer and other malt
beverages is more sensitive to contaminants than is the taste of
soft drinks, and that additional rinse water is therefore
required for beer cans than for soft drink cans. An additional
question examined is whether more water is necessary for light
beers than for heavier pilsners, lagers, or ales for the same
reason. The Agency examined canmaking .plants of four companies
which produce cans for both soft drinks arid beer, and additional
plants which produce cans for both light beer and other malt
beverages. EPA found that on the basis of information supplied
by the industry, wastewater flows in each plant do not vary with
the intended use of the can. Further, a number of the lowest
wastewater flow rates in the industry are found at plants which
manufacture cans primarily intended for beer. As a result, we
concluded that reduced flows are achievable regardless of whether
cans are manufactured for beer or for soft drinks.
Another factor examined is whether the quality of fresh makeup
water, which varies from location to location, restrains the
achievable flow reduction. The industry identified about three
plants as experiencing product quality problems related to the
quality of the fresh water supply. The Agency visited several of
those plants and talked with company officials, and we do not
believe that the specific product quality problems these plants
are experiencing are due to an excess of dissolved solids in the
fresh water supplied to the canwashers. In general, EPA
concludes that while site-specific water quality factors could
conceivably require additional water purification steps or the
addition of water treatment chemicals in a few instances, data
submitted by commenters and other data available in the record do
not support a contention that quality of makeup water limits the
degree of flow reduction achievable. The cost of such
pretreatment steps was examined and is included in Section VIII.
Another factor mentioned in comments is that routine production
stoppages restrict a company's ability to meet reduced water flow
315
-------
allowances, since water flow allowances are expressed as a
function of production. The Agency found no support for this
contention, since can plants can reduce or turn off the supply of
water to the washer during production stoppages.
Canwasher age and design, canwasher mat width, and can geometry
were also examined as factors which could affect a company's
ability to achieve the reduced water flow. EPA found only one of
these factors, age and design, to have any demonstrable relation
to water use. Water use at canmaking plants tends to vary with
age and design, but we visited several units of varying ages and
designs and found no engineering reason why improved recycle,
reuse, and water conservation practices cannot be implemented at
these canwashers to achieve the reduced flows of this reguation.
Commenters also asserted that the type of organic coating t6 be
applied, the type of lubricant to be washed off, the surface
finish on can tooling, and the type of label used all affect
achievable reductions in flow rates. Despite requests for
industry to provide data to substantiate these claims, only
general statements were provided for the record. In plant visits
and in subsequent information requests sent by EPA under the
authority of section 308 of the Act, attempts were made to
determine the possible effects of these factors, but no specific
data were obtained. The remaining factors identified by
commenters were similarly examined with similar results. The
Agency thus concludes that based on the record, these factors do
not appear to prevent any plant from achieving the flows used for
calculating t.he limitations and standards in this regulation.
In summary, the Agency has conducted numerous engineering plant
visits and exhaustively examined the information available in the
record, and finds no supportable reason why the BPT flow cannot
be -achieved in every canmaking plant. Since flow reductions for
BPT are demonstrated at at least 31 plants, the Agency concludes,
that the BPT flow can be achieved by all plants in the
subcategory.
BPT Treatment Effectiveness
The BPT model end-of-pipe treatment train for canmaking
wastewater consists of oil removal by skimming, dissolved air
flotation, chemical emulsion breaking, or a combination of these
technologies; chromium reduction when necessary; mixing and pH
adjustment of the combined wastewaters with lime to precipitate
metals; followed by Stokes1 law sedimentation ("lime and
settle"). This technology was selected as the model end-of-pipe
treatment technology since it is the most effective technology
for removing the pollutants of concern. Many plants in the
subcategory presently rely on dissolved air flotation (DAF) as
316
-------
the primary device for removing solids. The Agency noted this,
but determined that DAF is not as effective as lime and settle
for the removal of solids, based in part upon sampling data
submitted by the industry. See Tables V-13 (page 68) and V-17
(page 80) and the discussions in Section VII for further details.
Lime and settle technology is the model end-of-pipe treatment
technology for the removal of precipitated metals, fluoride,
phosphorus, and other solids. Lime (rather than caustic) is
necessary as a source of calcium in order to precipitate calcium
fluoride, which is the insoluble fluoride species. Eleven of the
62 plants for which we have complete data indicate that they
employ lime and settle technology. Further, four plants
indicated that they employ chromium reduction equipment, which
may be necessary in some cases to reduce hexavalent chromium to
trivalent chromium prior to precipitation and removal. Five
canmaking plants appear to have all elements of the model BPT
end,-of-pipe treatment technology described above already in
place.
Available sampling and analysis data from treated effluents in
the canmaking subcategory is inadequate to establish the
treatment effectiveness of lime and settle technology. As
described in Section V, the Can Manufacturers Institute (CMI) and
the United States Brewers Association (USBA) submitted sampling
and analysis data for fourteen plants. This data is presented in
Table V-16. Only three of these plants, Plants 530, 565, and
605, employ and optimally operate lime and settle treatment
technology, based on information submitted by companies and as
observed during plant visits. Of these, the first data day at
Plant 565 was rejected as anomalous, as inconsistent with
historical sampling at that plant, and with the remaining two
data days for the plant submitted by CMI and USBA. Thus, the
Agency determined that a total of eight days of sampling data
submitted by CMI and USBA was representative of optimally
operated end-of-pipe.treatment technology for removal of metals.
fluoride, phosphorus, and TSS.
For TSS, chromium, and zinc, the Agency determined that the
Combined Metals Data Base (CMDB) was the best available and most
appropriate basis for establishing the treatment effectiveness of
the model end-of-pipe treatment technology on wastewaters from
the canmaking subcategory. As described in Section VII, the CMDB
consists of 162 data points from 18 plants, (including one plant
in the canmaking subcategory), thus providing a larger data base
and better sampling reliability in comparison to the few other
data points available from the canmaking subcategory. Further,
this larger data base enhances the Agency's ability to estimate
317
-------
long-term performance and variability through statistical
analysis.
To determine whether this transfer of treatment effectiveness
data is appropriate, statistical tests of homogeneity were
applied prior to proposal to raw wastewaters from the canmaking
plants and the wastewaters of categories represented in the
combined metals data base. As described in Section VII, these
tests revealed .the canmaking raw wastewaters to be homogeneous
with the wastewaters of the categories represented in the
combined metals data base. Following proposal, the Agency
performed additional statistical analyses of untreated and
treated wastewaters, using EPA sampling data and data supplied by
CMI and USBA. These analyses confirmed the general homogeneity
of canmaking wastewaters with the wastewaters of the CMDB
categories, although this analysis showed the concentrations of
zinc in canmaking influent wastewaters are significantly lower
than those represented in the CMDB. Therefore, in the absence of
adequate data from optimally operating BPT end-of-pipe treatment
operating technology where it is installed at canmaking plants,
EPA considers transfer of treatment effectiveness data from the
combined metals data base to be appropriate.
This transfer of treatment effectiveness data is confirmed by the
eight data days of sampling submitted by CMI and USBA which
represent optimally operated lime and-settle treatment systems.
All eight of these data points meet the achievable concentrations
for TSS, chromium and zinc indicated by the CMDB and used in the
final regulation.
Due to the lack of adequate treatment effectiveness data for
aluminum in the canmaking subcategory, the achievable
concentration value for aluminum is based upon data from the
aluminum forming and coil coating categories. This value, 6.43
mg/1 as a daily maximum, is slightly increased from proposal to
reflect additional information received from the performance of
lime and settle treatment systems at aluminum forming plants. To
determine whether the transfer of this treatment effectiveness
data to the canmaking subcategory is appropriate, the Agency
compared the aluminum concentrations measured in raw and treated
wastewaters of the plants used to establish the treatment
effectiveness of aluminum with the concentrations of aluminum in
the wastewaters of canmaking plants. The comparison showed no
significant difference in the aluminum concentrations from the
two groups.
The aluminum concentration used in this regulation is confirmed
by Discharge Monitoring Report data (DMR) for one direct
discharger in the canmaking subcategory which employs and
optimally operates a lime and settle treatment system. These DMR
318
-------
data show that this plant met the aluminum concentration used in
this regulation for all but .two months in the past two years. In
addition, the Agency determined that this aluminum concentration
™JUe ^,WML?etu
-------
The' treatment^ effectiveness of the model oil removal technology
is well demonstrated, as presented in Section' VII. The final
concentration for oil and grease is presented in Table VI1-21,
and is the same as proposed! The sampling and analysis data
submitted by CMI and"USBA include 27 data days which represent
optimally operated oil removal technology, as presented in Table
V-16. Data for Plants 530, 578, 666, and 667 are not included in
this total since'these plants either 'do not employ the model oil '
removal technology or do not optimally operate the technology, as
determined by EPA during engineering visits to the plants. In
addition, the firfct day Desampling at .Plant 565 is not included
for the .treasons described ear.lier in the discussion of lime and
settle technology. / / ' ; .,
Based upon confidential information obtained by EPA during
engineering plant visits, the 13 influent samples provided by CMI
and'USBA were not representative of the total raw wastewater
since they exclude or pretreat oily wastewaters from the raw
wastewater prior'to the application of the model oil removal
technology. As a result, ?he data submitted by CMI and USBA were
useful for confirming the reasonableness of the BPT
concentrations but hot to establish these concentrations.
All the data supplied by CMl-and USBA which represent optimally
operated oil removal technology met the oil and grease
concentration used in 'this regulation. In addition, the Agency
has considered oil removal in DMR data from copper forming and
aluminum forming because these metal forming processes are
similar to canmaking processes and require the use of similar
lubricants. In particular, the treatment of oil and grease in
aluminum forming presents similar problems to canmaking. A11 °|
the 170 daily values for:oil and grease in aluminum forming DMR
data met the one-day limitation concentrations and all of the 46
monthly average values met 'the monthly average concentration
value. This provides a high degree of confidence that canmaking
plants can meet the oil and grease limitations.
Typical characteristics of total raw wastewater for the canmaking
subcategory are given in Table V-ll. The model end-of-pipe
treatment technology will reduce the concentration of regulated
pollutants to the levels described, in the lime and settle column
of Table VII-21. When these concentrations are multiplied by the
regulatory flow basis described above, the mass of regulated
pollutants allowed to be'discharged per 1000 cans is readily
calculated. Table IX-1 (page 322) shows the limitations derived
from this calculation.
EPA reviewed the data for regulated pollutant parameters to
determine how many plants are presently meeting the BPT mass
limits (see Table V-19, page 84 and Table V-20, page 85). Three
320
-------
sampled plants have all elements of the model treatment system in
plant js merin- f yfes: ^Lrpiiis^-affil
L
limitations for all pollutant parameters on a tSree sampMng
days, .while meeting the BPT regulatory flow on all three davs
a6
onallhrv, £°r uan rme
on all three sampling days, except for one aluminum data point.
Other sampled plants have some elements of the model treatment
system in place, but not all components. Including the thr el
plants described above, data from a total of fifteen plants SJJ2
examined: four plants sampled by EPA prior to proposal Ind
fourteen plants sampled by CMI and USBA after proposal (three
Ei™ ^W6rf Sam?Jed by both EPA and CMI and USBA). Each wS
sampled for three days for the eight regulated pollutant
datf16™^;,/1*1?1"9 a t0ta} °f 399 data points9 (tak?ngP mlslfng
met atP?i o? *lnS°«. acc?unt)' Mass limitations for chromium werl
n? ?A 5 \ 5* ?ata P°injs'' mass limits for zinc were met at 52
of 54 data points; mass limits for fluoride were met at 45 of 47
data points; and mass limits for phosphorus were met at 45 of 45
data points. TSS mass limits were met on 42 of 54 data points
BPT ^"iSy?,";;8 limits wf;e ?et,on 24 °f so data points.^ ?h2
BPT pH limits were met on 31 of the 49 sampling days for which pH
met onW?? o^??^' 2aSS;baSed °U and 9rea«e limitations were
SS ??mita?fons aS^easonlblS? these c-Pa^-nS/ the proposed
c°st §nd Effluent Reduction Benefits of BPT
Sn,?S£Si8^ing ?P^ the COSt Of aPPlying a technology must be
considered in relation to the effluent reduction benefits
by.such application. The quantity of pollutants removed
334> andthe "
anno.
