study of the operational conditions for anaerobic digestion of urban solid wastes

11
Study of the operational conditions for anaerobic digestion of urban solid wastes Edgar Fernando Castillo M. * , Diego Edison Cristancho, Victor Arellano A. Centro de Estudios e Investigaciones Ambientales, Universidad Industrial de Santander, Calle 9a Carrera 27, Aptdo. Aereo 678, Bucaramanga, Colombia Accepted 7 June 2005 Available online 29 November 2005 Abstract This paper describes an experimental evaluation of anaerobic digestion technology as an option for the management of organic solid waste in developing countries. As raw material, a real and heterogeneous organic waste from urban solid wastes was used. In the first experimental phase, seed selection was achieved through an evaluation of three different anaerobic sludges coming from wastewater treatment plants. The methanization potential of these sludges was assessed in three different batch digesters of 500 mL, at two temperature levels. The results showed that by increasing the temperature to 15 °C above room temperature, the methane production increases to three times. So, the best results were obtained in the digester fed with a mixed sludge, working at mesophilic conditions (38–40 °C). Then, this selected seed was used at the next experimental phase, testing at different digestion times (DT) of 25, 20 and 18 days in a bigger batch digester of 20 L with a reaction volume of 13 L. The conversion rates were reg- istered at the lowest DT (18 days), reaching 44.9 L/kg 1 of wet waste day 1 . Moreover, DT also has a strong influence over COD removal, because there is a direct relationship between solids removal inside the reactor and DT. Ó 2005 Elsevier Ltd. All rights reserved. 1. Introduction The current increase in quantities of solid waste at national and international levels is producing unfavor- able environmental effects. In particular, among devel- oping countries this situation creates the necessity to look for alternative solid wastes treatment options that can provide benefits over solid waste disposal. Anaero- bic digestion (AD), defined as ‘‘utilization of microor- ganisms, in anoxigenic conditions, to stabilize the organic matter by transforming it into methane and other inorganic products including carbon dioxide’’ (Kiely, 1999), could be a suitable choice for the biode- gradable fraction of urban solid wastes (USW) in devel- oping countries. With such implementation, the risk of producing important microbiological impacts will de- cline due to its anaerobic nature, since the sludge treat- ment would be correctly achieved. Besides, with this technology two residual effluents are produced: biogas (mainly methane and carbon dioxide) which can be used as an energy source, and a liquid effluent which could be used as a soil conditioner due to its physicochemical properties (Flotats et al., 1997). In addition, it is impor- tant to bear in mind that in contrast with USW from industrialized countries, generally USW generated in developing economies is characterized by a high organic content (mainly from food disposal) and relatively low plastic and metal contents. Nowadays, the main techno- logical option used in Colombia for the management of USW is the sanitary landfill. If AD is used as a prelimin- ary or local treatment, the required landfill volume will decrease, the useful lifespan of the landfill will also be greater and an important amount of energy would be generated from the waste stream. Currently, AD is used in many treatment processes such as wastewater, industrial wastes (breweries, wine, 0956-053X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2005.06.003 * Corresponding author. Tel./fax: +57 76 459919. E-mail addresses: [email protected], [email protected] (E.F. Castillo M.). www.elsevier.com/locate/wasman Waste Management 26 (2006) 546–556

Upload: edgar-fernando-castillo-m

Post on 02-Jul-2016

213 views

Category:

Documents


1 download

TRANSCRIPT

www.elsevier.com/locate/wasman

Waste Management 26 (2006) 546–556

Study of the operational conditions for anaerobicdigestion of urban solid wastes

Edgar Fernando Castillo M. *, Diego Edison Cristancho, Victor Arellano A.

Centro de Estudios e Investigaciones Ambientales, Universidad Industrial de Santander, Calle 9a Carrera 27, Aptdo. Aereo 678, Bucaramanga, Colombia

Accepted 7 June 2005Available online 29 November 2005

Abstract

This paper describes an experimental evaluation of anaerobic digestion technology as an option for the management of organicsolid waste in developing countries. As raw material, a real and heterogeneous organic waste from urban solid wastes was used. Inthe first experimental phase, seed selection was achieved through an evaluation of three different anaerobic sludges coming fromwastewater treatment plants. The methanization potential of these sludges was assessed in three different batch digesters of500 mL, at two temperature levels. The results showed that by increasing the temperature to 15 �C above room temperature, themethane production increases to three times. So, the best results were obtained in the digester fed with a mixed sludge, workingat mesophilic conditions (38–40 �C). Then, this selected seed was used at the next experimental phase, testing at different digestiontimes (DT) of 25, 20 and 18 days in a bigger batch digester of 20 L with a reaction volume of 13 L. The conversion rates were reg-istered at the lowest DT (18 days), reaching 44.9 L/kg�1 of wet waste day�1. Moreover, DT also has a strong influence over CODremoval, because there is a direct relationship between solids removal inside the reactor and DT.� 2005 Elsevier Ltd. All rights reserved.

