Vaccari D.A., Strom P.F., Alleman J.E. Environmental Biology for Engineers and Scientists

Подождите немного. Документ загружается.

Certainly, one of the most frequent problems with suspended-growth systems is exces-

sive ﬁlament presence, which leads to a condition referred to as bulking. Healthy, well-

structured ﬂoc usually contains a nominal-to-moderate level of ﬁlament presence

(Figures 16.25 and 16.26). In fact, it has been proposed that ﬁlaments may serve as a

‘‘backbone,’’ giving strength to ﬂoc and allowing them to grow larger. Recent advance-

ments with confocal microscopy have helped to conﬁrm this backbone role with the ﬁla-

ments found in ﬂoc partic les. The image shown in Figure 16.27 was clipped from an

animated video sequence whose sequential frames depict the progressive rotation of a

ﬂoc particle with a loosely arrayed, and highly nonhomogeneous, structure bearing

several backbone-type ﬁlament threads. However, these ﬁlamentous forms are sometimes

present in excessive amounts, extending out from the ﬂoc into the bulk solution, thereby

leading to bulking behavior. Under these conditions the ﬁlaments can impede sedimenta-

tion in the secondary settling tank by two mechanisms. First, they act as a sort of para-

chute, increasing viscous drag and reducing the settling velocity of the ﬂocs. Second, the

ﬁlaments act to fend the ﬂocs off from each other, preventing them from compacting

TABLE 16.5 Suspended-Growth Floc Problems and Possible Contributing Factors

Type of Floc Problem Operational Difﬁculties Possible Contributing Factors

Bulking ﬂoc Poor thickening, resulting in

secondary settler overloading

Low dissolved oxygen concentration

Low F/M loading

Insufﬁcient soluble BOD

Nutrient deﬁciency

Low pH

Completely mix reactor mode

Septic wastewater

High wastewater sulﬁde

concentration

Zoogleal ﬂoc Branched, ﬁngerlike ﬂoc ﬁlaments High F/M loading

Low SRT

Jelly ﬂoc Excessive ﬂoc exocellular slime Nutrient deﬁciency

Dispersed ﬂoc Floc destabilization and

breakdown

Hazy, cloudy efﬂuent

High inﬂuent salt concentration

High inﬂuent surfactant

concentration

Extreme high or low SRT

Pin ﬂoc Elevated efﬂuent suspended

solids

High SRT

Rising sludge Floc ﬂotation vs. settling Denitrifying sludge blanket and gas

release

Anaerobic sludge blanket and gas

release

Foaming Stabilization of bubbles

Floc solids bubble entrapment

High inﬂuent surfactant

concentration

Hydrophobic cell surfaces

Possible oil and grease recycle in

skimmings

608

BIOLOGICAL APPLICATIONS FOR ENVIRONMENTAL CONTROL

toward the bottom of the tank, such that the sludge then takes on a ‘‘bulky’’ condition,

with SVI values in excess of 150 mL/g.

Taken to an extreme, ﬁlaments may replace almost completely any semblance of a nor-

mal ﬂoclike structure (Figures 16.28 to 16.30). For many years, one particular sheathed

bacteria, Sphaerotilus natans, was blamed routinely as the responsible agent for ﬁlamen-

tous bulking problems, but further study demonstrated that several other bacterial forms,

and occasionally fungi, could be involved. During the early 1970s the work of D. H.

Eikelboom in The Netherlands developed a systematic microscopic method which

Figure 16.25 Suspended-growth ﬂoc with nominal ﬁlaments: (a) light microscope view at 400;

(b) SEM view at 600.

Figure 16.26 Suspended-growth ﬂoc (400) with moderate ﬁlament growth.

WASTEWATER TREATMENT 609

recognized more than two dozen different ﬁlament types that coul d contribute to bulking

events (Eikelboom and van Buijsen, 1981). Eikelboom’s classic work, which was then

continued and expanded by a group working with David Jenkins in the United States,

strongly suggested that several causes were possible, each as a result of the excessive

growth of a different type of ﬁlamentous organism. Table 16.6 provides an overview of

Figure 16.27 Confocal microscopy image of suspended growth ﬂoc structure.

