Chemical Engineering Journal95(2003)
221–234
A comprehensive insight intofloc characteristics and their impact on
compressibility and ttleability of activated sludge
Bo Jin a,∗,Britt-Marie Wilén a,b,Paul Lant a
a Department of Chemical Engineering,The University of Queensland,St.Lucia,Qld4072,Australia
b Water Environment Transport,Chalmers University of Technology,SE-41296Göteborg,Sweden
Abstract
This paper prents a comprehensive study of sludgefloc characteristics and their impact on compressibility and ttleability of acti-vated sludge in full scale wastewater treatment process.The sludgeflocs were characterid by morphological(floc size distribution, fractal dimension,filament index),physical(flocculating ability,viscosity,hydrophobicity and surface charge)and chemical(polymeric constituents and metal content)parameters.Compressibility and ttleability were defined in terms of the sludge volume index(SVI) and zone ttling velocity(ZSV).T
hefloc morphological and physical properties had important influence on the sludge compressibility and ttleability.Sludges containing largeflocs and high quantities offilaments,corresponding to lower values of fractal dimension(D f), demonstrated poor compressibility and ttleability.Sludgeflocs with highflocculating ability had lower SVI and higher ZSV,whereas high values of hydrophobicity,negative surface charge and viscosity of the sludgeflocs correlated to high SVI and low ZSV.The quantity of the polymeric compounds protein,humic substances and carbohydrate in the sludge and the extracted extracellular polymeric substances(EPS) had significant positive correlations with SVI.The ZSV was quantitatively independent of the polymeric constituents.High concentrations of the extracted EPS were related to poor compressibility and ttleability.The cationic ions Ca,Mg,Al and Fe in the sludge improved significantly the sludge compressibility and ttleability.
©2003Elvier Science B.V.All rights rerved.
Keywords:Activated sludgeflocs;Settleability;Compressibility;SVI;ZSV;Floc size;Fractal dimension;Filament index;Viscosity;Flocculating ability; Hydrophobicity;Surface charge;Polymeric constituents;EPS
1.Introduction
Biosolid–liquid paration by gravity ttling is one of the most critical operations in the activated sludge process. Settling and compaction ability of activated sludge is crucial to overall performance and efficiency of the treatment pro-cess as well as the quality of the receiving water.Previous investigations indicate that many activated sludge systems have experienced various biomass paration problems in the ttling tanks[1–3].There are two main types of t-tling problems:(i)bulking sludge due to the proliferation offilamentous bacteria and(ii)poorflocculation properties, e.g.formation of small and lightflocs.It is agreed that both aspects are equally important for the paration properties but the latter one has been much less investigated.In most cas,large,den and strongflocs are desirable for good ttling and compaction of activated sludge.
Theflocculation of activated sludge is an active process and depends on physical,chemical and biological factors.学生会职位
∗Corresponding author.Tel.:+61-7-3365-4479;fax:+61-7-3365-4199. E-mail address:bojin@cheque.uq.edu.au(B.Jin).Activated sludgeflocs are aggregates of suspended solids containing different groups of microorganisms and organic and inorganic particles embedded in a polymeric network of extracellular polymeric substances(EPS)[4–7].Due to the complex nature of theflocs,they display a wide varia-tion in physical,chemical and biological properties.It has been sugg
ested that the binding of the different entities is due to various types of intermolecular interactions such as DLVO-type interactions[8];bridging of EPS by means of divalent and trivalent cations[1,9]and hydrophobic interac-tions[10].Thefloc properties,such as size distribution and morphology,can differ substantially as a result of differ-ences in the environment in the treatment plant.A number offloc characteristics could be expected to exert some direct and/or indirect influence on the activated sludge properties. The complicated interrelationships with respect to physical, chemical and biological factors affecting thefloc charac-teristics and activated sludge properties are summarid in Fig.1.
It is recognid that the amount of EPS,surface properties (colloidal properties),floc size distribution,density,andfil-ament length are the major factors associated with activated
1385-8947/$–e front matter©2003Elvier Science B.V.All rights rerved. doi:10.1016/S1385-8947(03)00108-6
222 B.Jin et al./Chemical Engineering Journal 95(2003)
221–234
Fig.1.Network of physical,chemical and biological factors affecting sludge floc characteristics and activated sludge properties.
sludge properties [10–15].However,only a few isolated pa-rameters have been examined in any of the previous studies.The relative importance of each property and its impact on the bioflocculation of activated sludge have,therefore,not been established.In fact,the fundamental mechanisms of bioflocculation and their impact on the sludge characteris-tics have neither been thoroughly investigated nor success-fully quantified.Therefore,it is not well understood how the physical and biochemical characteristics of sludge flocs affect the ttling and dewatering properties of activated sludge.