1. Introduction

The current increase in quantities of solid waste atnational and international levels is producing unfavor-able environmental effects. In particular, among devel-oping countries this situation creates the necessity tolook for alternative solid wastes treatment options thatcan provide benefits over solid waste disposal. Anaero-bic digestion (AD), defined as ‘‘utilization of microor-ganisms, in anoxigenic conditions, to stabilize theorganic matter by transforming it into methane andother inorganic products including carbon dioxide’’(Kiely, 1999), could be a suitable choice for the biode-gradable fraction of urban solid wastes (USW) in devel-oping countries. With such implementation, the risk ofproducing important microbiological impacts will de-

0956-053X/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.wasman.2005.06.003

* Corresponding author. Tel./fax: +57 76 459919.E-mail addresses: [email protected], [email protected] (E.F.

Castillo M.).

cline due to its anaerobic nature, since the sludge treat-ment would be correctly achieved. Besides, with thistechnology two residual effluents are produced: biogas(mainly methane and carbon dioxide) which can be usedas an energy source, and a liquid effluent which could beused as a soil conditioner due to its physicochemicalproperties (Flotats et al., 1997). In addition, it is impor-tant to bear in mind that in contrast with USW fromindustrialized countries, generally USW generated indeveloping economies is characterized by a high organiccontent (mainly from food disposal) and relatively lowplastic and metal contents. Nowadays, the main techno-logical option used in Colombia for the management ofUSW is the sanitary landfill. If AD is used as a prelimin-ary or local treatment, the required landfill volume willdecrease, the useful lifespan of the landfill will also begreater and an important amount of energy would begenerated from the waste stream.

Currently, AD is used in many treatment processessuch as wastewater, industrial wastes (breweries, wine,

Table 1Physicochemical characterization of the substrate

Parameters Units Value

Particle size mm 4–6Real density kg/m3 1052Bulk density kg/m3 864Total solids kg/m3 168.3

% (p/p) 16.0–18.0Volatile solids kg/m3 158.8

% (p/p) 15.1% ST 94.4

Total organic carbon % (p/p) 39.0–48.0Total nitrogen % (p/p) 0.35–1.19Potassium % (p/p) 2.46Phosphorous % (p/p) 0.22C/N ratio 32.77C/K ratio 15.85C/P ratio 177.27

Source: CEIAM-UIS (2003).

Table 2General composition of substrate

Component %w/w dry

Easily biodegradable material 81.2Celullose 6.1Hemicelullose 2.73Lignin 4.95Ash 5.02

Source: CICELPA-UIS, (2003).

E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556 547

milk, feed and chemical industries), agricultural wastes(piggeries, poultry farming, bovine culture, farm wastesand harvest products) and urban wastes (sewage, and or-ganic urban solid wastes (OUSW)). In fact, it is possibleto say that AD is a well-known technology and its maincharacteristics and primary kinetic pathways are defined.However, in most cases it is necessary to know the per-formance of the particular seed–substrate–environmentcombinations. Because of that, many papers can befound in the literature about this subject, as the com-bined study of primary sludges with brewery industrywastes (yeast) combined with abattoir wastes (Sugrueet al., 1992), for instance. Some of this research has beenfinanced by the EU (Wheatley, 1991; Ferranti et al.,1987). According to Wellinger et al. (1992), there aremore than 500 anaerobic digestion plants in Europe thatoperate on agricultural farms.

However, in a parallel way, in Colombia we can findonly a few studies, such as the work by Diaz and Espitia(2001) about anaerobic systems for the treatment ofbrewery wastes and the work by Ramirez et al. (2001)for the start-up of an anaerobic reactor. Since relatedstudies on the treatment of OUSW have not been re-ported in Colombia, research specifically on this topicis considered useful. Consequently, the goal of the pres-ent investigation is to illustrate the behavior of an anaer-obic system for the treatment of OUSW and to acquireknowledge about the operating conditions of these sys-tems in order to be able to scale up the technology forindustrial treatment of OUSW in small and medium sizecities in Colombia.