Figure 16.28 Suspended-growth ﬂoc with heavy ﬁlament growth: (a) light microscope view at

400,(b) SEM view at 800.

610

BIOLOGICAL APPLICATIONS FOR ENVIRONMENTAL CONTROL

the various ﬁlamentou s forms they typically found, including some forms that have only

been distinguished with numbers (as originally assigned by Eikelboom) since they have

not been formally identiﬁed and given ofﬁcial genus and species names (Jenkins et al.,

1993). Table 16.7 provides a synopsis of the correlation observed between the various

microbial ﬁlaments and their contribution to bulking problems.

Figure 16.29 Solely ﬁlamentous suspended-growth matrix (400).

Figure 16.30 Magniﬁed images of ﬂoc ﬁlaments.

WASTEWATER TREATMENT 611

Knowing which condition is the cause of a particular bulking episode—by deter-

mining which ﬁlamentous types are present—is o f course ver y helpful in developing

control strategies. Bulking often appears in relation to extremely low organic loadings

(extremely low F/M or high SRT) or low dissolved oxygen concentrations in the aera-

tion tank. Occasionally, high sulﬁde concentrations, either from the inﬂuent wastewater

or produced in anaerobic zones within the treatment plant, can promote growth of

ﬁlamentous sulfur-oxidizing bacteria. Low pH can also lead to fungal bulking, but

this is a rather rare occurrence when pH control is available. Inadequate nutrients,

especially nitrogen and probably phosphorus, can also lead to a form of bulking or to

jelly sludge, in which excessive levels of exocell ular slime materials are produc ed.

However, this behavior i s more apt to occur with unique industrial wast ewaters (e.g.,

TABLE 16.6 Microbial Factors with Filamentous Bulking and Foaming in Activated Sludges

Percentage of Facilities Experiencing

Designated Problematic

Bacterial Growth

Rank Bacterial Agent Primary Secondary

1 Nocardia sp. 31 17

2 Type 1701 bacteria 29 24

3 Type 021N bacteria 19 15

4 Type 0041 bacteria 16 47

5 Thiothrix sp. bacteria 12 20

6 Sphaerotilus natans 12 19

7 Microthrix parvicella 10 3

8 Type 0092 Bacteria 9 4

9 Haliscomenobacter hydrossis 945

10 Type 0675 bacteria 7 16

Other forms Bacterial types 0803, 1851, 0961,

0581, and 0914, plus Nostoicoida sp.,

Beggiatoa sp.; undeﬁned fungi

Source: Adapted from Grady et al. (1999).

TABLE 16.7 Correlations Between Filamentous Organism Types

and Causes of Bulking

Cause of Bulking Indicative Filament Types

Major (occurring commonly)

Low organic load 0041, 0092, Microthrix parvicella,

Nocardia-like spp. 0581, 0803

Low dissolved oxygen 1701, 021N, Sphaerotilus natans, 1863

Minor (occurring infrequently)

Low pH Fungi

Sulﬁdes Thiothrix-like, Beggiatoa, 021N

Low N (or P) S. natans

612

BIOLOGICAL APPLICATIONS FOR ENVIRONMENTAL CONTROL

with canning operations), since domestic sewage usually has a considerable excess of

nitrogen and phosphorus.

Yet another type of bulking, which is rather less common than the ﬁlamentous version,

is caused by a particular type of ﬂoc structure largely composed of the bacterium,

Zoogloea ramigera, and for this reason this problem is referred to as zoogloeal bulking.

These ﬂocs are characterized by branched, ﬁngerlike projections that can have an effect

on settling similar to that of other ﬁlaments. Whereas zoogloeal ﬂocs are seen in small

numbers in perhaps 10% of suspended-growth systems, they are present occasionally in

excessive amounts and cause bulking. They are favored by high organic loading rates

(high F/M, low SRT). At one time, early investigators thought that these zoogl eal species

were the dominant, and perhaps only, ﬂoc-forming microorganisms in suspended-growth

systems. This assumption is now known to be incorrect, but the (improper) use of the term

zoogloea in reference to all ﬂoc forms (rather than the speciﬁc unusual type of ﬂoc) per-

sists among some in the ﬁeld.

