biomass stabilization in the anaerobic digestion of wastewater sludges

6
Biomass stabilization in the anaerobic digestion of wastewater sludges C. Arnaiz a, * , J.C. Gutierrez b , J. Lebrato c a Departamento de Ingenierı ´a Quı ´mica y Ambiental, Escuela Universitaria Polite ´cnica, Universidad de Sevilla, Virgen de Africa 7, 41011 Sevilla, Spain b Departamento de Ciencias Ambientales, Universidad Pablo de Olavide, Ctra. de Utrera km 1, 41013 Sevilla, Spain c Grupo Tratamiento de Aguas Residuales, Escuela Universitaria Polite ´cnica, Universidad de Sevilla, Virgen de Africa 7, 41011 Sevilla, Spain Received 25 June 2004; received in revised form 29 March 2005; accepted 20 May 2005 Available online 11 July 2005 Abstract Sludge stabilization processes include both volatile solid destruction and biomass stabilization. Traditionally, both processes have been considered together, in such a way that, when volatile solid destruction is achieved, the biomass is considered stabilized. In this study, volatile solids reduction and biomass stabilization in the anaerobic digestion of primary, secondary and mixed sludges from municipal wastewater treatment plants were researched in batch cultures by measurements of suspended solids and suspended lipid-phosphate. The estimated kinetic constants were higher in all sludge types tested for the biomass stabilization process, indicat- ing that volatile solids destruction and biomass stabilization are not parallel processes, since the latter one is reached before the former. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Anaerobic digestion; Living biomass; Municipal sludge; Phospholipid analysis; Volatile solids reduction; Wastewater treatment 1. Introduction Solid and semisolid materials removed from the li- quid stream of a municipal wastewater treatment plant are considered to be sludge. Organic waste from primary and secondary treatment constitutes most of the sludge, but it also includes grit, scum and screenings. The types of sludge are: primary sludge, secondary sludge and mixed primary and secondary sludge. Pri- mary sludge comes from primary sedimentation to re- move settleable solids that are readily thickened by gravity. Secondary sludge is biological sludge consisting of the conversion products from soluble wastes in pri- mary effluent and particles escaping primary treatment. Treatment processes such as activated sludge, trickling filter and rotating biological contactors produce second- ary sludges. Finally, sludges produced from combina- tion of primary and secondary sludges will have properties that are—approximately—proportional to their respective compositions (WEF/ASCE, 1992a). Sludge stabilization processes are the key to reliable performance of any wastewater treatment plant. These processes treat the sludges generated in the main treat- ment process, converting them to a stable product for ultimate disposal or use (WEF/ASCE, 1992b), and in- clude both volatile solid destruction and biomass stabil- ization. Volatile solid destruction is needed in order to reduce the volume of sludge requiring ultimate disposal, while the biomass stabilization process limits pathogens and provides a less odorous product, reducing the po- tential to impact negatively on human and biotic health. Traditionally, both processes have been considered to- gether, in such a way that, when volatile solid destruc- tion is achieved, the biomass is considered stabilized. 0960-8524/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2005.05.010 * Corresponding author. Tel.: +34 95 4552812; fax: +34 95 4282777. E-mail address: [email protected] (C. Arnaiz). Bioresource Technology 97 (2006) 1179–1184

Upload: c-arnaiz

Post on 26-Jun-2016

216 views

Category:

Documents


3 download

TRANSCRIPT

Page 1: Biomass stabilization in the anaerobic digestion of wastewater sludges

Bioresource Technology 97 (2006) 1179–1184

Biomass stabilization in the anaerobic digestionof wastewater sludges

C. Arnaiz a,*, J.C. Gutierrez b, J. Lebrato c

a Departamento de Ingenierıa Quımica y Ambiental, Escuela Universitaria Politecnica, Universidad de Sevilla, Virgen de Africa 7, 41011 Sevilla, Spainb Departamento de Ciencias Ambientales, Universidad Pablo de Olavide, Ctra. de Utrera km 1, 41013 Sevilla, Spain

c Grupo Tratamiento de Aguas Residuales, Escuela Universitaria Politecnica, Universidad de Sevilla, Virgen de Africa 7, 41011 Sevilla, Spain