Therefore, in this paper the operating conditions(digestion time (DT), temperature, pH, organic matterload) for the single-phase anaerobic digestion processof OUSW were preliminarily analyzed. For this pur-pose, sludges from the Wastewater Treatment Plant at‘‘Rıo Frıo’’ (Giron, Santander) and the anaerobic biodi-gester for the treatment of pig manure (Mesa de LosSantos, Santander) were considered as seed. The mainintrinsic variables used as performance criteria were spe-cific methane production, total solids destruction andthe volumetric composition of the biogas.

2. Materials and methods

As raw material for the evaluated AD process, a realand heterogeneous material from USW produced in thecity of Bucaramanga, Colombia was used. For charac-terization purposes, a set of four random samples wasmixed and evaluated. The organic fraction of USW(OFUSW) sample contained a non-scheduled mix ofvegetable and forestry wastes. A size reduction stepwas necessary to facilitate the handling of the material.Tables 1 and 2 summarize the main physical, chemicaland biodegradability characteristics of this material.

The experiments began with Phase I, selection of seedfor a high solids concentration AD process, using onlythe specific methane production as the primary selectionparameter. Other conditions taken into account were thestability of the pH, the required time for stable biogasproduction and the methane concentration in the bio-gas. These tests were carried out in three small batchdigesters of 500 mL, providing heat and agitationthrough a water bath with a serpentine (controlled bya thermostat), in which a mineral oil flow was heatedby a MLW Model U2c, and three magnetic agitators(Fig. 1). All reactors were fed on a daily basis, measur-ing daily pH and methane production. The compositionof the biogas was measured on a weekly basis. Totalalkalinity and volatile fatty acid concentrations werealso measured in order to estimate the progress in the ki-netic pathway of the AD process. To better understandthe experimental runs, the systems denomination for allreactors is summarized in Table 3.

The tests were carried out for three different types ofseed, each one fed at each batch reactor. The first reac-tor was fed with sludge from reactor UASB No. 2Wastewater Treatment Plant of ‘‘Rıo Frıo’’ (Giron,Santander, Colombia), named (PTAR). The secondreactor was fed with sludge from an anaerobic biodi-gester from a pig manure treatment process (Mesa delos Santos, Santander, Colombia), named (PIG) and

Fig. 1. Equipment used for seed selection.

Table 3Denomination of the digesters and their respective seed

Batch reactor Feed type

BID1 Sludge of reactor UASBBID2 Anaerobic biodigester pig manure feededBID3 Mixture 1:1 (v/v) of above sludges

548 E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556

finally the third reactor was inoculated with a mixture1:1 (v/v) of the sludges mentioned above, named (MIX).In order to quantify the methane production, gas metersof 1 L made in glass operating by liquid displacement(Mariotte bottle) were used. A summary of the experi-mental procedure followed is shown in Tables 4 and 5.

Table 4Variables for seed selection experimental design

Variable Values

Temperature T1 (20–30 �C)T2 (38–40 �C)

Solids percentage 8% y 12%

Seed type PTAR, PIG y MIX

Table 5Experimental design for seed selection

Experiment t (days) T (�C) %TS Seed

1 1–95 T1 8 PTAR2 95–120 T2 8 PTAR3 120–145 T2 12 PTAR4 1–95 T1 8 PIG5 95–120 T2 8 PIG6 120–145 T2 12 PIG7 1–95 T1 8 MIX8 95–120 T2 8 MIX9 120–145 T2 12 MIX

The pH adjustment in the anaerobic activity range forall reactors was made using sodium bicarbonate.

Once the seed selection was completed, the next stepwas the starting-up of a plastic made 20-L batch digester(see Fig. 1) with 13 L of reaction volume (Phase II). Thisreactor was equipped with an acrylic spy-hole, a metalcap holding the agitation and heating systems, two1.25-cm biogas evacuation valves, and various othervalves for feeding, discharging and sampling. Heat wassupplied through a 4-m stainless steel serpentine, whichcontained hot oil heated by a bath controlled with athermostat (MLW, Model U2c). Stirring was providedby a worm-gear and a helicoidal agitator moved by a1/2-HP electric motor. The usage of a sweeper on thebottom and a pulley system to decrease the speed to60 rpm was necessary. The agitation system was alwaysoperated in 2-h intermittent events controlled by a timer.