Note that not every ﬁlamentous type has been linked directly with a speciﬁc cause of

bulking. Also, there is still some debate about some of the relationships. For example, one

ﬁlament, generically referred to as type 021N, in addition to being a sulﬁde oxidizer, has

been associated by some with low dissolved oxygen (DO) and by others with low F/M.

Part of this disagreement may stem from the likelihood that some of the ﬁlamentous types

established using only the microscope may actually be composed of more than one spe-

cies. Future efforts to apply molecular biology tools will undoubtedly help to clear up

some of this confusion.

Reactor design may also inﬂuence bulking. For example, with completely mixed reac-

tor designs, wastewater and recycled sludge are vigorously mixed and rapidly distributed

throughout the tank, such that the incoming wastewater is rapidly diluted as it enters the

reactor. This circumstance then leads to uniformly low substra te concentrations within the

aeration tank, a condition that favors the low-F/M ﬁlamentous bacteria. Bacteria living

within these ﬂocs would be at a disadvantage to the extended ﬁlaments, which reach

out into the bulk solution to an extent where they have far higher access to substrates.

By comparison, an alternative plug-ﬂow reactor conﬁguration has no such initial mixing,

with wastewater and return sludge being introduced at the head end of the linear reactor

and then ﬂowing together down the length of the tank, or through several tanks in series.

In turn, there is far less initial dilution of the incoming substrates, and the elevated initial

substrate concentration facilitates a higher rate of diffusion into the ﬂoc to an extent which

then negates the physical advantage of extended ﬁlament growth. However, plug-ﬂow sys-

tems are more apt to experience higher initial oxygen uptake rates at the head end of their

operations, and unless care is taken to meet this higher demand, there may still be a con-

trary opportunity for ﬁlament growth due to low-DO bulking. Yet another innovative

approach to the control of ﬁlament bulking is to use a special reactor conﬁguration

designed speciﬁcally to optimize and select for the growth of ﬂoc-forming bacteria at

the expense of ﬁlaments. These selector systems with suspended-growth proce sses are

typically quite small, usually less than 1% of the total aeration tank volume and placed

just ahead of the main aeration tank and then mixed and/or aerated to promote aerobic or

anoxic conditions favorable to ﬂoc formers.

Another approach for control of nuisance ﬁlaments would be to apply chlorine or other

strong oxidant (e.g., hydrogen peroxide) to the return sludge on an intermittent basis.

Because the ﬁlaments are more exposed to the solution than are the ﬂoc formers, being

extended outward from the ﬂoc into the surr ounding bulk solution, they are also more

WASTEWATER TREATMENT 613

susceptible to toxins such as these oxidants. However, the recurring necessity for using

this method can be costly, and in the case of high-level chlorination there are also second-

ary concerns stemming from the potential formation of toxic trihalomethane compounds

within the mixed liquor.

Contrasted with the circumstance of excess ﬁlaments and bulking, though, there are

also problems with settling stemming from other difﬁculties tied to unusual ﬂoc confor-

mations or problems. One such condition, referred to as pin ﬂoc or pin-point ﬂoc, stems

from the absence of any ﬁlaments whatsoever. These small-diameter ﬂoc forms are

affected particularly by high-level aeration, which imposes excessive shear forces leading

to ﬂoc disaggregation. Since there is no ‘‘net’’ of ﬁlaments that would otherwise strain out

these particles physically, pin ﬂoc end up getting left behind during settling and thus con-

tribute to a higher-than-desired efﬂuent suspended solids concentrations.