P? S ^10" Of BPT is shown in Table X-5 (page 335). The
methodologies used in calculating these costs are presented in
Sections VIII The capital cost of BPT as an increment ICove tne
m?ri- in-place treatment equipment is estimated to be $0.743
million Annual cost of BPT for the canmaking subcategory is
r^i^SdK t0 be $°-645 milli°n- The quantity of pollStanil
J3??£5 2 ?V6Kra^ !faste by the BPT system for the subcategory is
estimated to be 3.79 million kg/yr including 2,234 kg/yr of toxic
pollutants. EPA believes that the effluent reduction berSf it
outweighs the cost of compliance with BPT. "auction oenent
321
-------
TABLE IX-1
BPT EFFLUENT LIMITATIONS
CANMAKIN<5 SUBCATEGOfcY
— ' ' " ' ' BPT Effluent Limitations
Pollutant or
Pollutant Property
a dbs)
*Chromium
Copper
T.^arl
Nickel
*Zinc
* Aluminum
*Fluoride 1
Iron
Manganese
*Phosphorus
*0il and Grease
*TSS
TTO
Maximum for Maximum tor
any one day monthly average
/I. 000. 000 cans manufactured
94.60
408.5
32.25
412.8
313.90
1382.45
2792.50
258 . 0
146.2
3590.50
4300.00
8815.00
68.8
*pH Within the range of 7.
(0,209) 38.70
(0.901) 215.0
(0.071) 27.95
(0.910) 273.05
(0.591) 131.15
(3.048) 688.00
(28.202) 5675.00
(0.569) 131.15
(0.322) 62.35
(7.916) 1468.45
(9.480) 2580.00
(19.434) 4192.50
(0.152) 32.25
0 to 10 at all times.
(0.085)
(0.474)
(0.062)
(0.602)
(0.289)
(1 .517)
(12.513)
(0.289)
(0.137)
(3.237)
(5.688)
(9.243)
(0.071 )
*Regulated pollutant
322
-------
I
oe
I
323
-------
7N
i
5?
1
IM
§
s
9M
2H
_,©©<•>
©
e
©
©
©
©
©
TOTAL rOlMTS - 82
MEAN- M2-M
LNMEAN - 1M-3Z
VALUE THAT ENVELOPES
HK OF THE POINTS - 116.0
iPT REGULATORY FLOW - 216.0
FLOW RANKING VALUE
FIGURE IX-2. ALL USABLE PRODUCTION NORMALIZED FLOW DATA
324
-------
SECTION X
BEST AVAILABLE TECHNOLOGY ECONOMICALLY ACHIEVABLE
The effluent limitations in this section apply to existina
dischargers. A direct discharger is a facility which dilcarn
or may discharge pollutants into waters of the Sni ted StatS
Thisr section presents information on direct discharaers and ?
addition presents total subcategory data. " *
economically achievable performance of pan of various
•BPT 'tSilZ ' PrOCe^SeS er °ther shared characteristics. As i?th
is inadISn^eg°rieS Wh°?e existin9 treatment system performance
TECHNICAL APPROACH TO BAT
a-^
available technologies applicable to the subcategory.
I?T thu- Pr°P°sed regulation for the subcategory, three levels of
BAT which accomplish reduction in the discharge of tlxic
pollutants greater than that achieved at BPT were equated.
llmi'tatlOM' baSSd °" the following
reduction of hexavalent chromium, when necessary
precipitation of cyanide when necessary
srskSivS1.!? 'SiStiSi chemtcal emulsion
hydroxide precipitation and sedimentation of metals
water reuse '.
is^SKn^ssrs'S.'* rinse
sludge dewatering
325
-------
The proposed BAT limitations were presented as BAT Option 1,
which included all of the treatment technologies described above.
BAT Option 2 included all the treatment and flow reduction
0"-
3 (pages 337 to 339)
The Agency received comments criticizing the requirement of
gaacS -st .
.B -as-as ss- -
basis for BAT. While at least three plants are known to use
countercurrent cascade rinsing and can be used to achieve the BAT
flow, the model flow reduction technology basis for the final
BAT regulation is counterflow rinsing, "hich is demonstrated at
fourteen plants. For the purposes of establishing a BAT flow in
the final Regulation, counterflow rinsing is defined as .having
all of the makeup water for stage 3 (the rinse Allowing can
etching or cleaning) taken from the overflow of stage 5 (the
rinse following metal surface treatment).
BAT OPTION SELECTION
The final BAT limitations are based on BAT Option 1 which
consists of: flow reduction using counterflow rinsing; removal of
oil and grease using skimming, chemical emulsion breaking, or
dissolved air flotation, or a combination of these technologies,
chromium reduction where necessary; and removal of other
Dollutants using lime and settle technology. Cyanide
plecipttation is not included in the final model fnd-of-pipe
treatment technology for the reasons presented in Section IX.
Usina the methodology described later in this section, the Agency
determined that the selected BAT (Option 1) will remove 135 kg/yr
of tSxic pollutants incrementally over the pollutant removal
achieved by BPT. BAT Option 2 achieves little incremental
removal of toxic pollutants beyond BAT Option 1 (25.5 kg/yr of
£oxic pollutants over BAT Option 1 ) as calculated on a model
plant basis (See Table X-2, page 332), at an additional capital
cost of $0.017 million and an additional annual cost of $0.011
million BAT Option 3 was rejected for the same reasons. As a
resulT,' these options were not selected f or the canmaking
subcategory. The economic impact analysis indicates that the
selected BAT option is economically achievable.
Industry Cost and Effluent Reduction Benefits of Treatment
Options
326
-------
5'n, 'n "*•!»* re etlnated
"
production nL^Jized" mals
,
REGULATED POLLUTANT PARAMETERS
327
-------
V-21, page 86). See also Table XII-2, page 359 for a complete
listing.
Th^ nercent removal of organics by oil skimming from aluminum
iSJels of oil in canmaking and aluminum forming are similar
permits. : .
The toxic metals selected for specific BAT regulation are total
other regulated pollutants. •
£Winl5i^^
IdvSSely affect receiving waters at these concentrations, and
assures the removal of other toxic pollutants.
proper operation of lime and settle technology.
Proper pH control is essential to optimal °P«ation of \\™™*
Parameter
anl'eilurl
proper control
328
-------
CANMAKING SUBCATEGORY BAT
BAT Flow Calculation
1AT Effluent Limitations Calculation
329
-------
frhat the treated effluent concentrations used in this
es e'tra^nefecttvenss of the model end-of-plpe
technology in the canmaking subcategory.
^s0rxnrct«^^
m(lHon cans produced can be calculated. leX-e.^paae 336,
removed cSiScidenSuy if the regulated pollutants are removed to
the specified levels.
DEMONSTRATION STATUS
treatment technologies are both demonstrated.
330
-------
S3
e oo o c oe wo n!o
mo eso
S3
Ai
&l
o o m m to ot *0 i
in.
•o
i
r
itiirff
331
-------
CM
S
CM
at
CO
co
en P* p* in co CM
O VO VO OO «4* O
co in o ^
r** 00 O 00
OO
in
*3« VO
CM CO P~ VO VO
r— r-* CO TT r—
CO
m co in
•— in r-
vo
P-oo^rp*cooocoinenor*co
voovocMoocoinocninr-coP-
csoooocoincocnvoinvocn^'
r— i-i^a'oencMr-Or-'p-p*
rj. rj»cMocMcocn«in'*
•<* CM r- CM
CM 00
O CO ^l1
in o co
vo co m
*r in co
co P-
m o o
O r** •*•
in P* i*
r- CM CM
00 • •
en
co
oo
O •— I
PI* en p<* CM o ^* ^* P^ o o en
CMOCMOoooencocMoooP-
o en vo
en oo —•
CO
in
vom^-cnovococnoo-'j'o
COr—CO^rCM COCOO
r" oo CM in r-
en
co P^
o m
vo
en P~ o m co
P~ r- r- CO P~
--oooco^cor-mvovoo
i^ .«3«voencMr-CMr-co
r+* r~r~OCMCOOO«O
ff CM r- CM
CM CO
p^ CO 00
tn CM
CM i-*
vo vo
moo
in r- r—
• X X
en P* ^
co CM CM
en
vo
CM
o
•
tn
CM
vo
vo
•
en
*»-
tn vo CM o co CM o co en vo vo en CM
CM
co
CO
CM
VO
CO O Q
vo
vo
OOOr—VOr-CMOOCOOCOin
m
cn
co
CM
r— cnr»
co o m
r-co'«*
- co vo
oen*a'
- o
co
i CM
en
CM
co
vo vo
en o o
pg f~ r"
.XX
TT VO ^9"
^* CM CM
vo
O
x
VO
oo
«
vo
in
vo
•
m
p-
vo
co vo en
CM
CM
CM
vo-!
r- CO
r— r^ CM
CO
VO VO
vo o o
m H- r~
^s
VO CM CM
00 • •
332
-------
en
•
CO
00
CM
r-
««•
oo
-------
en
co
00
o
CM
ut
co m
r- r-
H» coo
o en **
o in
vo vo vo
enooococo'<*csinooino r- o o
• «««***»**X** • X X
I
co ^
o
•
m
r-
01
oo
cr>
•
oo
• CO ^* ^* O VO m
en in
r- CN
m co r-
CO ^* OO
en en
^
•s
s!
vo
3 O '
• "x
VO VO
en o o
CO <•*• r**
• XX
CO O\ r*
co r^ r-
cs • •
eg ro co
vo
in
en
vo
OJ
m
in
Ol CO
CO CO GO
*d*
CM
VO VO
en o o
oo
in
. X X
r r— cs
oo r*
CN co co
0)
o
r— VO
\ o
I I
Cu Cu
ti
™ Srt 9J
"S18
O ±3 O
^d
5
CO
334
-------
r
J3
00
£
en
".
o
Bu|
Cl
m
o
Q.
e
i l i
Q ta Du
m
x
a.
3
& 2
VO
^* r^ CN
O 1— I—
crt
2
(0
1
"M
I
1
in
^* o ^
vo X r^
o oo oo
«*>
OV (^
C4 O
m
es
^
2
W4
a
M T3
a £
c
a
2
i
2
2 A.. ^
& S 2
a .5
1
01
g
13 u
^ 2
i— CS
f- CS
f- 14-1 14-1 IM
0) rH
i
•
S S S
I I I
< CQ O
335
-------
TABLE X-6
BAT EFFLUENT LIMITATIONS
CANMAKING SUBCATEGORY
Pollutant or
Pollutant Property
BAT Effluent Limitations
Maximum fdr Maximum
any o.ne d&y monthly
for
average
* •„"...
(lbs)/l.000.000 cans manufactured
*Chromium
Copper
T./aa^j
AJwdVt
Nickel
*Zinc
* Aluminum
*Flupride
Iron
Manganese
*Phosphorus
Oil and Grease
TSS
.L Wh»>
TTO
36.92
159.41
35.24
161 .09
122.49
539.48
4992.05
100.68
57.05 ;
1401 .13
1678. 00 ?
3439.9
26.85
(0,081)
(0.351)
(0.078)
(0.355)
(0.270)
(1.189)
(11.001)
(0.222)
(0.126)
(3;X)89)
(3.700)
(7.584)
(0.059)
15.10
83.9
16.78
106.55
51.18
268.48
2214.96
51 .18
24.33
573.04
1006.8
1636.05
12.59
(0.033)
(0.185)
(0.037)
(0.235)
(0.113)
(0.592)
(4.883)
(0.113)
(0.054)
(1.263)
(2.220)
(3.607)
(0.028)
*Regulated Pollutant
336
-------
Ill
!||
III
9 if
5
337
-------
oe
5
IE
•I
<
i
i
338
-------
2
I
I
8
I
•7
x
339
-------
110
©
140
120
u.
Q
100
©
©
II
©
=>
Q
O
E
©
©
40
TOTAL POINTS - 12
MEAN -
IN MEAN - 12.77
VALUE THAT ENVELOPES
SOX OF THE POINTS - W.i
BAT REGULATORY FLOW - I3.S
JL
J_
J.