The daily operational procedure was as follows: Apre-defined volume of 18% w/w total solids concentra-tion solution was fed to the batch reactor. Then, thesame volume was extracted from the bottom of the reac-tor, without considering its density. The DT was calcu-lated as the reactor operation volume to daily feedvolume ratio. In the 20-L batch digester, three differentDTs were tested, evaluating temperature, total solidsconcentration of the resulting sludge and the methaneproduction. In the same way, total and volatile solids re-moval for each DT was evaluated.

Because of stirring problems associated with the for-mation of aggregates inside the reactor, the agitator be-came a very important component of the digester, due tothe process requirement for maintaining a biomass sus-pended operation for improving mass and heat ex-changes between microorganism and substrate. Thesestirring problems were detected in the operation of thesmall batch digesters supplied with magnetic agitators.As a result, a new type of agitator (a helicoidal one)had to be designed. The helicoidal agitator moved thereactive mass up near the digester wall and the worm-gear moved the reactive mass down near to the axle-tree.In addition, a bottom sweeper was needed to agitate thesettled sludge in the bottom of the digester, and surfacepalettes were set to force the floating waste down. Theexperimental assembly constructed for the 20-L batchdigester is shown in Fig. 2.

This system was operated at 35 �C. Feeding particlesize was reduced with no extra water addition keepingthe total solids concentration close to 18% (w/w). Threedifferent DTs of 25, 21 and 18 days were tested. Theexperimental runs at Phase II were carried out in asequentially scheduled routine beginning with the 25-day DT period, followed by the 21-day DT period andfinally with the 18-day DT period. Before evaluationof the 18-day DT, a tentative 17-day DT period wastried, but at the end of the first 8-day period the systemshowed an intensive acidification process, due probably

Fig. 2. 20-L batch digester.

E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556 549

to the high increase in the organic loading into the reac-tor. This fact explains why the reactor had to be inocu-lated once again at this time.

In all experimental runs (both Phases I and II), thebiogas composition was measured with a biogas ana-lyzer BACHARACH model GA-94. This equipmenthas an infrared cell of dual wavelength and an electro-chemical cell.

Table 6Operational conditions followed for seed selection experiments

Temperature Total solids compositionof the feeding (w/w)

Day 1–95: room temperature(max = 29 �C; min = 20 �C)

Day 1–120: 8%

Day 95–145 (max = 40 �C; min= 38 �C)

Day 120–145: 12%

3. Results and discussion

As it was explained in the previous section, the re-search was divided into two parts: Phase I: selection ofthe appropriate seed for the solid waste AD and PhaseII: study of the 20-L batch digester behavior for severalDTs.

SPECIFIC METHAN(Room t

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0 5 10 15 20

Met

han

e p

rod

uct

ion

(m

3 CH 4

/m3

reac

tor/

day

)

BD1 (PTAR)

Fig. 3. Methane production rate of biodigesters

3.1. Phase I. Seed selection

Due to the absence of a regional solid waste treat-ment plant based on an AD system, it was necessaryto determine which of the different systems currentlyused in Colombia for wastewater treatment could pro-vide the most efficient seed, when operating with theOFUSW as substrate. The main selection parameterwas the specific production of methane, as well as thecomposition of the produced biogas. Other criteria ta-ken into account were the stability of the pH and thequality of the digested waste produced. Using the feed-ing proportions of sludge, substrate and water suggestedby Griffin et al. (1998), the batch reactors were startedup and set to room temperature. The system was oper-ated during 145 days at the conditions shown in Table 6.

One of the main goals of this research was to deter-mine the performance of the AD process when operatedat different total solids concentrations. For this reason,it was very important to evaluate how sensitive the seedwas to several organic loading conditions.

Figs. 3–11 show the pH variation and the specificmethane production rate reached at room temperatureconditions (Day 1–Day 95) and at high temperaturecondition (Day 95–Day 145) for all three reactors. Vol-umetric concentrations of CO2 and methane in the bio-

E PRODUCTION RATEemperature)

25 30 35 40 45Days

BD2 (PIG) BD3 (MIX)

operating at room temperature (Phase I).

pH vs TIME(Room Temperature)

5.0

5.5

6.0

6.5

7.0

7.5

8.0

100 105 110 115 120 125 130 135 140 145Days

pH

BD1 (PTAR) BD2 (PIG) BD3 (MIX)

Fig. 4. Variation of pH in digesters operating at room temperature.

