ARTICLE IN PRESS

Available at www.sciencedirect.com

WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 3 7 9 – 1 3 8 7

0043-1354/$ - see frodoi:10.1016/j.watres

�Corresponding auE-mail address:

journal homepage: www.elsevier.com/locate/watres

Anaerobic biodegradability and treatment of grey water inupflow anaerobic sludge blanket (UASB) reactor

Tarek A. Elmitwallia,�, Ralf Otterpohlb

aDepartment of Civil Engineering, Benha High Institute of Technology, Benha University, P.O. Box 13512, Benha El-Gedida, Benha, EgyptbInstitute of Wastewater Management and Water Protection, Hamburg University of Technology, Eissendorfer Strasse 42, D-21073 Hamburg,

Germany

a r t i c l e i n f o

Article history:

Received 8 May 2006

Received in revised form

7 December 2006

Accepted 8 December 2006

Available online 5 February 2007

Keywords:

Anaerobic digestion

Biodegradability

Domestic wastewater treatment

Ecological sanitation

Grey water

UASB reactor

nt matter & 2007 Elsevie.2006.12.016

thor. Tel.: +20 13 3230297;t_elmitwalli@yahoo.com

a b s t r a c t

Feasibility of grey water treatment in an upflow anaerobic sludge blanket (UASB) reactor

operated at different hydraulic retention time (HRT) of 16, 10 and 6 h and controlled

temperature of 30 1C was investigated. Moreover, the maximum anaerobic biodegradability

without inoculum addition and maximum removal of chemical oxygen demand (COD)

fractions in grey water were determined in batch experiments. High values of maximum

anaerobic biodegradability (76%) and maximum COD removal in the UASB reactor (84%)

were achieved. The results showed that the colloidal COD had the highest maximum

anaerobic biodegradability (86%) and the suspended and dissolved COD had similar

maximum anaerobic biodegradability of 70%. Furthermore, the results of the UASB reactor

demonstrated that a total COD removal of 52–64% was obtained at HRT between 6 and 16 h.

The UASB reactor removed 22–30% and 15–21% of total nitrogen and total phosphorous in

the grey water, respectively, mainly due to the removal of particulate nutrients. The

characteristics of the sludge in the UASB reactor confirmed that the reactor had a stable

performance. The minimum sludge residence time and the maximum specific methano-

genic activity of the sludge ranged between 27 and 93 days and 0.18 and 0.28 kg COD/

(kg VS d).

& 2007 Elsevier Ltd. All rights reserved.

1. Introduction

Although domestic wastewater is a pollutant, it can be

utilised after proper treatment as a resource for fertiliser,

water and energy (Otterpohl et al., 1999, 2003; Elmitwalli et al.,

2003). Grey water, which is the wastewater generated in the

household excluding toilet wastewater (black water), repre-

sents the major volume of the domestic wastewater (60–75%)

with low content of nutrients and pathogens (Otterpohl et al.,

1999; Jefferson et al., 1999; Eriksson et al., 2002). Most of

treatment plants of the grey water include a one- or two-step

septic tank system for pre-treatment (Otterpohl et al., 2003).

The grey water treatment needs both physical and biological

r Ltd. All rights reserved.

fax: +20 13 3229 263.(T.A. Elmitwalli).

processes for removal of particles and dissolved organic

matters (Jefferson et al., 1999). Recently, grey water treatment

has been studied either by application of high-rate aerobic

systems, such as rotating biological contactor (Nolde, 1999),

fluidised bed (Nolde, 1999), aerobic filter (Jefferson et al., 2000),

membrane bioreactor (Jefferson et al., 2000), or by application

of low-rate systems, such as slow sand filter (Jefferson

et al., 1999) and vertical flow constructed wetlands (Otterpohl

et al., 2003).