Received 25 June 2004; received in revised form 29 March 2005; accepted 20 May 2005Available online 11 July 2005

Abstract

Sludge stabilization processes include both volatile solid destruction and biomass stabilization. Traditionally, both processeshave been considered together, in such a way that, when volatile solid destruction is achieved, the biomass is considered stabilized.In this study, volatile solids reduction and biomass stabilization in the anaerobic digestion of primary, secondary and mixed sludgesfrom municipal wastewater treatment plants were researched in batch cultures by measurements of suspended solids and suspendedlipid-phosphate. The estimated kinetic constants were higher in all sludge types tested for the biomass stabilization process, indicat-ing that volatile solids destruction and biomass stabilization are not parallel processes, since the latter one is reached before theformer.� 2005 Elsevier Ltd. All rights reserved.

Keywords: Anaerobic digestion; Living biomass; Municipal sludge; Phospholipid analysis; Volatile solids reduction; Wastewater treatment

1. Introduction

Solid and semisolid materials removed from the li-quid stream of a municipal wastewater treatment plantare considered to be sludge. Organic waste from primaryand secondary treatment constitutes most of the sludge,but it also includes grit, scum and screenings.

The types of sludge are: primary sludge, secondarysludge and mixed primary and secondary sludge. Pri-mary sludge comes from primary sedimentation to re-move settleable solids that are readily thickened bygravity. Secondary sludge is biological sludge consistingof the conversion products from soluble wastes in pri-mary effluent and particles escaping primary treatment.Treatment processes such as activated sludge, trickling

0960-8524/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.doi:10.1016/j.biortech.2005.05.010

* Corresponding author. Tel.: +34 95 4552812; fax: +34 95 4282777.E-mail address: [email protected] (C. Arnaiz).

filter and rotating biological contactors produce second-ary sludges. Finally, sludges produced from combina-tion of primary and secondary sludges will haveproperties that are—approximately—proportional totheir respective compositions (WEF/ASCE, 1992a).

Sludge stabilization processes are the key to reliableperformance of any wastewater treatment plant. Theseprocesses treat the sludges generated in the main treat-ment process, converting them to a stable product forultimate disposal or use (WEF/ASCE, 1992b), and in-clude both volatile solid destruction and biomass stabil-ization. Volatile solid destruction is needed in order toreduce the volume of sludge requiring ultimate disposal,while the biomass stabilization process limits pathogensand provides a less odorous product, reducing the po-tential to impact negatively on human and biotic health.Traditionally, both processes have been considered to-gether, in such a way that, when volatile solid destruc-tion is achieved, the biomass is considered stabilized.

Page 2: Biomass stabilization in the anaerobic digestion of wastewater sludges

Nomenclature

k kinetic constant (d�1)MS mixed sludgePS primary sludgeSLP suspended lipid-phosphate (nmol ml�1)

SS secondary sludgeTSS total suspended solids (mg ml�1)VSS volatile suspended solids (mg ml�1)

Table 1Characterization of urban primary sludge and secondary sludge

Parameter Primary sludge Secondary sludge

pH 6.4 7.4TSS (mg ml�1) 44,880 17,980VSS (mg ml�1) 32,700 14,530

1180 C. Arnaiz et al. / Bioresource Technology 97 (2006) 1179–1184

The most widespread stabilization process in munici-pal wastewater treatment plants is anaerobic digestion.Anaerobic digestion offers significant advantages overaerobic systems, like low energy consumption, reducedsolids formation, low nutrient requirement and potentialenergy recovery from the methane produced (Stewartet al., 1995). This process is now widely used in manyenvironmental applications, in different configurationsand modes of operation.