4 § 6 71
FLOW RANKING VALUE
IS
11 12
FIGURE X-4. PRODUCTION NORMALIZED FLOW DATA FOR PLANTS
UTILIZING COUNTERFLOW RINSING
340
-------
SECTION XI
NEW SOURCE PERFORMANCE STANDARDS
This section presents effluent characteristics attainable by new
sources through the application of the best available
demonstrated control technology, processes, operating methods, or
other alternatives, including where practicable, a standard
permitting no discharge of pollutants. Possible model NSPS
technologies are discussed with respect to costs, performance,
and effluent reduction benefits. The rationale for selecting one
of the technologies is outlined. The selection of pollutant
parameters for specific regulation is discussed, and discharge
limitations for the regulated pollutants are presented.
TECHNICAL APPROACH TO NSPS
In the proposed regulation, five NSPS options were evaluated.
The options were identical to or built on BAT technology options.
The BAT options and the discussion and evaluation of them carried
out in Section X are. incorporated here by specific reference
rather than repeated in this section.
NSPS Options 1, 2 and 3 presented at proposal were identical to
BAT Options 1, 2 and 3 respectively, which are described in
Section X. The schematic diagrams of those systems are presented
in Figures X-l through X-3. Schematic diagrams of NSPS Options 4
and 5 are presented in Figure XI-1, page 350, and Figure XI-2,
page 351, respectively. In summary form, the two additional NSPS
treatment options were:
NSPS Option 4:
additional in-process water use reduction achievable by
addition of three additional stages to a six-stage canwasher
.or its equivalent
end-of-pipe treatment (identical to NSPS Option 1)
• chromium reduction, when required
»•' cyanide removal, when required
• oil removal by chemical emulsion breaking, dissolved air
flotation, oil skimming, or a combination of these
technologies »-,
• lime precipitation
• Stojces1 law sedimentation
NSPS Option 5: All of NSPS Option 4 plus end-of-pipe polishing
filtration.
An option requiring no discharge of process wastewater pollutants
was also considered at proposal. One plant was believed to be
""'• 341 ' " •
-------
achieving this level of pollutant reduction using water use
reduction, ultrafiltration, reverse osmosis, and water reuse,
although this plant was subsequently found to discharge at the
rate of 2.36 1/1000 cans. This system for pollutant reduction is
costly; investment costs greater than $1.7 million and annual
costs; greater than $0.97 million are projected for a new
canmaking plant. This option is not considered as the basis for
NSPS because of the high costs associated with this technology.
The Agency received comments criticizing the flow reductions
achievable by the addition of three stages to a six-stage
canwasher, which was the principal proposed flow basis for NSPS
Options 4 and 5. Industry believed that this flow reduction
technology was not fully demonstrated and would not achieve the
proposed NSPS flow. In response to these and other comments, the
Agency reevaluated the flow reduction basis for NSPS. As a
result, the NSPS flow in the final regulation is based on the
lowest demonstrated plant flow which is generally applicable in
the subcategory. This flow is achieved by using counterflow
rinsing and other water flow reduction techniques.
NSPS OPTION SELECTION
The final NSPS are based on NSPS Option 4, which consists of:
flow reduction using counterflow rinsing and other techniques to
achieve the lowest plant flow which is generally applicable in
the subcategory; removal of oil and grease using skimming,
chemical emulsion breaking, or dissolved air flotation, or a
combination of these technologies; chromium reduction where
necessary; and removal of other pollutants using lime and settle
technology. Cyanide precipitation is not included in the final
model end-of-pipe treatment technology for the reasons presented
in Section IX.
Using the methodology described in Section VIII and later in this
Section, EPA estimates that a new direct discharge canmaking
plant having the industry average annual production level would
generate a raw waste of 862 kg/yr of toxic pollutants. NSPS
Option 4 would reduce these toxic pollutants to 65 kg/yr. In
contrast, NSPS Options 1, 2, and 5 would result in the discharge
of 72, 47, and 37 kg/yr of toxic pollutants, respectively.
Options 1, 2 and 3 were not selected because Option 4 provides
greater removal of pollutants and is economically iachievable.
Option 5 was not selected because the addition of filtration to
the small effluent flow would achieve little additional toxic
pollutant reduction.
EPA selected the final NSPS because it provides a reduced
discharge of all. pollutants below the final BAT (compare Table
XI-1 with Table X-l). NSPS Option 5 achieves little incremental
342
-------
removal of pollutants beyond NSPS Option 4 (26.4 kg/yr of toxic
pollutants as calculated for a normal plant, at an additional
capital cost of $0.017 and an additional annual cost of $0.009
million). The Agency has determined that the new source
performance standards will not pose a barrier to entry.
REGULATED POLLUTANT PARAMETERS
The raw wastewater concentrations from individual operations were
examined to select appropriate pollutant parameters for specific
regulation. In Section VI each of the toxic pollutants was
evaluated and a determination was made as to whether or not to
consider them further for regulation.. Pollutants were not
considered for regulation if they were not detected, detected at
nonquantifiable levels/ or not treatable using technologies
considered. The pollutant parameters selected for NSPS
regulation in the canmaking subcategory are: oil and grease, TSS,
chromium, zinc, aluminum, fluoride, phosphorus, and pH.
Each of these pollutant parameters is discussed in detail in
Sections IX and X and those discussions are incorporated here by
reference. Further information may also be found in Section VI.
In addition to the pollutant parameters listed above, there is
some, amount of toxic organic pollutants in the canmaking
wastewaters. The Agency is establishing an oil and grease
standard for new sources in order to control the oil soluble
organics found in these wastewaters. Although a specific numeric
standard for organic priority pollutants is not established,
adequate control is expected to be achieved by control of the oil
and grease wastes. This is projected to occur because of the
slight solubility of the compounds in water and their relatively
high solubility in oil. This difference in solubility will cause
the organics to accumulate in and be removed with the oil (See
Tables VII-12, VII-13, and VII-29, pages 223, 224, and 235), and
see the discussion in Section X).\
Other pollutants are also found in canmaking wastewaters,
including copper, nickel, lead, and manganese. These pollutants
are not regulated specifically because the Agency determined that
they would be removed coincidentally with other pollutants when
the model end-of-pipe treatment system is employed and properly
operated.
343
-------
CANMAKING SUBCATEGORY NSPS
Calculation of. NSPS Flow and Effluent Limitations
The NSPS regulatory wastewater flow for the canmaking subcategory
is 63.6 1/1000 cans. This regulatory flow is based on the lowest
demonstrated plant flow which is generally applicable in the
subcategory and represents a 70% reduction from the BPT
regulatory flow. This flow is based on the demonstrated
performance of Plant 555, which utilizes counterflow rinsing and
other water conservation practices to achieve this flow. These
practices and techniques are described in Sections III and VII.
This flow is also achievable by countercurrent cascade rinsing
techniques, as described in Section VII.
Plant 438 achieves a lower plant flow than the NSPS flow: 2.36
1/1000 cans in actual operation or 20.3 1/1000 cans when unique
in-plant water reuse practices are factored out. This plant was
not used as the basis for NSPS since the plant was not considered
to be generally applicable to the subcategory.
Prior to establishing this NSPS flow, the Agency considered
thirteen specific factors which commenters presented as possible
barriers to the achievement of the NSPS flow. These factors are
presented and discussed in detail in Section IX. For the same
reasons presented in that section, the Agency has determined none
of these factors will prevent the achievement of the NSPS flow by
any plant.
Pollutant parameters selected for regulation for NSPS fare:
chromium, zinc, aluminum, fluoride, phosphorus, oil and grease,
TSS, and pH. The NSPS end-of-pipe treatment technology will
achieve the effluent concentrations of regulated pollutants equal
to those shown in Section VII, Table VI1-21 for lime and settle
technology. pH must be maintained within the range 7.0 - 10.0 at
all times.
The Agency determined- the expected pollutant concentrations in
waste .streams following the NSPS flow reduction and compared
these expected concentrations to the raw wastewater (see Table
XI-1, page 347) concentrations of pollutants in the combined
metals data base. The range of these expected concentrations is
within the raw waste concentrations in plants in the CMDB and in
other categories used to establish treatment effectiveness, thus
showing that the treated effluent concentrations used in this
regulation can be achieved by canmaking plants after : the
application of NSPS flow reduction. The CMDB and the elements of
the NSPS end-of-pipe treatment technology are described in detail
in Section VII, and Section IX presents the rationale for
344
-------
establishing the treatment effectiveness of the model end-of-pipe
technology in the canmaking subcategory.
When these concentrations are applied to the water use described
above, the mass of pollutant allowed to be discharged per
1,000,000 cans produced can be calculated. Table XI-3, page 349,
shows the standards derived from this calculation.
Cost and Effluent Reduction Benefits of_ NSPS
In calculating NSPS costs, the production from a 696 million
cans/yr "normal plant" was multiplied by the NSPS regulatory
flow, to derive the plant flows for cost estimation. The added
cost of pipes, pumps and other parts to achieve the NSPS flow was
estimated. No plant-specific, production or construction cost is
included.
Because the technology on which the new source flow is based is
the same as for BAT there would be no incremental cost above BAT.
However, the Agency considered that some new sources might
install additional technology to meet the new source flows. For
a worst case evaluation the Agency considered that three
additional stages of countercurrent cascade rinsing might be
added beyond BAT. The total capital investment cost for a new
model canmaking plant to install NSPS technology for a worst case
situation is estimated to be $0.493 million, compared with
investment costs of $0.382 million for a model plant to install
technology equivalent to BAT. Similar figures for total annual
costs are $0.302 million for NSPS, compared with $0.267 million
for BAT. Thus, if the more expensive technology were used, NSPS
investment and annual costs would be about ten percent greater
than BAT costs for existing sources. These incremental costs for
NSPS over BAT would represent less than 0.1 percent of expected
revenues for a new source model plant. The Agency has determined
that the new source performance standards will not pose a barrier
to entry.
For costing, the proposed in-process costing model
(installation of three additional stages to a six stage
canwasher) was retained because plants can achieve the new source
flow using this technique. There would be no additional costs
above BAT for a new source to achieve NSPS using counterflow
rinsing technology, which is used at the plant used as the basis
for new sources.
The pollutant reduction benefit was derived by (a) characterizing
untreated wastewater and effluent from each treatment system in
terms of concentrations produced and production normalized
discharges for each pollutant considered for regulation and (b)
calculating the quantities removed and discharged annually by a
345
-------
"normal plant." Since NSPS apply to new sources, no treatment
equipment in place is assumed. Results of these calculations are
presented in Table XI-2 (page 348). All pollutant parameter
calculations were based on mean raw wastewater concentrations for
plants sampled by EPA before proposal (see table V-11, page 65).
DEMONSTRATION STATUS
Each major element of the NSPS technology is demonstrated in one
or more canmaking plants; however no sampled canmaking plant uses
all of the NSPS technology. Plant 555, the plant which is the
basis for the NSPS flow, lacks lime addition and oil removal
technology.
The NSPS model system has all the same treatment components of
the BAT model system plus further flow reduction. The NSPS flow
is demonstrated at two plants (although one plant exhibits
anomalies which prevent the applicability of its performance to
the entire subcategory). As discussed in detail in Section IX,
five plants have installed all elements of the model end-of-pipe
treatment system and the treatment effectiveness of the model
treatment system is confirmed by numerous data points within the
canmaking subcategory (see Section IX). Therefore, NSPS
technology is demonstrated in the subcategory.
346
-------
~~ G\
,S.