The goal of this study was to gain a comprehensive insight into sludge floc characteristics and to identify the key factors which determine the solid–liquid paration properties of activated sludge.Extensive laboratory exami-nations of the physical and chemical properties of activated sludge samples from wage and industrial wastewater treatment plants (WWTPs)were performed.Of particular interest was the characterisation of the floc morphological (floc size distribution,fractal dimension,filament index),physical (flocculating ability,viscosity,hydrophobicity and surface charge)and chemical (polymeric constituents and metal contents)properties,and to identify th
eir impact on compressibility,ttleability and dewaterability of the acti-vated sludge.The results of this study are prented in two papers dealing with:(I)compressibility and ttleability;and (II)dewaterability [16].This paper prents part of an extensive study into the impact of various floc character-istics on the compressibility and ttleability of activated sludge.
2.Materials and methods
2.1.Integrated investigation and approach for activated sludge flocs
The investigations were conducted by collecting activated sludge samples from the target WWTPs and a ries of tests were carried out in the laboratory to obtain an extensive amount of information on the physical,chemical and biolog-ical properties of the sludge flocs.The major factors affecting the activated sludge properties can be classified accordingly:(1)influent and process conditions;
(2)chemical constituents of the sludge flocs;(3)microbial community and activity;
(4)flocculating ability and floc surface properties;and (5)
floc structure.
An overview of the experimental approach is prented in Fig.2.The results from the studies of the impacts of structural and microbial characteristics on the floc stability,and the influence of key chemical constituents in activated sludge on surface and flocculating properties are prented elwhere [6,7].
2.2.Activated sludge samples
Activated sludge samples were taken from ven different full-scale activated sludge WWTPs in Brisbane,Australia,including five wage treatment plants (STPs),treating domestic wastewater,and two industrial activated sludge
B.Jin et al./Chemical Engineering Journal 95(2003)221–234
223
Fig.2.Integrated investigation and approach to floc characteristics and sludge properties in activated sludge process.
treatment process for oil refinery and leachate effluent (Table 1).The sludge samples from each WWTP were col-lected from the aeration tanks and maintained in filled plas-tic containers placed in ice cooler during the transportation from WWTP site to laboratory.Sample tests started imme-diately and were completed within 20h,while being kept in a refrigerator at 4◦C.Sludges from each WWTP were generally examined twice with up to 4months in between.2.3.Extraction of EPS
The EPS were extracted from the activated sludge by mix-ing with a cation exchange resin (Dowex 50×80,Na +form,20–50mesh,Aldrich–Fluka 44445).This extraction proce-dure is bad on the method as described by Frølund et al.[4].This method has been widely accepted and ud for EPS extraction from activated sludge since this procedure results in high extraction efficiency and little or no cell ly-sis or exopo1ymer disruption [4,17,18].Two hundred grams of Dowex,previously washed in a buffer solution,and 0.5l of thickened sludge (8g/l)were transferred to an extraction
Table 1
Process descriptions of the wastewater treatment plants Treatment plant Source of wastewater Biological process Chemical dosage
SRT a (days)Wacol (A)
Domestic 50%;industrial 50%C,N,P 35Oxley Creek (B)Mainly domestic C 4Gibson Island (C)Mainly domestic C,N 18Thorneside (D)Mainly domestic C,N,P Lime and alum 12Capalaba (E)Mainly domestic C,N,P
Alum 15Caltex (F)Oil refinery effluent C 30Tip (G)
Leachate
C,N (quencing batch reactor)
Mg(OH)2
20
The biological process are classified as carbon removal (C);nitrogen removal (N);and phosphorus removal (P).
a Solids retention time (SRT)or sludge age.