Undesired ﬂoc breakdown and dispersed growth can also be encountered in a variety

of instances. High-level inﬂuent concentrations of salt or surfactants have been found to

destabilize ﬂoc structures, and short SRT systems with high-rate bacterial growth rates

have shown a tendency toward dispersed rather than ﬂocculated growth, due to a lack

of exocellular polymer release. Toxic shock impacts have also been proposed as a factor

behind dispersed growth, where a toxic shock initially kills off a portion of an existing

biomass, and then the remaining organisms grow extremely fast, in an unﬂocculated fash-

ion, because of the reduced competition for substrates.

Foaming is yet another problem that can occur in mixed li quors. Occasional ly, this

may be the result of a nonbio degradable surfactant (a surface-active agent) that enters

the system. Other times, degradable detergents and other foaming materials (including

proteins and DNA) that are present in the wastewater may not be degraded quickly enough

and may lead to surface tension reductions that trigger the release of foam. This typically

occurs either during startup of a plant and with very short SRTs, or after washout of

sludge during a storm. Both of these types of foam are usually white and can be controlled

with defoaming compounds or water sprays. The most common and serious foaming

problems, however, are the result of excessive growths of certain gram-positive ﬁlamen-

tous bacteria. These seem all to be actinomycetes, and most are highly branched. They

were originally called Nocardia, but it is now recognized that they also include several

related genera, including Gordonia, Skermania, Rhodococcus, and Tsukamurella.

Occasionally, the bulking organism Microthrix parvicella (which is not branched) also

causes this type of foaming. These bacteria seem to have a waxy coating, which makes

it hard to rewet them once they get lifted into the foam layer. They are slow-growing and

therefore are favored by longer SRT conditions. However, once they form a foam, they

may accumulate on the surface of the aeration and/or secondary settling tanks to an

extent that control of waste to achieve a desired SRT is extremely difﬁcult. In severe

cases, half of the system’s biomass may actually be in the foam, since it gets so thick.

The foam may even overﬂow the tank, or get pushed out by the wind, creating a

slippery safety hazard on walkways as well as an aesthetic problem. The foaming bacteria

appear to be able to utilize hydrocarbons and other fats and oils that may not readily

be available to many of the other organisms present. Defoaming compounds are not

effective in their control, and in some cases may even be utilized by the organisms as

a carbon and energy source. Instead, a combination of decreased sludge age and physical

removal of the foam (by skimming or vacuum) is generally used. Improved oil and

grease removal in primary treatment, and spray chlorination of the foam, may also be

beneﬁcial.

614 BIOLOGICAL APPLICATIONS FOR ENVIRONMENTAL CO NTROL

Occasionally, sludge will settle well in the clariﬁer and then suddenly bob back up to

the top. This results from gas bubble formation in the sludge blanket while it is sitting at

the bottom of the settling tank. This is called rising sludge if the bubbles are composed of

nitrogen gas (N

), formed as the result of denitriﬁcation. Since denitriﬁcation can take

place only if nitrate and/or nitrite are present while oxygen is absent, this means that

sludge rising can occur only if nitriﬁcation has ﬁrst occurred in the aeration tank and

then the sludge has gone anoxic in the settling tank. In cases where no nitrate or nitrite

are present, anaerobic sludge can ﬂoat to the top as a result of the production of other

gases, especially hydrogen (H

). These problems can be controlled by increasing the rate

at which sludge is returned to the aeration tank, so that settled biomass within the clar-

iﬁer’s sludge blanket does not have a chance to go anoxic or anaerobic.

Another important aspect in the microbial makeup of a suspended-growth reactor’s

biomass involves the absence, or presence, of special ized bacteria uniquely acclimated

for the metabolic uptake or conversion of speciﬁc contaminant species, including not

only nutrients such as nitrogen and phosphorus but also a wide range of potentially inhi-

bitory, and yet still biodegradable, compounds (e.g., phenolics, thiocyanates) (Bitton,

1999). The desired acclimation and retention of these specialized bacteria would be ana-

logous to that of their development within ﬁxed-ﬁlm systems, and there is limited evi-

dence to suggest that ﬂoc particles may also provide, at least partially, a comparable

sequencing of cometabolic aerobes and anaerobes located within the exterior and interior

regions resp ectively.