O O O O O ^ «* © O r*
I |«» VC
r" 09
•-•
r»
g§
;jg^^.«gg
> |u' ^
Ix OO»-JZN
-------
a
f** o ^"* ^* *4* oo vo CM o en r** o co vo en o
VO CM r- VO in f- r- Cn Cn «"!• r-» <» O •-* VO 00 VO CO
co ••* covococnoinr*cMvoo»inco vo«*r*-
VO T r* VO ^« i— CM CM r- CO O Ul
VO r- ^ r~ «* in
•fl> r~
VO ' VO VO
oocoini—'^i'vooococMoinen enop
^^^2 ^^ ^*°* °"C*'^° S ^^^
«tj" CM r— CM i-« r~
CM 00
j^ CM ^^ ^^ en ^i^ vo CM ^^ o^ oj ^^ ^j* co oo ^^ ^d* cs
vo CM t*^ vo CM CD vo f** en *"** co vo r^ CM vo en ^p en >UH
co »* covoinr-ia'cnr—ooe^ocMr»co ^» VOCMCM
^r CM *•• r— CM H- •-•
CM 00
01
348
-------
TABLE XI-3
NEW SOURCE PERFORMANCE STANDARDS
CANMAKING SUBCATEGORY
Pollutant or
Pollutant Property
NSPS
Maximum for
any one day
Maximum for
monthly average
q (lbs)/l.000,000 cans manufactured
*Chromium
Copper
Lead
Nickel
*Zinc
* Aluminum
*Fluoride
Iron
Manganese
* Phosphor us
*Oil and Grease
*TSS
TTO
27.98
120.84
26.71
122.11
92.86
408.95
3784.20
76.32
43.25
1062.12
1272.00
2607.60
20.35
*pH Within the range
(0.062)
(0.266)
(0.059)
(0.269)
(0.205)
(0.902)
(8.343)
(t).168)
(0.095)
(2.342)
(2.804)
(5.749)
(0.045)
of 7.0 to
11 .45
63.6
12.72
80.77
38.80
203.52
1679.04
38.80
18.44
434.39
763.20
1240.20
9.54
10 at all
(0.025)
(0.140)
(0.028)
(0.178)
(0.086)
(0.449)
(3.702)
(0.086)
<0.041 )
(0.958)
(1 .683)
(2.734)
(0.021 )
times
*Regulated Pollutant
349
-------
I
i
IU
i
350
-------
I
I
i
s
Ul
I
ri
X
351
-------
-------
SECTION XII
PRETREATMENT STANDARDS
The model control technologies for pretreatment of process
wastewaters from existing sources and new sources are described
An indirect discharger is defined as a facility which introduces
pollutants into a publicly owned treatment works (POTW).
Pretreatment standards for existing sources (PSES) are designed
to prevent the discharge of pollutants that pass through,
interfere with, or are otherwise incompatible with the operation
°^upubllcly owned treatment works (POTW). They must be achieved
within three years of promulgation. The Clean Water Act of 1977
requires pretreatment for pollutants that pass through the POTW
in amounts that would violate direct discharger effluent
limitations or interfere with the POTW's treatment process or
chosen sludge disposal method.
^ legislative history of the 1977 Act indicates that
pretreatment standards are to be technology-based, analogous to
the best available technology for removal of toxic pollutants
The general pretreatment regulations, which served as the
framework for the pretreatment regulations, are found at 40 CFR
fno? 4°!' See 43 FR 27736 June 26' 1978' 46 FR 9404 January 28,
1981, and 47 FR 4518 February 1, 1982.
PSNS are to be issued at the same time as NSPS. New indirect
dischargers, like new direct dischargers/ have the opportunity to
incorporate the best available demonstrated technologies. The
Agency considers the same factors in promulgating PSNS as it
considers in promulgating PSES.
Most POTW consist of primary or secondary treatment systems which
are designed to treat domestic wastes. Many of the pollutants
contained in canmaking wastes are not biodegradable and are
therefore ineffectively treated by such systems. Furthermore,
these wastes have been known to interfere with the normal
operations of these systems. Problems associated with the
uncontrolled release of pollutant parameters identified in
canmaking process wastewaters to POTW were discussed in Section
VI. The pollutant-by-pollutant discussions in that Section
covered pass through, interference, and sludge usability.
EPA has generally determined there is pass through of pollutants
if the percent of pollutants removed by a well operated POTW
achieving secondary treatment is less than the percent removed by
the BAT model treatment technology. POTW removals of the
353
-------
priority pollutants found in canmaking wastewater are presented
in Table XII-1 (page 358). The average removal of toxic metals
is about 50 percent. The BAT treatment technology removes more
than 92 percent of toxic metals (see Table X-2, page 332). This
difference in removal effectiveness clearly indicates pass
through of toxic metals will occur unless canmaking wastewaters
are adequately pretreated.
At BAT the toxic metals chromium and zinc are regulated, in
addition to aluminum (see Section X). Aluminum is regulated at
BAT because of its potential adverse affects upon receiving
waters and to control toxic metals that are not specifically
regulated. However, since alum (an aluminum sulfate) is often
added at POTW and since aluminum is not usually regulated for
pretreatment, standards for manganese and copper (which are
alloying constituents in the aluminum strip used in canmaking
processes) are substituted for aluminum in the final regulation.
Thus/ pretreatment standards are established for four metals:
chromium, zinc, copper, and manganese.
Pretreatment standards are also established for fluoride and
phosphorus since both pass through POTW. POTW remove no
fluoride. POTW removal of phosphorus is 10 to 20 percent. The
BAT treatment technology removes more than 80 percent of these
pollutants (see Table X-2).
As described in Section V, the Agency found fourteen specific
toxic organic compounds (collectively referred to as total toxic
organics or TTO) in canmaking wastewaters. The Agency considered
and analyzed whether these pollutants should be specifically
regulated. The removal of toxic organics is about 70 percent by
a secondary POTW (Table XII-1, page 358). This clearly indicates
that pass through of TTO will occur unless canmaking wastewaters
are adequately pretreated. Therefore TTO is regulated.
For PSES and PSNS, the pollutants which interfere with, pass
through or prevent sludge utilization for food crops must be
removed before discharge to the POTW. The model end-of-pipe
treatment technologies for PSES and PSNS are the same as those
for BAT and NSPS (see Figures X-2 and XI-1) and were selected for
the same reasons. The model treatment technology includes
removal of TTO-containing oil and grease by oil skimming,
chemical emulsion breaking, dissolved air flotation, or a
combination of these technologies; chromium reduction where
necessary; and removal of toxic metals and other pollutants by
lime and settle treatment technology.
The proposed PSES and PSNS were based upon reductions in flow, to
reduce the total mass of regulated pollutants discharged. Flow
reduction is retained in the final regulation. The PSES flow is
354
-------
„ J 2 °2 *Canf£ Which 1S identical to the BAT flow and which
n? Je?5?Snn0r the sameKreaSOnS (see Section X). The PSNS flow
63.6 1/1000 cans, which is identical to the NSPS flow and
which was chosen for the same reasons (see Section XI).
Industry Cost and Effluent Reduction of Treatment Options
™P°SSd^and, ,final PSES Options 0, 1, 2, and 3 are parallel to
BPT and BAT Options 1, 2, and 3, respectively. Also, proposed
and final PSNS Options are parallel to the NSPS Options
5SJ"5S5 °f _ phosphorus and manganese
are nonconventional pollutant parameters which pass through POTW
and are therefore regulated.
As previously discussed, manganese is an alloying constituent in
the aluminum strip used in canmaking processes, and its
™i!a^10n should adequately control all of the toxic metals in
canmaking wastewaters and assure the operating effectiveness of
a™ J£n T* fystem. The regulation also requires reporting of
any change to alloys which results in the use of aluminum alloys
in canmaking which contain less than 1.0 percent manganese. This
information will enable the Agency to determine whether changes
in this regulation are warranted. y
355
-------
PRETREATMENT STANDARDS
Mass based limitations are set forth below (Tables XI1-4 and XII-
5 pages 361 and 362). The mass based limitations are the only
method of designating pretreatment standards since the water use
reductions at PSES and PSNS are major features of the treatment
and control system. Only mass-based limits will assure the
implementation of flow reduction and the. consequent reduction of
the quantity of pollutants discharged. Therefore, regulation of
concentrations alone is not adequate.
The derivation of standards is explained'in Section IX. The PSES
flow is equal to the BAT flow (83.9 1/1000 cans) and its
derivation is presented in Section X. For PSNS, the calculation
is the same as NSPS which is presented in Section XI. The PSNS
flow, which is equal to the NSPS flow, is 63.6 1/1000 cans.
The effectiveness of the end-of-pipe treatment technology for the
removal of regulated pollutants is described in Section VII.
Section IX explains the derivations of treatment effectiveness
concentrations for chromium, zinc, fluoride, phosphorus, and oil
and grease (for alternative monitoring), which were used to
establish PSES and PSNS. Sections VII and IX also describe the
Combined Metals Data Base (CMDB) and the statistical tests which
were used to establish that canmaking wastewaters are comparable
to the wastewaters from the categories used to establish the
CMDB, and to the wastewaters of plants in other categories used
to establish treatment effectiveness. For PSES and PSNS,
treatment effectiveness concentrations for manganese and copper
were drawn from the CMDB to reflect properly operated lime and
settle treatment (see Table VII-21, page 230). For manganese and
copper, this transfer of treatment effectiveness data to the
canmaking subcategory is appropriate due to the inadequate
sampling data from within the subcategory and since canmaking
wastewaters have been determined to be comparable to the
categories used in the CMDB.
The removal of toxic organic pollutants by oil skimming from coil
coating, copper forming and aluminum forming plants is presented
in Section VII. Many of the toxic organic pollutants found in
canmaking wastewaters are found in coil coating, copper forming
or aluminum forming and have been shown to be removed by oil
removal. As established in Section VII, the average removal of
organics in aluminum forming by oil skimming is about 97 percent.
This removal rate is used for projecting the effectiveness of the
model oil removal technology in removing TTO in canmaking,
because some of the lubricants from aluminum forming are carried
on aluminum strip into canmaking operations and because the
concentrations of oil' in canmaking and aluminum forming are
similar (see Section IX for details).
356
-------
The achievable TTO concentration for PSES and PSNS was derived as
protocols. Following proposal, the presence of six of
:Sd??<2Sani? P?llutants.was confirmed and the presence of ev
SSJiftS?0 it0"S °r<**nics in- treatable amounts was .established
qualitatively. Following an analysis of this data, the Aaencv
determined that the mean concentration of. the fourteen TTO lS Sot
expected to exceed 2.73 mg/1 in wastewater froS a sinSlI
canmaking plant. The final mean treatment effect ivenlsl
concentration for TTO, therefore, is 0.08 mg/1 eirectiveness
9il removal is the model treatment technology for TTO and is
cSliUfSini: fc5e PSES and PSNS contro1 technologies and
believes ?Kat SL/nneSPSnding benefits and costs' The Agency
mfiJ ?2 2 J good oil and grease removal will allow a plant to
•5SS- ia SS,f?Jal ^°X1C ?rganics limitations. Since monitoring fo?
estaMisS?nny «"f req"ires sophisticated equipment, the Agency is
plrameter^r '..So;. ^ ^easeasa" alternative monWing
hinhJ.1?*' redu^tfons required by PSES and PSNS may result in
higher concentrations of pollutants in wastewaters prior to
end-of-pipe treatment This issue is discussed in Sections X and
2 u F • NSPS, respectively, since the model treatment-
technologies for BAT and NSPS are thesame as those f lr pill Ind
DEMONSTRATION STATUS
M technologies for PSES and PSNS are the
nc?!^ **' ^ **™*^™ ^atus *
357
-------
Table XII-1
KOW REMCWSLS OF THE PRIORITY POLLUTANTS FOUND IN CANMAKING WASTEWATER
Pollutant
Percent Removal by Secondary POIW
11. 1,1,1-Trichlorcethane
13, 1,1-Dichloroethane
15. l,l,2r2-Tetrachloroethane
18. Bis(2-chloroethyl) ether
23. Chloroform
29. 1,1-Dichloroethylene
44. Methylene Chloride
64. Pentachlorophenol
€6. Bis(2-ethylhexyl) phthalate
67. Butyl benzyl phthalate
68. Di-n-butyl phthalate
81. Phenanthrene \
85. Tetrachloroethylene ;
86. Toluene
119. Chromium
120. Copper
124. Nickel
128. Zinc
87
76
89
Not available
61
80
58
52
62
59
48
65
81
90
65
58
19
65
NOTES These data conpiled from Fate of Priority Pollutants in Publicly
n^ad Treatment Works, OS EPA, EPA No. 440/1-80-301, October,
••ftfaot and Determine National Ranpval Credits for Selected
SllutanfaTfoTPublicly Owned Treatment Vtorks, EPA No. 440/82-008,
September, 1982.