The volatile solids content is commonly used as anindicator of the amount of organic matter contained insludge. Therefore, the amount of volatile solids destruc-tion achieved in a sludge stabilization process can beused to measure its effectiveness to stabilize the organiccomponent of the sludge (WEF/ASCE, 1992b). On theother hand, the quantity of biomass is one of the funda-mental parameters in the design and control of biologi-cal wastewater treatment processes. The sludge retentiontime, the organic loading rate and the volumes of recir-culation and sludge are intimately related to the quan-tity of microorganisms.

Reactor performance of anaerobic sludge digestionfrom urban wastewater treatment plants is frequentlycharacterized in terms of volatile solids reduction.Regarding biomass, although new methods are progres-sively used, the most widespread method of biomassdetermination in these treatment plants is, also, theascertainment of the total suspended solids (TSS) or vol-atile suspended solids (VSS). Therefore, the main disad-vantage of measuring biomass in terms of volatile solidsconcentration is that its estimation includes not only liv-ing microorganisms, but inert mass, exopolymers andabsorbed organic matter in flocs or biofilms as well(Singh et al., 1994).

Phospholipids, a cell wall component, offer manyadvantages over other assays for selective cellular bio-mass estimation (high specific content and relativelyconstant amounts), and their determination by colori-metric methods is relatively simple, reproducible andsensitive (Lazarova and Manem, 1995). Phospholipids,which comprise up to 90–98% of the bacterial mem-branes, do not form part of cell reserves and are easilydegraded during bacterial lysis (White et al., 1979).Therefore, their estimation only includes living biomass.Lipids have been widely used in environmental samplesand recently applied to anaerobic wastewater biofilm(Arnaiz et al., 2003).

In this study, volatile solids and living biomass in theanaerobic digestion of different sludge types weremeasured. The purpose was to determine if biomassstabilization can be reached even when volatile soliddestruction is not completely achieved.

2. Methods

2.1. Batch reactors

The experiments were carried out in batch reactors(250 ml volume) each containing 20 ml of inoculumand 180 ml of sludge. The sludges—used as carbonand energy source—were primary sludge (PS), second-ary sludge (SS) and a 1:1 mixture of primary and sec-ondary sludge (MS), all of them obtained from theEast Urban Wastewater Treatment Plant of Sevilla,Spain. A characterization of these sludges is shown inTable 1. The inoculum for the reactors was obtainedfrom an anaerobic continuously stirred tank reactor thatprocessed a 1.5:1 mixture of PS and SS. The inoculationwas carried out under a non-oxygen gas flow, because ofthe strict anaerobic technique that relies on excluding alltraces of oxygen from the medium during culture prep-aration. In this case, the culture was bubbled with nitro-gen for 15 min. The flasks used had a one-way valve toallow gas to escape (Lebrato, 1990). The batch was incu-bated in darkness at a temperature of 35 �C, in an orbi-tal shaker-incubator, New Brunswick Scientific G25, at250 rpm. All the tests were run in triplicate. TSS, VSSand suspended lipid-phosphate (SLP) were routinelyanalyzed.

2.2. Analytical methods

TSS and VSS were measured according to StandardMethods (APHA-AWA-WPCF, 1992), understandingthat TSS is the portion of total solid retained by the fil-

Page 3: Biomass stabilization in the anaerobic digestion of wastewater sludges

0

20

40

60

80

100

0 14 27 41 54 68

% o

f V

SS r

emov

al

81

80

100

l

C. Arnaiz et al. / Bioresource Technology 97 (2006) 1179–1184 1181

ter and that VSS is the volatile fraction of TSS afterignition.