SPECIFIC METHANE PRODUCTION VS TIME (with heating)

0.00

0.30

0.60

0.90

1.20

1.50

1.80

2.10

125 127 129 131 133 135 137 139 141 143 145Day

m3

CH

4/(m

3·d

ay)

BD1 (PTAR) BD2 (PIG) BD3 (MIX)

Fig. 5. Methane production rate of the small digesters operating at 40 �C.

pH vs TIME (without heating)

5.0

5.5

6.0

6.5

7.0

7.5

8.0

100 105 110 115 120 125 130 135 140 145Days

pH

BD1 (PTAR) BD2 (PIG) BD3 (MIX)

Fig. 6. Variation of pH in the small batch digesters operating at 40 �C.

550 E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556

gas are also registered. As additional information, thetotal alkalinity and volatile fatty acids concentrationare shown as well, only for the period when the total sol-ids concentration of the feeding was increased (begin-ning Day 121).

At room temperature conditions, beginning Day 35the trend in all systems was an increase in methane pro-duction (Fig. 3). Considering only the average specificmethane production rate, the best result was reachedby the MIX reactor (0.21), followed by the PIG reactor

BIOGAS COMPOSITION-BD1

01020304050607080

0 20 40 60 80 100 120 140Days

Per

cen

tag

e

%CH4 %CO2

Fig. 7. Biogas composition with PTAR sludge.

BIOGAS COMSPOSITION - BD2

0

10

20

30

40

50

60

70

80

0 20 40 60 80 100 120 140Days

Per

cen

tag

e

%CH4

%CO2

Fig. 8. Biogas composition with PIG sludge.

BIOGAS COMPOSITION-BD3

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

0 20 40 60 80 100 120 140Days

Per

cen

tag

e

% CH4

% CO2

Fig. 9. Biogas composition with MIX sludge.

E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556 551

(0.18) and finally by the PTAR reactor (0.13). Thisbehavior agrees with the pH profile, generated due tothe addition of sodium bicarbonate for improving alka-linity capacity (Fig. 4). Considering Fig. 5, it is possibleto establish that when the batch digesters were operatedat a high temperature condition, the specific methaneproduction rate increased three times compared to the

corresponding value at room temperature. The averagespecific methane production rate for both the MIXand the PTAR reactors reached 1.20, while the PIGreactor only reached a value of 1.11. These results agreewell with general references about the AD process(Argelier et al., 1998; Beccari et al., 1993; Bouallaguiet al., 2004).

Fig. 10. Total volatile fatty acids content in all small digesters.

Fig. 11. Total alkalinity of small batch digesters operating at 40 �C.

552 E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556

It is important to note that beginning Day 121, thetotal solids concentration of the substrate was increasedfrom 8% to 12% w/w. In order to determine the totalalkalinity capacity of the system under higher organicloadings and its influence on the pH of the reactingmass; at this point the addition of sodium bicarbonatewas discontinued (see Fig. 6). The resulting profileshowed that there was a great variation in pH, the netconcentration of fatty volatile acids increased (see Fig.10) while the total alkalinity decreased (see Fig. 11),but it was possible to operate all small reactors understable conditions. This result became very importantfrom an economic point of view.

In general, although the methane composition in thebiogas decreased in all three systems when operated athigh temperature, the total flow of generated biogas in-creased. For the reactor fed with PTAR sludge, a finalmethane concentration of 44.8% was observed, while

the maximum concentration peak of 70.4% was reachedon Day 82 (Fig. 7). For the PIG system, a maximummethane composition of 65.1% was obtained on theDay 82, but then it dropped to a final value of 44.5%(Fig. 8). For the MIX system, a maximum methane con-tent of 63.3% was reached on Day 49, and a final meth-ane value of 53.6% was registered (Fig. 9).

According to the curves in Fig. 6, the MIX systemgenerated minor pH fluctuations, showing a greater sta-bility than the digesters fed with the other two seeds.This fact was supported when comparing the statisticaldata variances for each reactor. The corresponding va-lue for the MIX reactor was 0.18, while for the PTARreactor it was 0.44. It is also important to note thatthe experimental runs suggest that a biochemical equi-librium of the AD process at high temperature can bereached when the total alkalinity range lies between4500 and 5500 mg CaCO3/L, and the volatile fatty acidsconcentration is close to 2000 mg acetic acid/L.

As a final conclusion from experimental Phase I, itcan be stated that the MIX system offers the best pH sta-bility and also has the best specific methane productionrate, so it was selected as seed for the next experimentalruns.