vesl with four baffles and stirred at 900rpm for 4h at 4◦C.The extracted EPS was harvested by cen
trifugation at 12,000×g for 15min and filtration through a 0.45m cellu-lo acetate membrane.Total amount of EPS was expresd by sum of carbohydrate,protein,and humic substances.2.4.Flocculating ability带鸟的四字成语
The flocculating ability of activated sludge was deter-mined as the reflocculation ability of sludge flocs after dis-ruption.The method described by Jorand et al.[19]was modified and ud in this study.80ml of sludge sample with a concentration of mixed liquid suspended solids (MLSS)approximately 4g/l was transferred in a beaker placed on ice bath and sonicated at 50W for 15s.This was sufficient to disrupt the flocs,but did not cau cell rupture [5].A 10ml aliquot of the suspension was centrifuged at 1200rpm for 2min and the absorbance of the supernatant was mea-sured at 650nm (A).The rest of the sonicated suspension was stirred on a magnetic stirrer (t at a specific speed to keep the sludge flocs suspended)at ambient temperature
夫妻恩爱224 B.Jin et al./Chemical Engineering Journal 95(2003)221–234
for 15min after which a 10ml aliquot was analyd in the same way as before (B).The standard deviations for tripli-cate samples were 0.2–5%.The flocculating ability of the flocs was calculated as Eq.(1):flocculating ability (%)=
1−B
A
×100(1)
2.5.Relative hydrophobicity
The relative hydrophobicity of the sludge flocs was mea-sured as adherence to hydrocarbons,as detailed by Chang and Lee [20].30ml of the sludge samples were washed and suspended in Tris buffer (0.05mM at pH =7.1).The activated sludge suspension (thickened to approximately MLSS 4g/l)was homogenid by sonication (50W for 2min)at 4◦C to disrupt the flocs to single cells and small micro-colonies.The suspension was agitated uniformly for 5min with 15ml hexadecane (Sigma)in a paratory funnel.After 30min when the two phas had parated completely,the aqueous pha was transferred into another glassware.The standard deviation for the three duplicate samples tested was 1–6%.The relative hydrophobicity was expresd as the ratio of MLSS concentration in the aqueous pha after emulsification (MLSS e )to the concentration of MLSS in the aqueous pha before emulsification (MLSS i ):
relative hydrophobicity (%)=
1−
芋圆糖水
MLSS e
MLSS i ×100(2)2.6.Surface charge
The measurement of surface charge of the sludge flocs was performed by colloidal titration [21].A known vol-ume of sludge sample diluted in distilled water was re-acted with an excess amount of polybrene (Polyscience Inc.)(0.001N),followed by back-titration with polyvinyl sulphate (Aldrich–Fluka)(0.001N)to a colorimetric end-point indi-cated by Toluidin Blue.The surface charge was expresd as milliequivalents per gram of mixed liquor suspended solids of positive or negative colloidal q./g MLSS.The standard deviations for triplicate samples were 3–10%.2.7.Compressibility and ttleability
The compressibility and the ttleability of the activated sludges were evaluated by the sludge volume index (SVI),and the zone ttling velocity (ZSV),respectively.SVI is the volume of 1g of the total suspended solids after 30min of ttling.A higher SVI is related to a poorer compressibility.The ZSV is a key parameter to evaluate the sludge ttling properties,since it determines how much the condary t-tlers can be loaded.ZSV and SVI were measured according to Standard Methods [22]
in a 2l ttling cylinder.When the ttled sludge volume exceeded 25%of the total vol-ume (2l),the sludge was diluted accordingly to minimi the wall effects during ttling.
2.8.Floc size distribution
功亏一篑The floc size distributions were determined by a Malvern Mastersizer/E instrument with a 300mm lens which enables the measurement of particles in the range 0.9–546m.This instrument measures the size of particles by means of light scattering.The samples were diluted in filtrated effluent (0.45m millipore filter)to avoid multiple scattering.The activated sludge suspension was then continuously recycled through the sample cell of the Malvern with a peristaltic pump to be expod to a 2mW He:Ne lar (wavelength 633nm).Each sample was measured three times with a stan-dard deviation 0.1–4.5%.The scattered light is detected by means of a detector that converts the signal to a size distri-bution bad on volume.The average size of the flocs was given as the mean bad on the volume equivalent diameter (D )[4,3].
2.9.Fractal dimension
The structure of the flocs was quantified in terms of fractal dimensions (D f ),D f corresponds to the space filling capacity of an object and is thus a measure of the aggregate structure.The D f dimensio
ns were calculated from the raw light scattering data from the Malvern Mastersizer/E instru-ment according to the method by Spicer et al.[23].This technique is bad on a power law relationship between the total scattering intensity of the light from the aggregates and the magnitude of the scattering vector.The scatter-ing intensity at each detector was calculated from the raw scattering data by means of information from the Malvern Mastersizer/E.By plotting the log of the light scattering intensity as a function of the log of the light scattering vector,the linear slope is the D f .D f varies between 1and 3.Guan et al.[24]ud this method to calculate the fractal dimensions for activated sludge.The high value of the D f is related to compact and den flocs.2.10.Filament index
The sludge flocs were examined by light microscopy and images were captured on a Nikon Microphot FXA mi-croscope via a charge-coupled device connected to a PC.Filamentous organism content was quantified as filament index using the method by Jenkins et al.[25].The number of filamentous organisms was rated on a scale of 1–5,where 1corresponds to no filamentous organism prents and 5corresponds to excessive growth of filamentous organisms,according to Eikelboom and van Bijn [11].2.11.Sludge viscosity
The apparent viscosity was determined using a rotational viscosity meter (Model LVDVII,Brookfield,England).It was measured at the shear rate 100s −1for 5min to keep
B.Jin et al./Chemical Engineering Journal95(2003)221–234225 the sludge in suspension and is expresd as mPa s.The
measurements were carried out at similar suspended solids
concentrations,approximately3g/l,at20◦C.The apparent
viscosity of the sludge is a reflection of internal and external
interactions and forces occurring within the sludgeflocs and
fluids,and describes the deformation of theflocs under the
influence of stress[26–28].