In the particular case of pursuing complementary oxidative and reductive nitrogen

removal steps, design and control measures to secure the presence and vitality of

nitrogen-oxidizing and nitrogen-reducing bacteria within biomass will determine whether

nitriﬁcation and denitriﬁcation were able to occur. Given the low growth rates commonly

maintained by nitrifying bacteria, the key to achieving nitriﬁcation within an activated

sludge is that of retaining reactor biomass for periods (maintaining an SRT) in excess

of 4 days (and perhaps as much as 8 to 10 days during cold -weather periods). At this

point, a sufﬁcient (albeit, even then rather small—at levels of a few percent or lower of

the total viable cells) inventory of these nitrifying lithotrophs can be kept to effect the

desired oxidation of ammonia to nitrate.

This strategy for securing nitriﬁers has subtle aspects that are inherently similar to that

associated with ﬁxed-ﬁlm systems. A competitive advantage for nitriﬁers in ﬁxed-ﬁlm

systems is maintained by limiting the amount of incoming organics available to compet-

ing heterotrophs; on the other hand, suspended-growth nitrifying reactors maintained at

high solids retention times are able to produce similarly low bulk solution levels of avail-

able organic carbon, such that the faster-growing heterotrophs are energetically con-

strained and unable to impose competitive stress. In fact, the process of nitriﬁcation

has now become a rather com monplace requirement for a considerable number of waste-

water treatment facilities given the inhibitory impact caused by low-level (sub-ppm) free

ammonia concentrations on ﬁsh living within downstream receiving waters. Taken one

step further, many newer activated sludge systems are also being designed to encompass

full nitrogen removal, ther eby matching ammonia oxidation (via aero bic nitriﬁcation)

with nitrate reduction (via anoxic denitriﬁcation).

This combined nitriﬁcation–denitriﬁcation practice provides a number of potential

beneﬁts, including (1) the ability to achieve enhanced levels of overall nitrogen removal,

thereby reducing the downstream impact of this prospective nutrient; (2) the ability to

secure a recovery of the oxidizing ‘‘power’’ provided by nitrates, thereby conserving

WASTEWATER TREATMENT 615

the energetic investment previously made to create these oxidized products during aerobic

nitriﬁcation; and (3) the ability to achieve a similar recovery of alkalinity during deni tri-

ﬁcation, thereby partially offsetting the alkalinity loss realized during prior nitriﬁcation.

Suspended-growth systems designed for combined nitriﬁcation and denitriﬁcation sys-

tems tend to fall into one of two alternative design categories, including zoned and tem-

porally sequenced (i. e., ﬂip-ﬂopped aeration and mixing) arrangements to secure the

necessary aerobic and anoxic phases. One such popular, zoned design strategy, known

as the modiﬁed Ludzack–Ettinger (MLE) scheme is depicted in Figure 16.31. The ﬁrst,

anoxic stage in this MLE system mixes incoming raw wastewater with a nitrate-bearing

ﬂow brought back from the trailing aerobic nitriﬁcation reactor by way of an internal

feedback recycle stream, thereby facilitating the conditions necessary for efﬁcient reduc-

tion and removal of this recycled nitrogen. Within this initial anoxic zone, therefore, deni-

triﬁers employ incoming organics as their electron donor while reductively using nitrates

as an electron acceptor, thereby producing nitrogen gas.

Example 16.2: Preliminary Activated Sludge Design After presenting their preliminary

design estimate for a standard trickling ﬁlter with the Deer Creek, Illinois, community

(see Example 16.1), the community’s consulting engineer was asked to develop yet

another preliminary design estimate for a standard secondary activated sludge system

which might be used alternatively for this facility’s secondary treatment process. As

explained in Example 16.1, the expected average daily wastewater ﬂow for this commu-

nity will be 605.7 m

/day, and the carbonaceous organic loading (CBOD) entering this

secondary activated sludge reactor system will be 72.7 kg BOD/day.