A significant amount (41%) of organic carbon in the

domestic wastewater originates from grey water (Otterpohl

et al., 2003). Therefore, Elmitwalli et al. (2006b) treated

grey water in an upflow anaerobic sludge blanket (UASB)

ARTICLE IN PRESS

Nomenclature

ABmax maximum anaerobic biodegradability, %

COD chemical oxygen demand concentration, mg/L

CODt total COD concentration, mg/L

CODss suspended COD concentration, mg/L

CODcol colloidal COD concentration, mg/L

CODdis dissolved COD concentration, mg/L

CODfraction_wastewater_time COD fractions (t, ss, col or

dis) concentration of the wastewater (r, f or m) in

the biodegradability experiments at time (i or e),

mg/L

Hr hydrolysis based on the removed particulate COD,

%

HRT hydraulic retention time, h

Ht hydrolysis based on influent particulate COD, %

Mr methanogenesis based on the removed CODt, %

Mt methanogenesis based on the influent CODt, %

NH4–N ammonium concentration, mg N/L

PO4–P phosphate concentration, mg P/L

Q influent wastewater flow rate to the UASB reactor,

L/d

Qw excess sludge flow rate from the UASB reactor, L/d

SMAmax maximum specific methanogenic activity, kg

COD/(kg VS d)

SRTmaxmaximum sludge residence time, d

SRTmin minimum sludge residence time, d

TKj–N total Kjeldahl nitrogen concentration, mg N/L

TS total solids concentration, g/L

UASB upflow anaerobic sludge blanket

V UASB reactor volume, L

VS volatile solids concentration, g/L

X average sludge concentration in the UASB reactor,

g VS/L

Xe effluent concentration, g VS/L (from the results,

1 g VSE2.05 g COD)

Xw concentration of the excess sludge from the UASB

reactor, g VS/L

Yproduct_substrate yield of product from substrate in the

biodegradability experiments, mg/mg

Subscripts

col colloidal

dis dissolved

e at the end of the experiment

f paper filtered wastewater

i at the start of the experiment

m membrane filtered wastewater

ss suspended solids

WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 3 7 9 – 1 3 8 71380

reactor at ambient temperature (14–24 1C). They found that a

CODt removal of 41% can be achieved at hydraulic retention

time (HRT) of 12 h and 23 1C, while a removal of 31% can be

obtained at HRT of 20 h and 18 1C.

The maximum conversion to methane (ABmax) is an

important parameter for evaluating the anaerobic treatment

of any wastewater (Elmitwalli et al., 2001). Determination of

the ABmax is carried out in batch experiments either by

addition of inoculum or without. Inoculum addition repre-

sents inaccurate method due to production of dissolved

organic matter and biogas from decay of the inoculum, while

experiment without inoculum addition suffers from a long

operational period. Elmitwalli et al. (2004) determined the

ABmax of the pre-treated (in septic tank) grey water with

inoculum addition at 30 1C. They found a high ABmax for the

pre-treated grey water (86%).

The COD removal is limited in high-rate anaerobic systems

at low temperatures and, therefore, a long HRT is needed for

providing sufficient hydrolysis of particulate organic in

domestic wastewater (Zeeman and Lettinga, 1999; Elmitwalli

et al., 2002a). The grey water has a relatively higher

temperature (18–38 1C), as compared to the domestic waste-

water, because the grey water originates from hot water

sources, like shower (29 1C), kitchen (27–38 1C) and laundry

(28–32 1C) (Eriksson et al., 2002). Therefore, high-rate anaero-

bic systems might run efficiently for on-site treatment of grey

water in low-temperature regions.

The main objectives of this research are to study the

feasibility of grey water treatment in the UASB reactor and

determination of the maximum removal in the UASB reactor

and ABmax of the grey water and its fractions.

2. Materials and methods

2.1. Continuous experiment

An UASB reactor (7 L) was installed in the Institute of

Wastewater Management and Water Protection, Hamburg

University of Technology, Germany. Fig. 1 shows a sche-

matic diagram of the system. The system was installed

in a temperature-controlled room at 30 1C. The grey water

was collected from ‘Flintenbreite’ settlement in Luebeck,

Germany. A container of 1 m3 storage capacity was installed

in the settlement to collect the grey water from the

inlet sewer pipe of the first septic tank. The collected grey

water represented a wastewater of one day and was

transported directly to the Institute, where it was stored at

4 1C until usage for reactor feeding. Table 1 shows the

characteristics of the grey water. The UASB reactor was

operated for a period of 272 days and started operation after

addition of seed-sludge from an anaerobic digester treating

primary and secondary sludge. The UASB reactor was

operated at HRT of 16, 10 and 6 h for 95, 91 and 86 days,

respectively.