The procedure used in this study in order to determineSLP was a modification of that found in Findlay et al.(1989). The procedure consisted of: (a) 0.5 ml of samplewas added to a 70 ml screw-cap test tube. Then, 20 ml ofchloroform, 20 ml of methanol and 20 ml of deionizedwater were added. The extraction mixture was gentlyshaken for 10 min and allowed to stand up to completephase separation. (b) In order to facilitate recovery ofthe chloroform, the aqueous (upper) phase was aspiratedfrom the test tube with the aid of a vacuum pump andsub-samples of 5 ml of the chloroform layer were trans-ferred into 10 ml screw-cap test tubes. At this point, lip-ids can be stored at �20 �C. (c) The chloroform wasremoved under a stream of nitrogen and phosphatewas liberated from lipids by adding 2.7 ml of a potassiumpersulfate solution (5 g added to 100 ml of 0.36 NH2SO4) and the sealed test tubes were heated in an ovenat 105 �C for 1 h. (d) Phosphate release by persulfatedigestion was determined by adding 0.6 ml of an ammo-nium molybdate solution (2.5% of (NH4)6Mo7O24 Æ4H2Oin 5.72 N H2SO4 allowed to stand for 10 min) and 2.7 mlof a malachite green solution (0.111% polyvinyl alcoholdissolved in water at 80 �C was allowed to cool, and0.011% malachite green was then added and allowed tostand for 30 min). (e) The absorbance at 610 nm was thenread using a spectrophotometer (HITACHI U-2000).The concentrations of phosphate were calculated byusing the regression line from a standard curve obtainedby digesting 10, 20, 40, 60, 80, 100 and 150 ll of a 1 mMglycerol-phosphate solution.

In the phospholipid analysis, solvents for lipid ex-traction were of high quality, lipid standard [DL-a-phos-phatidylethanolamine, dipalmitoyl (C16:0)], calciumglycerol phosphate and malachite green were of reagentquality (Sigma) and polyvinyl alcohol was 98% hydro-lyzed (average molecular weight, 13,000–23,000; AldrichChemical Co., Inc.). Glassware was washed with phos-phate-free detergent, rinsed 5 times with tap water and2–3 times with deionized water, and air-dried. Glasswarewas rinsed with chloroform just before use. Potassiumpersulfate solution was replaced monthly.

0

20

40

60

0 14 27 41 54 68 81

Time (d)

% o

f SL

P r

emov

a

Fig. 1. VSS and SLP removal efficiency for primary sludge (j),secondary sludge (n) and mixed sludges (·). Results are expressed asmean ± standard deviation of three replicate analyses.

2.3. Data analysis

The volatile solids reduction and biomass stabili-zation can be expressed by the first-order reactions de-scribed in Eqs. (1) and (2), respectively.

log VSS � log VSS0 ¼ �kt=2.303 ð1Þlog SLP � log SLP0 ¼ �kt=2.303 ð2Þ

where VSS0 is the initial volatile suspended solids(mg ml�1), VSS is the volatile suspended solids at timet (mg ml�1), SLP0 is the initial suspended lipid-phos-

phate concentration (nmol ml�1), SLP is the suspendedlipid-phosphate concentration at time t, k is the kineticconstant (d�1) and t is the time (d).

The slope obtained from these lines is �k/2.303.All statistical analyses and linear regressions were

performed using the standard functions within Micro-soft Excel.

3. Results and discussion

Efficiency of batch reactors processing different typesof urban sludges was measured as volatile solids reduc-tion. Biomass concentration was determined as VSS andSLP. Data obtained during 81 days of incubation areshown in Fig. 1. Data are the mean values of threereactors.

Fig. 1 shows that VSS reduction described a curvewith two different phases: an exponential or log phasefollowed by a stationary phase. The stationary phasewas quickly reached in the reactors fed with secondarysludges, by day 13. In the reactors fed with mixedsludges, the stationary phase was reached by day 29.Finally, in those fed with primary sludges this phasewas reached later, by day 62.