Preliminary Design Details

Hydraulic

loading,

system

sizing,

and retention

times

Design hydraulic retention time (HRT): 4 h (see Table 16.4)

Design solids retention time (SRT): 8 days (see Table 16.4)

Design reactor depth: 5 m

Projected reactor volume: (605.7 m

/day) (4 h ) ¼ 100.9 m

Projected reactor surface area: 100.9 m

/5 m ¼ 20.2 m

Clarifier

and

(Nitrification)

Internal feedback recyle

returning nitrates back to

the denitrifying anoxic zone

Raw

wastewater

influent

Waste

sludge

Underflow recyle

returning settled biomass back to

the denitrifyin

anoxic zone

Treated

effluent

Aerobic

Organic carbon oxidation

and

nitrate reduction

(Denitrification)

Anoxic

Organic carbon oxidation

and

ammonia oxidation

(Nitrification)

Figure 16.31 Representative two-tank plus two-recycle design scheme for biological nitrogen

removal. (Note: This particular arrangement is referred to as the modiﬁed Ludzack–Ettinger system,

based on its modiﬁed addition of an internal feedback recycle stream to the original Ludzack–

Ettinger design scheme.)

616

BIOLOGICAL APPLICATIONS FOR ENVIRONMENTAL CONTROL

Biomass

content

Estimated theoretical biomass yield (Y

): 0.5

Estimated biomass endogenous decay and death rate: 0.05 day

1

Estimated active reactor biomass concentration:

ðSRT=HRTÞðY

=ð1 þ k

 SRTÞðCBOD

influent

 CBOD

effluent

Þ¼

1920 g=m

Estimated active biomass mass: (1920 kg/m

) (100.9 m

) ¼193.8 kg

Estimated total suspended solids mass: (active biomas s/0.75) ¼

258.4 kg

Organic

Projected organic loading:

ð72:7 kg CBOD=dayÞ=ð100:9m

Þ¼0:73 kg=m

 day

Projected F/M (food/microorganism) ratio:

ð72:7 kg BOD=dayÞ=ð258:4kgÞ¼0:38 day

1

Related notes 1. As with Example 16.1, this preliminary design was based on a sin-

gle reactor receiving an average daily ﬂow, as compared to multi-

ple reactors and peak etc. design ﬂow conditions.

2. The total suspended solids vs. active biomass differences was

based on a typically observed biomass fraction of 75% vs. an

additional inert solids content of 25%.

3. The biomass concentration derivation given above takes into

account composite biomass growth for CBOD depletion, plus

decay and death, with an assumed efﬂuent BOD of 10 mg/L);

further details regarding this analysis can be found in Wastewater

Engineering (Metcalf and Eddy, 2003).

4. In comparison to the prior trickling ﬁlter design estimate, there are

several important design differences, including:

a. Considerably different contact times (hours at activated

sludge vs. seconds or minutes for the trickling ﬁlter).

b. Considerably different total reactor volumes (the activated

sludge system is 2þ times smaller).

c. Roughly comparable active biomass levels but sizably different

total biomass levels (with a total of 4þ times higher in the

trickling ﬁlter, due to attached underlying anaerobic and inert

bioﬁlm).

Going beyond nitrogen removal, biological phosphorus removal can also be achieved

in full biological nutrient removal (BNR) systems designed speciﬁcally to provide

sequenced anaerobic and aerobic zones which would encourage the growth of a highly

unique bacterial group largely believed to fall within the Acinetobacter genus. These unu-

sual cells consume low-molecular-weight volatile fatty acids (VFAs) during an initial

anaerobic zone while releasing internal reserves of polymerized phosphate (i.e., a poly-

phosphate known as volutin). Moving from the anaer obic to aerobic zones, though, these

same cells reverse the process, balancing their accelerated upta ke of soluble phosphorus

against the oxidation of previously stored VFAs. The end result, therefore, is that of

a biochemically mediated phase transfer of phosphate, shifting incoming soluble

WASTEWATER TREATMENT 617