3.2. Study of the 20-L biodigester performance at several

DTs

The second experimental section developed in this re-search was focused on the behavior of a 20-L anaerobicbatch digester with a working volume of 13 L. The reac-tor was inoculated with a mixture of 50% v/v of PIG andPTAR sludges, according to the previous results.

At startup of the anaerobic bio-digester, an acidi-fication process was noted and it was necessary tore-inoculate the reactor and to change the feeding ratefrom 24-h intervals to 12-h intervals. Only when the sys-tem reached a stable range of pH and volatile fatty acidsconcentration, the bio-digester operating evaluation atseveral DTs began. The results for specific productionrate for methane, the pH variation and the methane vol-umetric composition in the biogas at several DTs areshown in Figs. 12–14.

The experimental results showed that after daily timefeeding of the reactor, a great pH variation occurs andthe concentration of methane in the biogas decreases.This effect varies in intensity according to the differentDTs tested, probably caused by an acidification process.To avoid acidification, the last DT tested was evaluatedby feeding twice a day, decreasing the shock caused by agreat amount of fresh substrate in the bio-digester. Inthis way, it was possible to obtain a more constant meth-ane production rate and new stable conditions were ob-tained in less time.

As shown in Figs. 12–14, region DT corresponds tothe 17-day DT period evaluation when the reactor was

pH variation

5.5

6.0

6.5

7.0

7.5

8.0

135 140 145 150 155 160 165 170 175 180 185 190 195 200Day

pH

DT = 21 days DT = 18 daysDT = 25 days D T = 17 daysA

Fig. 12. pH variation during 20-L digester evaluation.

E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556 553

fed only once a day. A rapid pH decrease was registeredwhile acidification process progressed. In the 18-day DTperiod evaluation, although the pH was lower than thepH values for the other DTs, the system showed more

DT = 2DT = 25 days

Met

han

e p

rod

uct

ion

(m

3 CH

4/m

3 rea

cto

r/d

ay)

D1160155150145140135

3.0

2.5

2.0

1.5

1.0

0.5

Fig. 13. Methane production rate dur

30

35

40

45

50

55

60

65

70

75

80

135 140 145 150 155 160 165

TRH = 25 days TRH = 21

Met

han

e p

erce

nta

ge

days

BIOGAS CO

Fig. 14. Biogas composition durin

stability in spite of the increase in the organic load tothe reactor.

Fig. 13 shows the methane production rate at differ-ent DTs. Only a small increase in the methane produc-tion can be observed when compared with the 25-dayDT experiment and the 21-day DT experiment. But,when the 18-day DT period was tested, due to the pHstability, a more significant increase in the methane pro-duction rate can be seen.

Fig. 14 shows the volumetric composition of thebiogas for the DTs evaluated. The reason why onlythe methane volumetric composition is shown, is dueto the fact that methane and carbon dioxide accountfor 99% of the total volumetric fraction of the biogas.During the evaluation of the first two DTs, severalpeaks in the methane volumetric composition can beobserved, caused probably by the great pH irregularityduring the period when these operational conditionswere maintained. In contrast, when the reactor oper-

DT = 18 daysADT1 days

ays20019519018518017517065

ing the 20-L digester evaluation.

170 175 180 185 190 195 200

days TRHA TRH = 18 daysMPOSITION

g 20-L digester evaluation.

4 0

5 0

6 0

7 0

8 0

9 0

135 140 145 150 155 160 165 170 175 180 185 190 195 200

ST SV

TR H = 25 days TR H = 21 days TR H A TR H = 18 days

Rem

ova

l per

cen

tag

e

days

TOTAL AND VOLATILE SOLIDS REMOVAL

Fig. 15. Total and volatile solids removal during 20-L digester evaluation.

Table 8Nutrients content of the liquid effluent (DT = 21 days)

Parameter Value

Total organic carbon 16.00%Total nitrogen 1.95%Potassium 3.63%Phosphorus 1.40%C/N 8.2C/K 4.4C/P 11.4

554 E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556

ated with an 18-day DT period and fed twice a day, agreat stability in the biogas composition was obtained.

Fig. 15 shows the total and volatile solids removalsfor the DTs tested. A direct relationship between DTand the removal percentage of both, total and volatilesolids, can be observed. During the acidifying periodDT, no solid testing was carried out.