358
-------
TABLE XII-2
TOXIC QRGANICS COMPRISING TTO
Pollutant,
11. Iflr
13. 1,1,-Dichloroethane
15. 1,1,2,2-Tetrachloroethane
18. Bis (2-chloroethyl) ether
23. Chlorofonn
29. 1,1-Dichloroethylene
44. Methylene chloride
64. Pentachlorophenol
66. Bis (2-ethylhexyl) phthalate
67. Butyl benzyl phthalate
68. Di-n-butyl phthalate
81. Phenanthrene
85. Tetrachloroethylene
86. Toluene
TOTAL
Mean Raw Waste
At Proposal
(a)
0.561
0.093
0.022
1.55
0.022
0.464
0.016
2.727
Postproposal
Data
(b)
0.561
0.018(c)
0.055
0.066
0.012(d)
0.093
0.022
0.030(d)
0.869
0.228
0.464
0.044
0.018(d)
0.135
2.615
(a) Mean concentrations of toxic organics found above quantifiable limits
O0.010 mg/1) in raw wastewaters sampled by EPA at proposal (See Table
V-ll).
(b) Mean concentrations of toxic organics including postproposal data.
(c) Toxic organics found above quantifiable limits (>0.010 mg/1) in treated
effluent samples analyzed and submitted by Reynolds Aluminum Company
(See Table V-21).
(d) Toxic organics found above quantifiable limits O0.010 mg/1) in
treated wastewaters sampled by EPA after proposal (See Table V-19).
359
-------
2
M
X
(0
3
CO
00
i
c,
r CM vo <*> * o
9. -: ~. ^ *. x
vo "in^ing^s ^^ ^gg
vo vo vo vo vo
M^ ^41 • ^J' f^ ^NJ f^ OJ * Cft ^" ^^* On
/M% ^B« «^ ^^> [^ tf^ VO O^ VO ^^ f^ OM
•••i ^«« ^C) ^^ C^ ^O ^^ ' VO
r"
o\ =^^ on in c
^^" C*JP* ^"t™ ^" ^™ VO
,-- r- r— in w
vo vo vo vo vo
OOOOOCAO^J'^'OOf—OOOO J£>22
inoooo\«--'— r~.vo'x'x '"m .*>> on o^
,_. vo^-csro com
o r- oo vo o in% oo •* in CM^O'O vo jo^o"®
§ « en in oj o ^ •* vo CM^- ^ ^ aj o ~ ~
S »RRS53S»a2ffl*S5 »S2
e> i— in <«p ^J1 on
-------
TABLE XII-4
PRETREATMENT STANDARDS FOR EXISTING SOURCES
CANMAKING SUBCATEGORY
Pollutant or
Pollutant Property
PSES
MaxIiUi for
any one day
Maximum for
monthly average
g (lbs)/100Q.QQQ cans manufactured
*Chromium
IF
• "S
36
go
tn
I
\
I
ic
in
»
alternate
monitoring)
78.00
*f
-85
(3.699)
(7'584)
(0.059)
1006.80
1636.05
12.59
*Regulated Pollutant
,~
1
(2.220)
3607
(0.028)
361
-------
TABLE XI1-5
pfETREATMENT STANDARDS FOR NEW SOURCES
CANMAKING SUBCATEGORY
Pollutant or
Pollutant Property
Maximum for
.any one day
PSNS
Maximum for
monthly average
a (
•Chromium
•Copper
Lead
Nickel
•Zinc
Aluminum
•Fluoride
Iron
•Manganese
•Phosphorus
•Oil "S^ Grease
alternate
monitoring)
*TTO
lbs)/1 . 000 r 000 cans manufactured
27.98
120.84
26.71
122*11
92.86
408.95
3784.20
76.32
43.25
1 062 . 1 2
(for
1272.00
2607.60
20.35
(0.062)
(0.266)
(0.059)
(0.269)
(0.205)
(0.902)
(8.343)
(0.168)
(0.095)
(2.342)
(2.804)
(5.749)
(0.045)
11.45
63.60
12.72
80.77
38.80
203.52
1679.04
38.80
18.44
434.39
763.20
1240.20
9.54
(0.025)
(0.140)
(0.028)
(0.178)
(0.086)
(0.449)
(3.702)
(0.086)
(0.041)
(0.958)
(1.683)
(2.734)
(0.021 )
•Regulated Pollutant
362
-------
SECTION XIII
BEST CONVENTIONAL POLLUTANT CONTROL TECHNOLOGY
The 1977 Amendments added Section 30Ub)(2)fFl *•« «•»,« A *.
BCT is not an additional limitation but replaces BAT for *-h«
out the
the
=
363
-------
-------
SECTION XIV
ACKNOWLEDGMENTS
This document has been prepared by the staff of the Effluent
Guidelines Division with assistance from technical contractors,
other EPA offices and other persons outside of EPA. This Section
is intended to acknowledge the contribution of the persons who
have contributed to the development of this report.
The initial effort on this project was carried out by Sverdrup &
Parcel and Associates under Contract No. 68-01-4408; Hamilton
Standard Division of United Technologies, under Contract No.
68-01-4668, assisted in some sampling and analysis.
The field sampling programs were conducted under the leadership
of Garry Aronberg of Sverdrup & Parcel assisted by Donald
Washington, Project Manager, Claudia O'Leary, Anthony Tawa,
Charles Amelotti, and Jeff Carlton. Hamilton Standard's effort
was managed by Daniel J. Lizdas and Robert Blaser and Richard
Kearns.
In preparation of this document, the Agency was assisted by
Versar Inc., under contract 68-01-6469, and two subcontractors to
Versar, Whitescarver Associates, Inc., and JFA, Inc. Versar's
effort was managed by Lee McCandless and Pamela Hillis with
contributions from Jean Moore and others. John Whitescarver
Robert Hardy, Robert Smith, V. Ramona Wilson, Jon Clarke, and
Lisa Taschek of Whitescarver Associates assisted in the
preparation of the final development document. JFA's efforts
were managed by Geoffrey Grubbs, with substantial assistance from
Thomas Wall.
Ellen Siegler of the Office of General Counsel provided legal
advice to the project. Josette Bailey was the economic project
officer for the project. Henry Kahn and Barnes Johnson provided
statistical analysis and assistance for the project. Alexandra
Tarnay provided environmental evaluations and word processing was
provided by Pearl Smith, Carol Swann, and Glenda Nesby.
Technical direction and supervision of the project was provided
by Ernst P. Hall. The technical project officer was Mary L.
Belefski, with assistance from V. Ramona Wilson.
Finally, appreciation is expressed to the Can Manufacturers
Institute (CMI), the United States Brewers Association, UJSBA) and
the participating can manufacturing companies7 for their
assistance and technical advice. -'
365
-------
-------
SECTION XV
REFERENCES
'• movinri.^^r ss 'i^&iS-n*1^1?1- .«* —
Finishing Abstracts. Third
^
aa*aaa»fti jsjii"'^' w
University Plaza Hackensack, New Jersey 90601? '
iitffefiflsfl. ^"^ Saatinas, sdited by Dr. H. W.
» • ***B^W«I- AIM JL \»/j_ 11 cr i_« i pint^ninrYa "I~IW>
-------
Mr. Michael Quinn, Mr. Walter Cavanaugh, Mr. James Maurar,
Mr. John Scalise . * . .,
Division o£ Oxy Metals Industries
P. 0. Box 201
Detroit, MI 45220
Amchem Corporation:
Lester Steinbrecker
Metals Research Division
Brookside Avenue
Ambler, PA 19002
Diamond Shamrock
Metal Coatings Division
p. O. Box 127
Chardon, OH 44024 .
Wyandotte Chemical:
Mr. Alexander W. Kennedy
Mr. Gary Van Ve Streek
Wyandotte, MI
"•
1977.
11. Handbook of Chemistrv., Lange, Norbert, Adolph, McGraw Hill,
New York 1973.
,2. Dangerous ProEertles of Industrial Materials, Sax N. Irving,
Van Nostrand Reinhold Co. New York.
13- «*• '^^
»•
15. industrial Pollution, Sax, H. Irving, Van Nostrand iRelnhold
Co., New York 1974.
"• ""• 'sssss '•
"•
368
-------
18. "Treatability of the Organic Priority Pollutants - Part C -
ime3 dayav9> TreatedEffiuents
-
TreatedEffiuents Concentration
19. Water Quality Criteria Second Edition, edited by Jack Edward
McKee and Harold W. Wolf, 1963 The Resources Agency of
No! 3°A ' r Quality Control Board, Publication
20" Srl-C?ndens^d Chemigal^Dictionary, Ninth Edition, Revised by
24 '
21. Wastewater Treatment Technology. James W. Patterson.
22 ' B^
23. "Development Document For Proposed Existing Source
J-treat^t^the Buffalo Municipal
t I
25 me °f Industrial Wastes", Seminar Handout, U.S.
.
27 Wate?Ctfo?fu??nSer °f ^54° Bi?lo9ical Sewage Treatment",
||i§|4l Pollution Control Federation Journal . February 1963 p
28. Wastewater Engineering. 2nd edition, Metcalf and Eddy.
29' NlwJI^o?kL,TT9720l0qY/ L'W< °°dd/ et> al>/ Bar"eS and Noble'
30 '
•fSSJVSSinVH2 Condensation of 4-aminoantipyrene
HW«it-«niV.» ? S . Hydroxybenzene - II. Influence of
Hydronium lon^ Concentration on Absorbtivity," Samuel D.
Faust and Edward W. Mikulewicz, Water Research 1967
Pergannon Press, Great Britain ~ ' '
31' "fj^o^s Influencing^the Condensation of 4-aminoantipyrene
with derivatives of Hydroxylbenzene - I. a Critique," Samuel
369
-------
D. Faust and Edward W. Mikulewicz, Water Research, 1967,
Pergannon Press, Great Britain
32 Scott, Murray C., "Sulfext, - A New Process Technology for
lemoval of Hea^y Metals from Waste Streams " presented at
1977 purdue Industrial Waste Conference, May 10, 11, and 12,
1977.
33 "SulfexT. Heavy Metals Waste Treatment Process," Technical
Bulletin, Vol. XII, code 4413.2002 (Permutit®) July, 1977.
34 Scott, Murray C., "Treatment of Plating Effluent by Sulfide
Process," Products Finishing, August, 1978.
35. Lonouette, Kenneth H., "Heavy Metals Removal," Chemical
Engineering, October 17, pp. 73-80.
36. Curry, Nolan A., "Philosophy and Methodology of Metallic
Waste Treatment," 27th Industrial Waste Conference.
37. Patterson, James W., Allen, Herbert E. and Seal a, John J.,
"Carbonate Precipitation for Heavy Metals Pollutants,
journal of Water Pollution Control Federation, December,
1977 pp. 2397-2410.
38. Bellack, Ervin, "Arsenic Removal from Potable Water,"
Journal American Water Works Association, July, 1971.
39. Robinson, A. K. "Sulfide -vs- Hydroxide Precipitation of
Heavy Metals from Industrial Wastewater," Presented at
EPA/AES First Annual Conference on Advanced Pollution
Control for the Metal Finishing Industry, January 17-19,
1978.
40 Sora Thomas J., "Treatment Technology to meet the Interim
Primary Drinking Water regulations for Inorganics," Journal
American Water Works Association, February, 1978, pp. 105-
112.
41. Strier, Murray P., "Suggestions for Setting Pre^e|t™ent
Limits for Heavy Metals and Further Studies of POT W s
meSorlnduS to Carl J. Schafer, Office of Quality Review,
U.S. E.P.A., April 21, 1977.
42. Rohrer, Kenneth L., "Chemical Precipitants for Lead Bearing
Wastewater s," Industrial Water Engineering, June/July, 1975.
- -
370
-------
Hydroxides," International Journal of Air and Water
Pollution. Vol. 8, 1964/pp. 537-556. " ~~
44. Bhattacharyya; O., Jumawan, Jr., A.B., and Grieves, R.B.
Separation of Toxic Heavy Metals by Sulfide Precipitation,"
Separation Science and Technology. 14(5), 1979, pp. 441-452.