PS contains more readily degradable organic matterthan SS does. As a result, higher total volatile solids

Page 4: Biomass stabilization in the anaerobic digestion of wastewater sludges

y = -0.0227x

R2 = 0.8726

y = -0.0150xR2 = 0.9860

-1.0

-0.8

-0.6

-0.4

-0.2

0.00 9 18 27 35 44 53 62

Time (d)

log(

VSS

/VSS

0)

-1.6

-1.3

-1.0

-0.6

-0.3

0.0

log(

SLP

/SL

P0)

y = -0.0272xR2 = 0.9757

y = -0.0094xR2 = 0.8589

-0.7

-0.6

-0.4

-0.3

-0.1

0.00 9 18 27 35 44 53 62

log(

VSS

/VSS

0)

-1.8

-1.4

-1.1

-0.7

-0.4

0.0

log(

SLP

/SL

P0)

PS

MS

y = -0.2009x

y = -0.0192xR2 = 0.9876

-0.3

-0.2

-0.2

-0.1

0.00 2 4 6 7 9 11 13

log(

VSS

/VSS

0)

-1.6

-1.3

-1.0

-0.6

-0.3

0.0

log(

SLP

/SL

P0)

SS

Fig. 2. Logarithmic representation of the volatile solids reduction(m —) and biomass stabilization (s - - -) of each kind of sludge. Theequation for the regression analysis and the R2 are shown in eachchart. Primary sludge (PS); secondary sludge (SS); mixed sludges (MS).

1182 C. Arnaiz et al. / Bioresource Technology 97 (2006) 1179–1184

destruction efficiency was achieved for PS, 87.2%, whileSS showed 42.6% (Fig. 1). It has become common prac-tice to combine PS and SS for digestion. When this wasdone, destruction of volatile solids in the digested sludgefell between the characteristics of either sludge takenalone (60.9%). While PS showed two typical growthcurves for batch cell culture (from day 0 to 13, and fromday 13 to 81), SS and MS showed just one, probably re-lated to a longer phase of acclimatization for the loweramount of biomass supplied to the reactors fed withPS. It should be mentioned that SS sludge is mainlycomposed of biological sludge. Therefore, PS reactorscontained less biomass.

On the other hand, if biomass concentration asphospholipid concentration is represented against incuba-tion time, it can be seen that curves of biomass reductionor biomass stabilization are not similar to those of VSSreduction (Fig. 1). Whereas the latter diminished con-stantly, in the former a fall followed by an increase ofSLP was observed for reactors fed with primary and mixedsludges. It means that the SLP contents of PS and MSshowed a small phase of exponential growth before thephase of death, even when the reduction in VSS was con-tinuous, showing that VSS (a gross measure) is not a goodestimation of the amount of biomass. Moreover, SLP con-tent decayed much faster in SS than in PS and MS.

Fig. 1 shows that the stationary phase of biomassreduction was reached quickly in the reactors fed withsecondary sludges, by day 7. In the reactors fed withmixed sludges the stationary phase was reached by day29 while, in those fed with primary sludges, this phasewas reached later, by day 48. The higher total biomassreduction efficiency once the stationary phase wasreached was achieved for SS, 96.1%, while PS and MSshowed 91.7% and 83.1%, respectively.

In Fig. 1, it can be observed that the efficiency of re-moval increased with the incubation period when thiswas short. However, at longer incubation times, the per-centage removal did not significantly change, and, there-fore, attempts to operate at longer incubation times toimprove efficiency were useless. On the other hand, itcan be seen that below a certain value of incubationtime, a very small reduction resulted in a drastic reduc-tion of the removal efficiency. In order to ensure correctoperation of the reactors, only the exponential or logphases of curves from Fig. 1 are of interest in industrialdesign and can be expressed by Eqs. (1) and (2).