The values for the volatile fatty acids concentrationand the total alkalinity were maintained constant, ex-cept when the reactor became acidified. The standardvalue of fatty acids concentration was 11.000 mg aceticacid/L, showing peaks of 13.000 mg acetic acid/L. Thesevalues are similar to the corresponding ones publishedby Kayhanian and Rich (Kayhanian and Rich, 1995)whose investigation was conducted under similar oper-ating conditions. Alkalinity was kept between 6.000and 7.000 mg CaCO3/L.

Table 7 summarizes the specific methane produc-tion rate, the methane volumetric composition reachedin the biogas, as well as the total and volatile solidsconcentration values once each system became stable.Other parameters included in Table 7 are the totalamount of methane produced per unit mass of wet so-

Table 7Comparison of the 20-L batch reactor performance at different DTs

Parameter Units

Biogas

Specific methane productionm3CH4

m3reactor�day

Methane production by wet waste massL CH4

kg waste

Methane volumetric composition % volumeGross energy produced kJ

day

Gross energy produced by waste mass unit kJkg waste

Effluent

Solids concentration in effluent g/LTotal solids removal %Volatile solids in effluent g/LVolatile solids removal %

lid waste and the gross energy produced in the bio-di-gester. The term gross energy refers to bio-gasgenerated energy potential not considering thesystems energy consumption. The energy lost by con-vection was estimated to be close to 300 kJ/day(average).

Finally, Table 8 shows the nutrient contents of thedigested effluent from the 20-L reactor operating at aDT of 21-days. This analysis was carried out for eval-uating the potential of this stream as a soilconditioner.

DT

25 days 21 days 18 days

1.429 1.655 2.62333.8 33.1 44.960 57 55

692.0 843.6 1385.71259.1 1297.9 1824.6

60.7 62.1 69.063.6 62.1 59.336.0 40.7 51.577.1 74.1 67.4

E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556 555

4. Conclusions

Considering the results obtained in the 20-L reactorand also considering the goals of this research, severalconclusions can be reached:

It is possible to maintain a stable anaerobic processfor the OFUSW treatment with an important energygeneration using a mixture of sludges from conventionalwastewater treatment plants in Colombia. Variables ofpH variation, total solids in the feed, specific methaneproduction rate and volumetric composition of the bio-gas can be controlled with no major constraints.

It is feasible to obtain an important fixing carbon asmethane from the OFUSW anaerobic treatment inColombia, creating a great potential for developing en-ergy producing systems at a local and regional scale inthe country, in addition to the significant reductions inthe quantities of waste disposed in sanitary landfills.

The best operational conditions determined in the 20-L reactor were: an 18-day DT period, temperature of35 �C, 18% w/w of total solids concentration in feedingsubstrate and feeding frequency of twice per day. ThepH value reached at steady state was close to 7 andthe specific methane production rate was close to2.6 l day�1 reactor�1.

The methane production rate per unit weight of wetwaste was 35% higher when operating with an 18-dayDT. In general and according to well-known references,the methane production by waste mass (wet, total solidor volatile solids) is normally inversely proportional tothe DT. This means that when the amount of substratetreated increases (the DT is lowered), the volume of bio-gas obtained from a unit of mass substrate decreases aswell. This behavior can be easily seen in the firsts twoDTs tested, which were fed once a day. But accordingto the experimental results for the 18-day DT period,feeding the reactor twice a day allowed higher methano-genic microorganism activity. Feeding twice in a day,the pH decrease between feedings was eliminated, andthe pH–DT relationship described earlier cannot be ap-plied. Moreover, the difference in solids removal (espe-cially volatile solids) is not as noticeable as the pHdifferences. This occurs due to the fact that the main sol-ids removal takes place during the hydrolysis stage andthe bacteria in charge of the hydrolysis process are notinfluenced by the pH as the methanogenic bacteria are.

According to a carbon balance, it can be stated thatmost of the carbon removed from the substrate duringthe 18-day DT (Tables 1, 2, 7, and 8), was released asgas in the form of methane and carbon dioxide (approx-imately 81% of the organic carbon that is fed to the di-gester). This removal allows an important increase in theproportion ratios of the other nutrients. This also indi-cates that most of the carbon content carried in theliquid waste effluent has low or moderate bio-degrada-tion capacity. Moreover, the total and volatile solids re-

moval shows that the digested matter, especially theorganic matter treated at a greater DT, became a so-called ‘‘stable material’’ since the ratio between volatileand total solids is close to 0.6, a value limit for a bio-solid to be stable.