45. Patterson, James W., "Carbonate Precipitation Treatment for
Cadmium arid Lead," presented at WWEMA Industrial Pollutant
Conference, April 13, 1978.
46. "An Investigation of Techniques for Removal of Cyanide from
Electroplating Wastes," Battelle Columbus Laboratories,
Industrial Pollution Control Section, November, 1971.
47. Patterson, James W. and Minear, Roger A., "Wastewater
Treatment Technology," 2nd edition (State of Illinois,
Institute for Environmental Quality) January, 1973.
48, Chamberlin, N.S. and Snyder, Jr., H.B., "Technology of
Treating Plating Waste," 10th Industrial Waste Conference.
49. Hayes, Thomas D. and Theis, Thomas L., "The Distribution of
Heavy Metals in Anaerobic Digestion," Journal of Water
Pollution Control Federation. January, 1978. pp. 61 -72:.
50. Chen, K.Y., Young, C.S., Jan, T.K. and Rohatgi, N., "Trace
Metals in Wastewater Effluent," Journal of Water Pollution
Control Federation. Vol. 46, No. 12, December^1974^—ppT
2663—2675.
51. Neufeld, Ronald D., Gutierrez, Jorge and Novak, Richard A.,
A Kinetic Model and Equilibrium Relationship for Metal
Accumulation," Journal of Water Pollution Control
Federation. March. 1977. pp. 489-498. '.— ~~
52. Stover, R.C., Sommers, L.E. and Silviera, D.J., "Evaluation
of Metals in Wastewater Sludge," Journal of Water Pollution
Control Federation. Vol. 48, No. 9, September;1976, pp.
2165—2175.
53. Neufeld, Howard D. and Hermann, Edward R., "Heavy Metal
Removal by Activated Sludge," Journal of Water Pollution
Control Federation. Vol. 47, No. 2, Februarys 1975^—ppT
310-329. > ' ff
54.; Schroder, Henry A. and Mitchener, Marian, "Toxic Effects of
Trace Elements on the Reproduction of Mice and Rats,"
Archives of Environmental Health. Vol. 23, August, 1971, pp.
102—106.
371
-------
55. Venugopal, B. and Luckey, T.D., "Metal Toxicity in Mammals"
(Plenum Press, New York, N.Y.), 1978.
56. Poison, C.J. and Tattergall, R.N., "Clinical Toxicology,"
(J.B. Lipinocott Company), 1976.
57. Hall, Ernst P. and Barnes, Devereaux, "Treatment of
Electroplating Rinse Waters and Effluent Solutions,"
presented to the American Institute of Chemical Engineers,
Miami Beach, Fl., November 12, 1978.
58. Mytelka, Alan I., Czachor, Joseph S., Guggirio, William B.
and Golub, Howard, "Heavy Metals in Wastewater and Treatment
Plant Effluents," Journal of Water Pollution control
Federation, Vol. 45, No. 9, September, 1973, pp. 1859-1884.
59. Davis, III, James A., and Jacknow, Joel, "Heavy Metals in
Wastewater in Three Urban Areas, "Journal of Water Pollution
Control Federation,. September, 1975, pp. 2292-2297.
60. Klein, Larry A., Lang, Martin, Nash, Norman and Kirschner,
Seymour L., "Sources of Metals in New York City Wastewater,"
Journal of Water Pollution Control Federation, Vol. 46, No.
12, December, 1974, pp. 2653-2662.
61. Brown, H.G., Hensley, C.P., McKinney, G.L, arid Robinson,
J.L., "Efficiency of Heavy Metals Removal in Municipal
Sewage Treatment Plants," Environmental Letters, 5 (2), .
1973, pp. 103-114.
62. Ghosh, Mriganka M. and Zugger, Paul D., "Toxic Effects of
Mercury on the Activated Sludge Process," Journal of. Water
Pollution Control Federation, Vol. 45, No. 3, March, 1973,
pp. 424-433.
63. Mowat, Anne, "Measurement of Metal Toxicity by Biochemical
Oxygen Demand," Journal of Water Pollution Control
Federation, Vol. 48, No. 5, May, 1976, pp. 853-866.
64. Oliver, Barry G. and Cosgrove, Ernest G., "The Efficiency of
Heavy Metal Removal by a Conventional Activated Sludge
Treatment Plant," Water Research, Vo. 8, 1074, pp. 869-874.
65. "Ambient Water Quality Criteria for Chlorinated Ethanes",
PB81-117400, Criteria and Standards Division, Office of
Water Regulations and Standards, U.S. EPA.
66. "Ambient Water Quality Criteria for Chloroalkylethers,"
PB81-117418, Criteria and Standards Division, Office of
Water Regulations and Standards, U.S. EPA.
372
-------
67. "Ambient Water Quality Criteria for Dichloroethylenes "
PB81-117525, Criteria and Standards Division, Office of
Water Regulations and Standards, U.S. EPA.
68. "Ambient Water Quality Criteria for Halomethanes," PB81-
117624, Criteria and Standards Division, Office of Water
Regulations and Standards,, U.S. EPA..
69. "Ambient Water Quality Criteria for Phthalate Esters "
PB81-I17780 Criteria and Standards Division, Office of Water
Regulations, and Standards, U.S. EPA.
70. "Ambient Water Quality Criteria for Toluene", PB81-117855,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
71. "Ambient Water Quality Criteria for Arsenic," PB81-117327,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
72. "Ambient Water Quality Criteria for ..Cadmium," PB81-117368,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
73. "Ambient .Water Quality Criteria for Chromium," PB81-117467,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
74. "Ambient Water Quality Criteria for Copper," PB81-117475,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
75. "Ambient Water Quality Criteria for Cyanide/1 PB81-117483,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
76. "Ambient Water Quality Criteria for Lead," PB81-117681,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
77i "Ambient Water Quality Criteria for Mercury," Criteria and
Standards Division, Office of Water Regulations and
; Standards, U.S. EPA
78. "Ambient Water Quality Criteria for Nickel," PB81-117715,
Criteria and Standards Division, Office of Water Regulations
and Standards U.S. EPA.
373
-------
79. "Ambient Water Quality Criteria for Zinc," PB81-117897,
Criteria and Standards Division, Office of Water Regulations
and Standards, U.S. EPA.
80. Treatability Manual, U.S. Environmental Protection Agency,
Officeof Research and Development, Washington, D.C. July
1980, EPA - 600/8-80-042a,b,c,d,e.
81. Electroplating Engineering Handbook, edited by H. Kenneth
Graham, Van Nostrand Reinhold Company, New York, 1971.
82. Can Manufacturers Institute, "Directory - Cans Manufactured
for Sale," 1982.
83. Can Manufacturers Institute, "Metal Can Shipments Report,"
1980.
84. Can Manufacturers Institute, "Metal Can Shipments Report,"
1981.
85. Church, Fred L. "Can Equipment Sales Ride Wave of Plant
Expansions." Modern Metals, April, 1978, pp. 32-40.
86. "Computer Control Increases Productivity, Cuts Downtime at
Canmaking plant." The BREWERS DIGEST, July, 1975, pp. 36-38.
87. "Deep-drawn Oval Fish Cans." Iron and Steel Engineer, July,
1974, p. 55.
88. "Design Data." Machine Design, February 14, 1974, pp. 148-
150.
89. "Experts Tell What's New in Forming." American Machinist,
April 1, 1975, pp. 45-46.
90. "Industry Environmental Activities." The BREWERS DIGEST,
August, 1976, p. 14.
91. Knepp, J. E. and L. B. Sargent, Jr. "Lubricants for Drawing
and Ironing Aluminum Alloy Beverage Cans." Lubrication
Engineering, April, 1978, pp. 196-201.
92. Kuhner, John G. "Pearl's Total Aluminum Can Program." The
BREWERS DIGEST, January, 1976, pp. 45-50.
93. "Lone Star Adopts Ultra-Lightweight Seamless Steel Can." The
BREWERS DIGEST, May, 1975, pp. 46-47.
94. Lubrication, published by Texaco, Inc. N.Y., N.Y. Volume 61,
April-June 1975, pp. 17-18.
374
-------
95. Lund, H., editor, Industrial Pollution Control Handbook,
McGraw-Hill 1971, pp. 612-613. ~~™~ ~
96. Church, Fred L., "Aluminum's Next Target: Cost-Competative
Food Cans," Modern Metals. Vol. 32, May 1976, pp. 81-87.
97. American Society for Metals, Metals Handbook. 8th Edition,
98. Maeder, Edward G. "The D&I Can: How & Why it Does More With
Less Metal." Modern Metals. August, 1975, pp. 55-62.
99. Mastrovich, J. D. "Aluminum Can Manufacture." Lubrication
Vol. 61, April-June, 1975,pp. 17-36.
100. Mathis, Jerry N. "We See a future For Steel Two-Piece
Cans.." advertisement, The BREWERS DIGEST. January, 1977,
PP .13."
101. Mungovan, James. "New Can Plant on Target: 2 Million
Containers a Day." Modern Metals. Vol. 33, July, 1977, pp.
27—36...
102. "Olympia's Plans for Lone Star." The BREWERS DIGEST, Julv.
1977, pp. 20-23. '
* . ..-.•... .
103. "Schmidt's Christens New $7 Million Packaging Facility."
Food Engineering. October, 1977, pp. 47-49.
104. Spruance, Frank Palin, Jr. U.S. Patent 2,438,877, September
6, 1945.
105. Sullivan, Barry C. "Lone Star Turns It Around With
Returnables, Youth Emphasis." The BREWERS DIGEST. May, 1976,
pp. 28-30.
375
-------
-------
SECTION XVI
GLOSSARY
Accumulation - in reference to biological systems, is the
concentration which collects in a tissue or organism which
does not disappear with time. wnicn
Acidulated Rinse - See Sealing Rinse
Act - The Federal Water Pollution Control Act (P.L. 92-500) as
amended by the Clean Water Act of 1977 (P.L. 95-217)?
Activator - A material that enhances the chemical or physical
change when treating the metal surface. Y»^«*A
Adsorption - The adhesion of an extremely thin layer of molecules
Sfi-S g?? £r,-ilquid t(? the surface of the solid or liquid
with which they are in contact.
Agency - The U.S. Environmental Protection Agency.
Alc*icfde 7 Chemical used in the control of phytoplankton (algae)
i n Wei tGir •
Alkalinity - The quantitative capacity of aqueous media to react
with hydrogen ions.
Alumi""m .i§£is Material - Means aluminum and aluminum alloys
which are processed in canmaking.
Anionic Surfactant - An ionic type of surface-active substance
that has been widely used in cleaning products. The hydro-
philic group of these surfactants carries a negative charge
in the washing solution. y
SSSBSI.™." An process of controlled CBISMIIKKHUHP
prodiit-ing a hard, transparent oxide up to several
mils in thickness.
Area Processed - See Processed Area.
;g!?.j;nq . * The Process of cleaning a filter or ion exchange
column by reversing the flow of water.
377
-------
Baffles - Deflector Vanes, guides, grids, gratings, or similar
* "devices constructed or placed in flowing water or sewage to
(1) check or effect a more uniform distribution of
velocities; (2) absorb energy; (3) divert, guide, or agitate
the liquids; or (4) check eddy currents.
Basis Material or Metal - That substance of which the cans are
*""""' "made and that receives the coating and the treatments in
preparation of coating.
BAT - The best available technology economically achievable under
~Section 304(b)(2)(B) of the Act
BCT - The best conventional pollutant control technology, under
Section 304(b)<4) of the Act
BDT - The best available demonstrated control technology
processes, operating methods, or other alternatives,
including where practicable, a standard permitting no
discharge of pollutants under Section 306(a)(l) of the Act.
Biochemical Oxygen Demand (BOD) - (1) The quantity of oxygen
*— required for the biological and chemical oxidation of
waterborne substances under conditions of test used in the
biochemical oxidation of organic matter in a specified time,
at a specified temperature, and under specified conditions.
(2) Standard test used in assessing wastewater strength.