The volatile solids reduction and the biomass stabil-ization are represented in Fig. 2 by means of Eqs. (1)and (2). The graph for SS in Fig. 2 uses a timescale of7 days for SLP and a timescale of 13 days for VSS,because SLP and VSS were almost all removed veryquickly in SS. In spite of these few points, data obtainedon the 7th day of incubation for SLP and on the 13thday of incubation for VSS in SS are the mean value ofthree reactors. As shown in Fig. 1, the standard devia-

tion of these points was small. Actually, mean ± stan-dard deviation of the three replicate analyses was96.07 ± 0.62 for SLP and 42.60 ± 4.77 for VSS.

Table 2 shows the k values for the volatile suspendedsolids reduction and biomass stabilization for eachsludge type tested, calculated from the straight lines ofFig. 2. It can be seen that these values are higher in allcases for the biomass stabilization process: 0.052 against0.035 d�1 in PS, 0.463 against 0.044 d�1 in SS and 0.063against 0.022 d�1 in MS. These data indicate that vola-tile solids destruction and biomass stabilization are notparallel processes, since the latter was reached quite ear-lier than the former.

The k value, calculated for the biomass stabilization(expressed as phospholipid concentration) in the reac-tors fed with secondary sludges as carbon and energysource, was very high. In this case, the reduction rateof biomass was very quick, with a k of 0.463 d�1,whereas in primary and mixed sludges the values of k

were 0.052 d�1 and 0.063 d�1, respectively. Therefore,

Page 5: Biomass stabilization in the anaerobic digestion of wastewater sludges

Table 2k values (d�1) estimated for the volatile suspended solids reduction andbiomass stabilization of each sludge type

Primary sludge Secondary sludge Mixed sludge

VSS SLP VSS SLP VSS SLP

0.035 0.052 0.044 0.463 0.022 0.063

C. Arnaiz et al. / Bioresource Technology 97 (2006) 1179–1184 1183

the assessed kinetic constants for the decay of the viablebiomass indicated that the viable biomass in SS decayedmuch faster than the viable biomass in PS and MS.

Another objective in measuring lipid-phosphate con-centration was to find the living cell biomass of the totalVSS. Fig. 3 shows VSS against SLP during the incuba-tion period for reactors processing each type of sludge.If the relation VSS SLP�1 is represented for all thereactors (Fig. 4), the straight line of regressiony = 38.3x + 6,436 is obtained (units in lgVSS nmol�1

Pi ),with a coefficient of correlation of 0.742.

An important point of the relationship between totalbiomass amount measured as volatile solids and total

PS

SS

y = 0.0514x + 4.7799

R2 = 0.7964

y = 0.0688x + 8.348

R2 = 0.7757

0

10

20

30

40

50

0 105 210 315 420 525

SS (

mg

mL

-1)

y = 0.0171x + 7.1492

R2 = 0.7946

y = 0.0166x + 11.129

R2 = 0.7682

0

5

10

15

20

0 82 164 246 328 410

SS (

mg

mL

-1)

y = 0.0371x + 6.7961

R2 = 0.9156

y = 0.0395x + 13.011

R2 = 0.8893

0

6

12

18

24

30

0 82 164 246 328 410

SLP (nmol mL-1)

SS (

mg

mL

-1)

MS

Fig. 3. TSS versus SLP (m —) and VSS versus SLP (s - - -) during theincubation period of each sludge type. The equation for the regressionanalysis and the R2 are shown in each chart. Primary sludge (PS);secondary sludge (SS); mixed sludges (MS).