2.12.Chemical analysis
The sludge and the extracted EPS were analyd for
protein,carbohydrate and humic substances as detailed by
Frølund et al.[4].All analys were carried out in triplicate.
Mixed liquid suspended solids and volatile suspended solids
(VSS)were measured according to Standard Methods[22].
Metals in the sludge were analyd byflame-AAS in cen-
trifuged samples adjusted to pH1with supra pure HNO3.
The standard deviation for the determinations was1–10%
for protein and carbohydrate,1–20%for humic substances
and1–5%for cationic ions.
2.1
3.Statistical analysis
猪肺汤怎么做止咳润肺Statistical analys were performed to identify the ma-
jor cau and effect relationships.To simplify the analy-
sis,univariate linear correlations were ud.All statisti-
cal analys were carried out using the software Statistica
(Statsoft,Tulsa,OK,USA).Since a normal distribution
was not obtained for many of the properties examined,a
distribution-free statistical method was ud.The Pearson’s
product momentum correlation coefficient(r p)was ud to
estimate linear estimations.The Pearson’s coefficient r p is
always between−1and+1,where−1means a perfect neg-
ative correlation and+1a perfect positive correlation while
0means abnce of relationship.Non-linear relationships
were conducted by regression analysis.Correlations were
considered statistically significance at a95%confidence in-
terval(P<0.05).
Table2
Summary offloc characteristics of the sludgeflocs and physical properties of the activated sludge
Parameter Unit Activated sludge sample
A B C D E F G
Floc sizem122±2311±2176±3122±563±355±2124±3
D f– 2.16±0.23 1.96±0.06 2.12±0.02 2.15±0.01 2.30±0.04 2.44±0.04 2.09±0 Filament index–24–553–42–411 Flocculating ability%60±155±355±156±237±469±132±5 Hydrophobicity%65±268±370±260±260±664±648±1 Negative surface
charge
meq./g MLSS0.26±0.070.54±0.020.34±0.060.32±0.120.30±0.010.19±0.030.13±0.05
Viscosity mPa s 4.75±0.13 4.59±0.4310.5±0.46 5.19±0.38 4.53±0.13 4.96±0.02 3.97±0.12 SVI ml/g97±18235±11255±5148±5109±1174±845±5 ZSV m/h 2.89±1.26 1.49±1.100.52±0.22 3.04±0.40 4.02±0.60 2.51±0.26 5.94±0.66
V olatile fraction (VSS/MLSS)%80±273±580±579±371±284±459±1
3.Results and discussions
3.1.Sludgefloc characteristics
Thefloc characteristics and physical properties of the ac-
北京到上海飞机几个小时tivated sludges are summarid in Table2,and chemical
composition of the sludge and the extracted EPS is detailed
in Table3.It can be obrved that the sludge samples were
intentionally chon to cover a broad range offloc charac-
teristics.
The mean size of the sludgeflocs varied between40and
320m and most samples had an averagefloc size smaller
than150m.The D f was measured between1.9and2.5.The
水仙花语
quantity offilamentous organisms determined asfilament
index(FI)had a wide range from1to5.Most of the sludges
had an FI lower than3,corresponding to low to moderate
numbers offilamentous bacteria.The sludges from the STPs
contained different levels offilamentous microorganisms,
whereas nofilamentous species was found in the sludge
samples from the two industrial treatment plants.
The sludgeflocs performed differentflocculating abilities
varying between30and70%,but had relatively small vari-
ations in relative hydrophobicity between50and70%.The
sludges had negatively charged surfaces varying from−0.08
to−0.55meq./g MLSS.Table2shows that more than60%
of the biosolids were in the volatile fraction.Apart from one
of the STP sludge sample(C),which contained excessive
quantities offilamentous microorganisms,the viscosity of
the sludge was between4.5and5.5mPa s.
Protein,humic substances and carbohydrate were the ma-
jor polymeric materials in both the sludge and the extracted
EPS.Sum of the amount of protein,humic substances and
carbohydrate is referred as total biopolymers in this study.
Protein was the dominating biopolymer and reprented
more than45%of the total polymeric fractions,followed
by humic substances,and carbohydrate constituting ap-
proximately15–30and13%of the biopolymers in the
sludge and the extracted EPS,respectively.All sludge sam-