Biodegradable - The part of organic matter which can be oxidized
—' By bi©processes, e.g., biodegradable detergents, food
wastes, animal manure, etc.
Biological Wastewater Treatment - Forms of wastewater treatment
in which bacteria or biochemical action is intensified to
stabilize, oxidize, and nitrify the unstable organic matter
present.
BMP - Best management practices under Section 304(e) of the Act
Bodvmaker - The machine for drawing, or drawing and ironing
two-piece can bodies.
BPT - The best practicable control technology currently available
under Section 304(b)(l) of the Act.
Buffer - Any of certain combinations of chemicals used to
stabilize the pH values or alkalinities of solutions.
Cake - The material resulting from drying or dewatering sludge.
378
-------
Calibration - The determination, checking/ or rectifying of the
graduation of any instrument giving quantitative
measurements;
Canmakinq - The manufacturing operations used to produce various
shaped metal containers subsequently used for storing foods.
beverages, and.other products.
Captive Operation - A manufacturing operation carried out in a
facility to support other manufacturing, fabrication, or
assembly operations.
Carcinogenic - Referring to the ability of a substance to produce
or incite cancer.
Central Treatment Facility - Treatment plant which co-treats
process wastewaters from more than one manufacturing
operation or cotreats process wastewaters with noncontact
cooling water, or with nonprocess wastewaters, miscellaneous
runoff, etc.).
Chemical Coagulation - The destabilization and initial
aggregation of colloidal and finely divided suspended matter
by the addition of a floe-forming chemical. The amount of
oxygen expressed in parts per million consumed under
specific conditions in the oxidation of the organic and
oxidizable inorganic matter contained in an industrial
wastewater corrected for the influence of chlorides.
Chemical Oxygen Demand (COD) - (i) A test based on the fact that
all organic compounds, with few exceptions, can be oxidized
to carbon dioxide and water by the action of strong
oxidizing agents under acid conditions. Organic matter is
converted to carbon dioxide and water regardless of the
biological assimilability of the substances. One of the
chief limitations is its ability to differentiate between
biologically oxidizable and biologically inert organic
matter. The major advantage of this test is the short time
required for evaluation (2 hrs). (2) The amount of oxygen
required for the chemical oxidation of organics in a liquid.
Chemical Oxidation - A wastewater treatment in which a pollutant
is oxidized.
Chemical Precipitation - Precipitation induced by addition of
chemicals.
Chlbrination - The application of chlorine to water or wastewater
generally for the purpose of disinfection, but frequently
for accomplishing other biological or chemical results.
379
-------
Chromate Conversion Coating - A process whereby an aqueous
acidified chromate solution consisting mostly of chromic
acid and water soluble salts of chromic acid together with
various catalysts or activators is applied to the can body.
Chromium Process Controller - A device used to maintain a
" desirable and constant hexavalent chromium concentration.
Clarification - The removal of suspended solids from wastewater.
Cleaning - The process of removing contaminants from the surface
of a coil. .
Clean Water Act - The Federal Water Pollution Control Act
Amendments" of 1972 (33 U.S.C. 1251 et seg.), as amended by
the Clean Water Act of 1977 (Public Law 95-217)
Colloids - A finely divided dispersion of one material called the
" ~*dTspersed phase" (solid) in another material which ^ is
called the "dispersion medium" (liquid). Normally
negatively charged.
compatible Pollutant - A specific substance in a waste stream
which alone can create a potential pollution problem, yet is
" used to the advantage of a certain treatment process when
combined with other wastes.
Composite - A combination of individual samples of water or
wastewater taken at selected intervals and streams and mixed
in proportion to flow or time to minimize the effect of the
variability of an individual sample.
Concentration Factor - Refers to the biological concentration
factorwhichis the ratio of the concentration within the
tissue or organism to the concentration outside the tissue
or organism.
Concentration, Hydrogen Ion - The weight of hydrogen ions in
~~ grams per liter of solution. Commonly expressed as the pH
value that represents the logarithm of the reciprocal of the
hydrogen ion concentration.
Contamination - A general term signifying the introduction of
~ microorganisms, chemicals, wastes or sewage which renders
the material or solution unfit for its intended use.
Contractor Removal - The disposal of oils, spent solutions, or
"sludge by means of a scavenger service.
380
-------
JChe process of applying a chromate,
phosphate, complex oxide or other similar protective coating
to a metal surface.
Cooling Tower - A device used to cool water used in the manufac-
turing processes before returning the water for reuse.
x
Cupping - Process whereby a flat sheet of metal is formed into a
cup by means of a die punch operation ('a cupper).
Degreasing - The process of removing grease and oil from the sur-
face of the material.
Deionized Water - Water from which dissolved impurities (in the
form of free ions) have been removed to reduce its
electrical conducting properties and the potential for
contamination of the manufacturing process.
Dewaterinq - A process whereby water is removed from sludge.
Die - Part on a machine that, punches shaped holes in, cuts, or
forms sheet metal, cardboard, or other stock.
Direct Discharger - A facility which discharges or may discharge
pollutants into waters of the United States.
Dissolved Solids - Theoretically the anhydrous residues of the
dissolved constituents in water. Actually the term is
defined by the method used in determination. In water .and
wastewater treatment, the Standard Methods tests are used.
Draqout - The solution that adheres to the can and is carried
past the edge of the treatment tank.
Drawing - A process where a sheet of metal is pushed into a mold
or die by a solid piece of metal (punch), thus flowing over
the punch to form a cup.
Draw-redraw - Process in which a second drawing step follows an
initial drawing to form a deeper cup.
Drying Beds - Areas for dewatering of sludge by evaporation and
seepage.
Dump - The discharge of process waters not usually discharged for
maintenance, depletion of chemicals, etc.
Effluent - The wastewaters which are\ discharged to surface
waters, directly or indirectly. \
381
-------
Emergency Procedures - The various special procedures necessary
to protect the environment from wastewater treatment plant
failures due to power outages, chemical spills, equipment
failures, major storms and floods, etc.
Emulsion Breaking - Decreasing the stability of„ dispersion of one
liquid in another.
End-of-Pipe Treatment - The reduction and/or removal of
pollutants by chemical treatment just prior to actual
discharge. ..
Equalization - The process whereby waste streams from different
sources varying in pH, chemical consitutents, and flow rates
are collected in a common container. The effluent stream
from this equalization tank will have a fairly constant flow
and pH level, and will contain a homogeneous chemical
mixture.
Extrusion - Process of shaping by forcing basis material through
a die.
Feeder, Chemical - A mechanical device for applying chemicals to
water and sewage at a rate controlled manually or auto-
matically by the rate of flow.
Flanging - The forming of a protruding rim or collar on the end
of the can body to allow attachment of the end.
Float Gauge - A device for measuring the elevation of the surface
of a liquid, the actuating element of which is a buoyant
float that rests on the surface of the liquid and rises or
falls with it. The elevation of the surface is measured by
a chain or tape attached to the float.
Floe - A very fine, fluffy mass formed by the aggregation of fine
suspended particles.
Flocculatbr - An apparatus designed for the formation of floe in
water or sewage.
Flocculation - In water and wastewater treatment, the agglomera-
tion of colloidal and finely divided suspended matter after
coagulation by gentle stirring by either mechanical or
hydraulic means. In biological wastewater treatment where
coagulation is not used, agglomeration may be accomplished
biologically.
382
-------
Flow-Proportioned Sample - A sampled stream whose pollutants are
apportioned to contributing streams in proportion to the
flow rates of the contributing streams.
Grab Sample - A single sample of wastewater taken at neither set
time nor flow.
Grease ~ .In .wastewater,. a group of substances including fats
waxes, free fatty acids, calcium and magnesium soaps!
mineral oil, and certain other nonfatty materials. The tvoe
of solvent and method used for extraction should be stated
for quantification.
Hardness - A characteristic of water, imparted by salts of cal-
cium, magnesium, and iron such as bicarbonates, carbonates,
sulfates, chlorides, and nitrates that cause curdling of
soap, deposition of scale in boilers, damage in some
industrial processes, and sometimes objectionable taste. It
may be determined by a standard laboratory procedure or
computed from the amounts of calcium and magnesium as well
as iron, aluminum, manganese, barium, strontium, and zinc,
and is expressed as equivalent calcium carbonate.
Heavy Metals - A general name-given to the ions of metallic ele-
ments such as copper, zinc, chromium, and nickel.
Holding Tank - A reservoir to contain preparation materials so as
to be ready for immediate service.
Indirect Discharger - A facility which introduces or may
introduce pollutants into a publicly owned treatment works.
Industrial Wastes - The wastes used directly or indirectly in
industrial processes as distinct from domestic or sanitarv
wastes. . *
In-Process Control Technology - The regulation and conservation
of chemicals and rinse water throughout the operations as
opposed to end-of-pipe treatment.
Ion—Exchange - A reversible chemical reaction between a solid
(ion exchanger) and a fluid (usually a water solution) by
means of which ions may be interchanged from one substance
to another. The superficial physical structure of the solid
is not affected.
Ironing - A process where the side walls of a drawn cup are
pressed against the punch, making them thinner and longer,
and creating a deeper can of larger volume.
383
-------
Lagoon - A man-made pond or lake for holding wastewater for the
——removal of suspended solids. Lagoons are also used as
retention ponds.
Landfill - An approved site for dumping of waste solids.
Lime - Any of a family of chemicals consisting essentially of
* calcium hydroxide made from limestone ..(calcite).
Limiting Orifice - A device that limits flow by constriction to a
relatively small area. A constant flow can be obtained over
a wide range of upstream pressures.
Lubricant - A substance such as oil, grease, etc., used for
lessening friction.
Make-Up Water - Total amount of water used by process.
Mandrel - A shaft or bar the end of which is inserted into a
workpiece to hold it during machining.
Milligrams Per Liter (mq/1) - This is a weight per volume desig-
nation used in water and wastewater analysis.
Mutaqenic - Referring to the ability of a substance to increase
the frequency or extent of mutation.
National Pollutant Discharge Elimination System (NPDES) - The
federal mechanism for regulating discharge to surface waters
by means of permits. A National Pollutant Discharge
Elimination System permit issued under Section 402 of the
Act.
Necking - Forming of a narrower portion at the top of a can body.
Neutralization - Chemical addition of either acid or base to a
solution such that the pH is adjusted to approximately 7.
Noncontact Cooling Water - Water used for cooling which does not
come into direct contact with any raw material, intermediate
product, waste product or finished product.
Nonionic Surfactant - A general family of surfactants so called
becausein solution the entire molecule remains associated.
Nonionic molecules orient themselves at surfaces not by an
electrical charge, but through separate grease-solubilizing
and water-soluble groups within the molecule.
- National Pollutant Discharge Elimination System.
384
-------
NSPS - New source performance standards under Section 306 of the
Act.
Orthophosphate - An acid or salt containing phosphorus as PO4.
Outfall - The point or location where sewage or drainage
discharges from a sewer, drain, or conduit.
Paint - A liquid composition of plastic resins, pigments and sol-
vents which is converted to a solid film after application
as a thin layer by a drying or heat curing process step.
Painted Area - (Expressed in terms of square meters). The
dimensional area that receives an enamel, plastic, vinvl or
laminated coating.
Palletizing - The placing of finished cans into a portable
Storage container prior to their being filled.
Parshall Flume - A calibrated device developed by Parshall for
measuring the flow of liquid in. an' open conduit. It
consists essentially of a contracting length, a throat, and
an expanding length. At the throat is a sill over which the
flow passes as critical depth. The upper and lower heads
are each measured at a definite distance from the sill. The
lower head cannot be measured unless the sill is submerged
more than about 67 percent.
EH - The negative of the logarithm of the hydrogen ion concen-
tration.
pJi Ad-just - A means of maintaining the optimum pH through the use
of chemical additives.
Phosphate Coating - In canmaking the process of forming a
conversion coat on aluminum by spraying a hot solution of
phosphate containing titanium or zirconium.
Pollutant - The term "pollutant" means dredged spoil, solid
wastes, incinerator residue, sewage, garbage, sewage sludge,
munitions, chemical wastes, biological materials,
radioactive materials, heat, wrecked or discarded equipment,
rock, sand, cellar dirt and industrial, municipal and
agricultural waste discharged into water.