biomass amount measured as lipid-phosphate concen-tration is that was not of y = ax type, as might havebeen expected, but of y = ax + b type (Figs. 3 and 4):when living biomass was completely stabilized, therewere still volatile solids in suspension into the reactors(y-intercept). Thus, both PS and SS consisted of a biode-gradable fraction and an inert fraction, which could notbe biologically degraded. Moreover, the higher SLP/VSS ratio of PS (0.0514) as compared to the SLP/VSSratio of SS (0.00171) indicates that the biodegradablepart of the VSS in PS contained relatively more viablebiomass (�3 times more). It is remarkable that previousworks (Arnaiz et al., 1998; Findlay et al., 1989; Zhangand Bishop, 1994) did not show intersection with they-axis for the conversion factors. Based on data fromthis study, the inert fraction would be represented bythe ordinate at the y-axis of the straight lines shown inFigs. 3 and 4. In SS (biological sludge), dead-end prod-ucts and the rest of the extracellular matrix constitutethis inert fraction. In PS, organic materials such as cel-lulose and others constitute this inert fraction (Eckenfel-der, 1980; Marais, 1984). More recent studies do reportsimilar results to those in this work (Arnaiz et al., 2003)in anaerobic wastewater biofilms.

If the conversion factors are expressed aslgTSS nmol�1

Pi , a comparison between sludge types showsa slope of 68.80 lg nmol�1 for PS, 16.07 lg nmol�1 forSS and 42.76 lg nmol�1 for MS (Fig. 3). These valuesare quite different to those reported previously (Findlayet al., 1989; Zhang and Bishop, 1994) (4.785 and4.81 lgTSS nmol�1

Pi , respectively). It should be mentionedthat, in both studies, conversion factors were obtainedby enrichment cultures of free-living cells from envi-ronmental biofilm and aerobic wastewater biofilmrespectively, so extracellular polymeric substances couldappear at lower levels. When conversion factors for allsludge types were calculated, the relationship between

y = 0.0475x + 10.948

R2 = 0.6679

y = 0.0383x + 6.4364

R2 = 0.742

0

10

20

30

40

50

0 105 210 315 420 525

SLP (nmol mL-1)

SS (

mg

mL

-1)

Fig. 4. TSS versus SLP (m —) and VSS versus SLP (s - - -) during theincubation period for all the reactors tested. The equation for theregression analysis and the R2 are shown in each chart.

Page 6: Biomass stabilization in the anaerobic digestion of wastewater sludges

1184 C. Arnaiz et al. / Bioresource Technology 97 (2006) 1179–1184

SS and SLP was found to be y = 47.5x + 10,948 (unitsin lgTSS nmol�1

Pi ), R2 = 0.668 (Fig. 4).Theoretical ratio between lipid-phosphate concentra-

tion and volatile solids concentration was not calculatedin this work. This theoretical ratio must be obtainedfrom enrichment cultures of free cells detached fromthe biofilms (Lazarova and Manem, 1995). However,selective enrichments involve cell reproduction and onlysome of the total active cells are susceptible to regrowth.In addition, the reproducibility of results depends onincubation conditions, biomass type (free cultures orcells from biofilms) and the choice of nutritive medium.Therefore, theoretical ratio obtained by these methodshas nothing to do with biomass into the reactors studied(Cao and Alaerts, 1995; Arnaiz, 2000; Arnaiz et al.,1998, 2003). However, the similarity between the specificslopes in reactors with PS (0.0688 and 0.0514), SS(0.0166 and 0.0171) and MS (0.0395 and 0.0371)suggests the possibility that these specific slopes arethe theoretical ratio between lipid-phosphate concentra-tion and volatile solids concentration of the specificmicrobial cell mass into the reactors under the experi-mental conditions of this work (Fig. 3). This possibilityhas been suggested elsewhere (Arnaiz et al., 2003). Moredata in this sense are needed for general conclusions.

4. Conclusions

In this study, volatile solids reduction and biomassstabilization in the anaerobic digestion of differenturban sludge types were measured and the kinetic con-stants estimated.

The volatile solids in primary sludge were more bio-degradable than in secondary sludge (87% versus 43%).

Based upon the SLP/VSS ratios, it has to be con-cluded that biodegradable fraction in the VSS containedmore viable biomass in primary sludge than in second-ary sludge.