The main characteristic of the effluent stream gener-ated during this research is the high moisture content(greater than 90%). This property constrains seriouslyits use as soil conditioner. These effluents produced inan AD process may be joined in a compost process bymixing them with the biodegradable portion of USW(urban solid wastes), as it was done in Italy (Tchobanog-lous et al., 1994). However, its composition characteris-tics do not surpass the minimum NPK contentsnecessary for fertilizers; therefore, its application is re-stricted to organic soil conditioning. The direct use ofthese effluents on crops is not possible until this stepof the research, mainly because a complete micro-biological characterization must be done.

Acknowledgements

This work was developed thanks to a research grantfrom COLCIENCIAS (Colombia) and the academicsupport of CEIAM-UIS (Centro de Estudios e Investi-gaciones Ambientales – Universidad Industrial deSantander).

References

Argelier, S., Delgenes, J.Ph., Moletta, R., 1998. Design of acidogenicreactors for the anaerobic treatment of the organic fraction of solidfood waste. Bioprocess Engineering 18, 309–315.

Beccari, M., Longo, G., Majone, M., Rolle, E., Scarinci, A., 1993.Modeling of pretreatment and acidogenic fermentation of theorganic fraction of municipal solid wastes. Water. Sci. Techol. 27(2), 193–200.

Bouallagui, H., Touhami, Y., Cheikh, R.B., Hamdi, M., 2004.Bioreactor performance in anaerobic digestion of fruit andvegetable waste. Process Biochemistry.

Diaz, B., Espitia, V.S., 2001. Caracterizacion microbiologica yfisicoquımica de lodos utilizados en plantas de tratamientoanaerobias de la industria cervecera. Proyecto COLCIENCIAS-Universidad Nacional de Colombia. [Physicochemical and micro-biological characterisation of sludges from brewery waste watertreatment plants. Project developed in join by Colciencias-NationalUniversity of Colombia].

Ferranti, M., Ferrero, L., Hermit, P., 1987. Anaerobic Digestion:Results of Research and Demonstration Projects. Elsevier AppliedScience, Essex, England.

Flotats, X., Campos, E., y Bonmati, A., 1997. Aprovechamientoenergetico de residuos ganaderos. Curs d�Enginyeria Ambiental:Aprofitament enegetic de residus organics (3�: 1997: Lleida).Memorias. 1-21. [Energetic profit from livestock farming, 3�:1997: Lleida. Memories:1-21].

Griffin, M.E., Mcmahon, K.D., Mackie, R.I., Raskin, L., 1998.Methanogenic population dynamics during start-up of anaerobicdigesters treating municipal solid waste and biosolids. Biotechnol-ogy and Bioengineering 57, 342–354.

556 E.F. Castillo M. et al. / Waste Management 26 (2006) 546–556

Kayhanian, M., Rich, D., 1995. Pilot-scale high solids thermophilicanaerobic digestion of municipal solid waste with an emphasis onnutrient requirements. Biomass and Bioenergy 8 (6), 433–444.

Kiely, G., 1999. Ingenierıa Ambiental. Fundamentos, entornos,tecnologıas y sistemas de gestion (Environemental Engineering.Fundamentals, Technologies and Management Systems).McGraw-Hill, Madrid, pp. 870–872.

Ramirez, L.F., Molina, F., Monroy, F., Valencia, N., 2001. Optimi-zacion de la etapa de arranque de reactores anaerobios mediante elmejoramiento de la calidad de semillas en condiciones dinamicas deoperacion. Proyecto COLCIENCIAS-Universidad de Antioquia-Universidad del Valle. [Startup optimisation of anaerobic reactorsby improving quality seed under dynamic operation conditions.

Research project developed in join by Colciencias-Antioquia�sUniversity-Valle�s University (Colombia)].

Sugrue, K. Kiely, G., Mckeogh, E., 1992. A pilot study of a mix ofmunicipal and industrial sludges. In: Proceedings of IAWQConference, Washington, DC.

Tchobanoglous G. Theisen H. Virgil S. 1994. Gestion Integral deResiduos Solidos. Mc Graw-Hill, Madrid, pp. 58, 160, 407, 408.

Wellinger, Wyder, A.K., Metzler, A.E., 1992. Kompogas – a newsystem for the anaerobic treatment of source separated waste. In:Proceedings of International Symposium on Anaerobic Digestionof Solid Waste, Venice.

Wheatley, A., 1991. Anaerobic Digestion: A Waste TreatmentTechnology. SCI and Elsevier Applied Science, Essex, UK.