Pollutant Parameters - The characteristics or constituents of a
waste stream which may alter the chemical, physical,
biological, or radiological integrity of water.
385
-------
Polvelectrolvtes - Used as a coagulant or a coagulant aid in
water—and" wastewater treatment. They are synthetic or
natural polymers containing ionic constituents. They may be
cationic, anionic, or nonionic.
POTW - Publicly Owned Treatment Works.
Prechlorination - (1) Chlorination of water prior to filtration.
(2) Chlorination of sewage prior to treatment.
Precipitate - The solid particles formed from a liquid solution
due to the saturation of the solid in the solution having
been achieved.
Precipitation. Chemical - Precipitation induced by addition of
chemicals.
Pretreatment - Any wastewater treatment process used to reduce
pollution load partially before the wastewater is introduced
into a main sewer system or delivered to a treatment plant
for substantial reduction of the pollution load.
Printing - The technique of rolling a design on a painted strip.
Priority Pollutant - The 129 specific pollutants established by
the—EPA from the 65 pollutants and classes of pollutants as
outlined in the consent decree of June 8, 1976.
Process Water - Any water which during manufacturing or
processing, comes into direct contact with or results from
the production or use of any raw materials, intermediate
product, finished product, by-product, or waste product.
PSES - Pretreatment standards for existing sources of indirect
discharges under Section 307(b) of the Act.
Publicly Owned Treatment Works (POTW) - A central treatment works
serving a municipality.
Raw Wastewater - Plant water prior to any treatment or use.
RCRA - Resource Conservation and Recovery Act (PL 94-580) of
1976, Amendments to Solid Waste Disposal Act.
Recirculated Water - Process water which is returned as process
—water in the same or in a different process step.
Rectangular Weir - A weir having a notch that is rectangular in
shape.
386
-------
Recycled—Water - Process water which is returned to the same
process after treatment.
Reduction Practices - (1) Wastewater reduction practices can mean
the reduction of water use to lower the volume of wastewater
requiring treatment and (2) the use of chemical reduction to
lower the valance state of a specific wastewater pollutant.
Reduction - The opposite of oxidation treatment wherein a
reductant (chemical) is used to lower the valence state of a
pollutant to a less toxic form e.g., the use of SO2 to
reduce hexavalent chromium to trivalent chromium in" an
acidic solution.
Retention—Time - The retention time is equal to the volume of a
tank divided by the flow rate of liquids into or out of the
uclflJC •
Rinse - Water for removal of dragout by dipping, spraying,
•fogging, etc.
Sanitary Sewer - A sewer that carries water or wastewater from
residences, commercial buildings, industrial plants, and
institutions together with minor quantities of ground,
storm, and surface waters that are not admitted
intentionally.
Sealing Rinse - The final rinse in the conversion coating process
which contains a slight concentration of chromic acid.
Seaming - in canmaking the joining of two edges of a rolled metal
blank to form a cylinder and the joining of ends or tops to
can bodies.
Seamless - In canmaking refers to can bodies formed without side
seams. Cans are formed by drawing of flat sheet metal into
a cupped shape.
Secondary—Waste Water Treatment - The treatment of wastewater by
biological methods after primary treatment by sedimentation.
Sedimentation - Settling by gravity of matter suspended in water,
Service Water - The water in general use throughout a plant
Usually in canmaking this is a municipal or potable water
but it may be specifically treated water in those ar-eas
where the readily available water is not suitable for
canmaking.
387
-------
Settleable Solids - O) That matter in wastewater wh.ich will not
"* stay in suspension during a preselected settling period,
such as one hour, but either settles to the bottom or floats
to the top. .(2) In the Imhoff cone test, the volume of mat-
ter that settles to the bottom of the cone in one hour.
Skimmer - A device to remove floating matter from wastewaters.
Sludge - The solids (and accompanying water and organic matter)
which are separated from sewage or industrial wastewater.
Sludge Dewatering - A process used to increase the solids
concentration of sludge.
Sludge Disposal - The final disposal of solid wastes.
Solvent - A liquid capable of dissolving or dispersing one or
more other substances.
Spills - A chemical or material spill is an unintentional dis-
charge of more than 10 percent of the daily usage of a
regularly used substance. In the case of a rarely used (one
per year or less) chemical or substance, a spill is that
amount that would result in 10% added loading to the normal
air, water or solids waste loadings measured as the closest
equivalent pollutant.
Stamping - Forming or cutting of can tops by the application of a
die.
Suspended Solids - (1) Solids that either float on the surface
of,orare in suspension in water, wastewater, or other
liquids, and which are largely removable by laboratory
filtering. (2) The quantity of material removed from
wastewater in a laboratory test, as prescribed in "Standard
Methods for the Examination of Water and Waste Water" and
referred to as nonfilterable residue.
Teratoqenic - Referring to the ability of a substance to form
developmental malformations and monstrosities.
Three-piece cans - Cans formed by combining a cylindrical portion
andtwoends. Usually, the sides are formed by wrapping a
metal around a mandrel and locking the seam.
Total Cyanide - The total content of cyanide including simple
and/orcomplex ions. In analytical terminology, total
cyanide is the sum of cyanide amenable to chlorination and
that which is not according to standard analytical methods.
388
-------
Total Solids - The total amount of solids in a wastewater in
solution and suspension.
Toxicity - Referring to the ability of a substance to
jury to an organism through chemical activity.
cause in-
Treatment Facility Effluent - Treated process wastewater before
discharge.
Trimming - Removal of excess metal from the top of a
ng -
ody.
shaped can
body
Turbidity - (1) A condition in water or wastewater caused by the
presence of suspended matter, resulting in the scattering
and absorption of light rays. (2) A measure of fine
suspended matter in liquids. (3) An analytical quantity
usually reported in arbitrary turbidity units determined by
measurements of light diffraction.
Two-piece cans - Cans formed by drawing a flat metal plate into a
cup and attaching a top.
Viscosity - That property of a liquid paint or coating material
which describes its ability to resist flow or mixing. Paint
viscosity is controlled by solvent additions and its control
is essential to effective roller-coater operation and
uniform dry films thickness.
Waste plate - Tin plate with defects too severe to repair. It is
used for making cans for products such as paint which will
not be adversely affected by the defects.
Water Balance - An accounting of all water entering and leaving a
unit process or operation in either a liquid or vapor form
or via raw material, intermediate product, finished product,
by-product, waste product, or via process leaks, so that the
difference in flow between all entering and leaving streams
is zero.
Water Use - The quantity of process water used in processing a
specified number of cans (expressed as 1/1,000 cans).
Weir - A diversion"dam. (2) A device that has a crest and some
containment of known geometric shape, such as a V,
trapezoid; or rectangle and is used to measure flow of
liquid* The liquid surface is exposed to the atmosphere.
Flow is related to upstream height of water above the crest,
to position of crest with respect to downstream water
surface, and to geometry of the weir opening.
389
-------
OIL AND GREASE ANALYTICAL METHOD
s.
For determining the concentration of oil and grease in wastewater
samples from all subcategories of coil coating, the following
methodology which is based on Standard Methods, 15th Edition,
Methods 503A and 503E is followed. In this method,, a partition
gravimetric procedure is used to determine hydrocarbqn (petroleum
based) oil and grease.
f
(1) Apparatus
(i) Separatory funnel, 1 liter, with TFE* stopcock.
(ii) Glass stoppered flask, 125 ml.
(iii) Distilling flask, 125 ml.
(iv) Water bath.
(v) Filter paper, 11 cm diameter.2
(vi) Glass funnel.
(vii) Magnetic stirrer and Teflon coated stir bar
(2) Reagents
(i) Hydrochioric acid, HCi, 1+1. :
(ii) Trichlorotrifluoroethane^ (1,1,2-trichloro-l,2,2-tn-
fluoroethane), boiling point 47<>C.
The solvent should leave no measurable residue on
evaporation; distill if necessary.
Do not use any plastic tubing to-transfer solvent
between containers.
(iii)Sodium sulfate, Na2S04, anhydrous crystal
(iv) Silica gel, 60 to 200 mesh*.
Dry at 110°C for 24 hours and store in a tightly sealed
container.
(3) Procedure
To determine hydrocarbon oil and grease, collect about 1
liter of sample and mark sample level in bottle for later
determination of sample volume. Acidify to pH 2 or lower;
generally, 5 ml HCI is sufficient. Transfer to a separatory
funnel. Carefully rinse sample bottle with 30 ml
trichlorotrifluoroethane and add solvent washings to separatory
funnel. Preferably shake vigorously for 2 minutes. However, if
it is suspected that a stable emulsion will form, shake gently
for 5 to 10 minutes. Let layers separate. Drain solvent layer
through a funnel containing solvent-moistened filter paper into a
tared clean flask. If a clear solvent layer cannot be obtained,
add 1 g Na2SO4 to the filter paper cone and slowly drain
emulsified solvent onto the crystals. Add more Na2S04 if
necessary. Extract twice more with 30 ml solvent each but first
rinse sample container with each solvent portion. Combine
extracts in tared flask and wash filtetf with an additional 10 to
390
-------
20 ml solvent. Add 3.0 g silica gel. Stopper flask and stir on
a maqnetic stirrer for 5 minutes. Filter solution through filter
g Ind wash silica gel and filter paper with 10 ml solvent and
Lne with filtrate in tared distilling flask. Distill solvent
from distilling flask in a water bath at 7QoC. Place flask on a
water bath at" 70°C for 15 minutes and draw air through it with an
applied vacuum for the final 1 minute. Cool in a desiccator for
30 minutes arid weigh.
(4) Calculations
Calculation of O&G-E; If the organic solvent Is free of
residue the gai~in weight of the tared distilling flask is due
to hydrocarbon oil and grease. Total gain in weight, E, is the
amount of hydrocarbon oil and grease in the sample (mg);
mq (hydrocarbon oil and grease)/! = E x 1000
^ > ' • • • ml sample
(5) Use of O&G-E; The value, O&G-E shall be used^ as the
measure 0r~co¥pTTi[HEe with the oil and grease limitations and
standards set forth in this regulation except where total O&G is
specifically required.
* Teflon® or equivalent
2 Whatman No. 40 or equivalent
3 Freon or equivalent
* Davidson Grade 950 or equivalent
391
-------
1
METRIC UNITS
CONVERSION TABLE
MULTIPLY (ENGLISH UNITS)
ENGLISH UNIT ABBREVIATION
acre ac
acre - feet ac ft
British Thermal
Unit BTU
British Thermal
Unit/pound BTU/lb
cubic feet/minute cfm
cubic feet/second cfs
cubic feet cu ft
cubic feet cu ft
cubic inches cu in
degree Fahrenheit *F
feet ft
gallon gal
gallon/minute gpm
horsepower hp
inches 1n
inches of mercury in Hg
pounds lb
million gallons/day mgd
mile mi
pound/sguare
inch (gauge) psig
square feet sq ft
square inches sq in
ton (short) ton
yard yd
* Actual conversion,,not a multiplier
by TO OBTAIN (METRIC UNITS)
CONVERSION ABBREVIATION METRIC UNIT
hectares
cubic meters
kilogram « calories
kilogram calories/kilogram
cubic meters/minute
cubic meters/minute
cubic meters
liters
cubic centimeters
degree Centigrade
meters
liters
liters/second
killowatts
centimeters
atmospheres
kilograms
cubic meters/day
kilometer
atmospheres (absolute)
square meters
square centimeters
metric ton (1000 kilogram)
meter
0.405
1233.5
0.252
0.555
0.028
1.7
0.028
28.32
16.39
0.555(°F-32)*
0.3048
3.785
0.0631
0.7457
2.54
0.03342
0.454
3,785
1.609
(0.06805 psig +1)*
0.0929
6.452
0.907
0.9144
ha
cu m
kg cal
kg cal /kg
cu m/min
cu m/min
cu m
1
cu cm
•C
m
1
I/sec
kw
cm
atm
kg
cu m/day
km
atm
sq m
sq cm
kkg
m
aUJ.aOVHWMENTFWNTINaOmCI: 198
-------