The assessed kinetic constants for the decay of theviable biomass indicated that the viable biomass in sec-ondary sludge decayed much faster than the viable bio-mass in PS.

Acknowledgements

This work was supported by research Grant No.IN92-D28480461-92 from the Ministry of Educationand Cultures of Spain to the first author.

References

APHA-AWA-WPCF, 1992. Standard Methods for the Examination ofWater and Wastewater. In: Clesceri, L.S., Greenberg, A., Trussell,R. (Eds.), American Public Health Association, 18th ed. Washing-ton, DC, USA, pp. 2–71.

Arnaiz, C., 2000. Depuracion biologica de aguas residuales industri-ales. Desarrollo de tecnologıa con lechos fluidizados. Thesis Doc.,University of Sevilla, Spain.

Arnaiz, C., Ruiz, C., Gomez, E., Garcia, I., Escot, E., Aguilar, E.,Medialdea, J.M., Gutierrez, J.C., Lebrato, J., 1998. Evaluation ofthe effectiveness of materials as anaerobic wastewater treatmentsupports by phospholipid analysis. In: Proceedings of the Interna-tional Specialty Conference on ‘‘Microbial Ecology of Biofilms:Concepts, Tools and Applications’’, Lake Bluff, Illinois, USA,pp. 294–300.

Arnaiz, C., Buffiere, P., Elmaleh, S., Lebrato, J., Moletta, R., 2003.Anaerobic digestion of dairy wastewater by inverse fluidization: theinverse fluidized bed and the inverse turbulent bed reactors.Environ. Technol. 24, 1431–1443.

Cao, W.S., Alaerts, G.J., 1995. Influence of reactor type and shearstress on aerobic biofilm morphology, population and kinetics.Water Res. 29, 107–118.

Eckenfelder, W.W., 1980. Principles of Water Quality Management.CBI Publishing Co., Boston, USA.

Findlay, R.H., King, M.G., Watling, J., 1989. Efficacy of phospholipidanalysis in determining microbial biomass in sediments. Appl.Environ. Microbiol. 55, 2888–2893.

Lazarova, V., Manem, J., 1995. Biofilm characterization and activityanalysis in water and wastewater treatment. Water Res. 29, 2227–2245.

Lebrato, J., 1990. Obtencion de biogas a partir de residuos organicosurbanos: Experiencias en lecho fluidizado, Ph.D. Thesis, Universityof Sevilla, Spain.

Marais, G.R., 1984. Theory, Design and Operation of NutrientRemoval Activated Sludge Processes, Water Research Commis-sion, Pretoria, South Africa.

Singh, A., Kuhad, R.C., Sahai, V., Ghosh, P., 1994. Evaluation ofbiomass. Adv. Biochem. Eng. Biotechnol. 51, 47–70.

Stewart, J.M., Bhattacharya, S.K., Madura, R.L., Mason, S.H.,Schonberg, J.C., 1995. Anaerobic treatability of selected organictoxicants in petrochemical wastes. Water Res. 29, 2730–2738.

Water Environment Federation & American Society of Civil Engi-neers, 1992a. Sludge handling and concentration, in Design ofMunicipal Wastewater Treatment Plants, ed. by WEF/ASCE,pp. 1107–1253.

Water Environment Federation & American Society of Civil Engi-neers, 1992b. Sludge stabilization, in Design of MunicipalWastewater Treatment Plants, ed. by WEF/ASCE, pp. 1255–1382.

White, D.C., Bobbie, R.J., Herron, J.S., King, J.D., Morrison, S.,1979. Biochemical measurements of microbial mass and activityfrom environmental samples. In: Costerton, J.W. (Ed.), NativeAquatic Bacteria: Enumeration, Activity and Ecology. ASTMSpec. Tech. Publ., University of Calgary, MD, USA.

Zhang, T.C., Bishop, P.L., 1994. Density, porosity and pore structureof biofilms. Water Res. 28, 2267–2277.