2.2. Batch experiments

The ABmax at a temperature of 30 1C was determined in

duplicate for raw, paper-filtered and membrane-filtered grey

water. For obtaining representative values, the biodegrad-

ability was determined two times for different wastewater

samples. The experiment was performed in serum bottles

ARTICLE IN PRESS

Biogas

Grey waterst ag k

P

Mixer

Sam

plin

g po

int s

Pump

Water tank for biogas

displacement

...

UASB reactor

15

200

cm

2520

1510

2515

4525

5

Diameter = 7 cm

Biogas

Mixer

Sam

plin

g po

int s

Pump

Water tank for biogas

displacement

...

UASB reactor

Diameter = 7 cmTemperatureControlledroom at 30°C

Beaker (2) forevaporationdeterminationin beaker (1)

Beaker (1) forthe waterreplaced bybiogas

Effluent

Grey water

storage tank

Fig. 1 – Schematic diagram of the UASB system treating grey water.

Table 1 – Characteristics of the grey water of ‘Flintenbreite’ settlement in Luebeck, Germany

CODt CODss CODcol CODdis TotalPO4–P

OrthoPO4–P

ParticulatePO4–P

TKj–N NH4–N ParticulateN–N

Average 640 325 190 125 9.8 8.0 1.8 27.2 4.2 23.0

Standard

deviation

127 132 96 37 0.6 0.6 0.5 3.5 2.0 3.6

WAT E R R E S E A R C H 41 (2007) 1379– 1387 1381

and each bottle was flushed with nitrogen gas for 5 min to

guarantee anaerobic conditions. The experiments were

carried out without inoculum addition. Therefore, the ABmax

was determined after a long time of 125 and 121 days for the

first and the second experiment, respectively. COD fractions

(suspended, colloidal and dissolved) were measured at the

start and the end of each experiment.

At the end of continuous operation of the UASB reactor,

batch recirculation experiments were carried out two

times to determine the maximum removal of COD fractions

in grey water. In each batch recirculation experiment,

the storage tank was filled with 28 L of grey water (4 times

of the UASB reactor volume). Then, the UASB reactor was

fed with the wastewater for a period of three HRT

without wastewater recirculation. The aim of this step

is to guarantee that the wastewater in the reactor will be

similar to the recirculated wastewater. After that, the

remaining grey water in the storage tank was recirculated in

the UASB reactor for a period of five days. The COD fractions

of the recirculated wastewater in the storage tank were

measured in time.

Anaerobic digestibility and SMAmax tests were performed at

a temperature of 30 1C for the sludge in the UASB reactor at

the end of each operational phase. The digestibility and the

SMAmax were carried out for a mixture of the sludge from

each sampling port in the reactor. The anaerobic digestibility

tests were performed in 250 mL serum bottles and each

bottle contained 150 mL of sludge followed by flushing with

nitrogen gas for 5 min to guarantee anaerobic conditions. The

tests were carried out without inoculum addition. Therefore,

the anaerobic digestibility was determined after long period

of 76, 84 and 79 days for the sludge of phases 1, 2 and 3,

respectively. The SMAmax was measured according to Elmit-

walli et al. (2002b), but by measuring CODdis depletion

(instead of acetate depletion) to simplify the test. The SMAmax

was determined based on the CODdis depletion in the second

feed with an initial acetate concentration of 1.5 g COD/L for

each feed and a sludge concentration of 2 g VS/L. The

anaerobic digestibility and SMAmax tests were done in

duplicate for each sample.

2.3. Analysis

COD, TKj-N, NH4-N, ortho and total P, TS and VS were

determined as described by Standard Methods (APHA, 1998).

Raw samples were used for CODt, 4.4mm folded paper-filtered

samples for CODf and 0.45mm membrane-filtered samples for

CODdis. The CODss and CODcol were calculated by the

differences between CODt and CODf, CODf and CODdis,

respectively.

ARTICLE IN PRESS

Table 2 – Maximum anaerobic biodegradability of CODfractions and the yield between the fractions

CODfraction

ABmax_

(%)Yield (mg/mg)

Yss_ Ycol_ Ydis_

Total (t) 74 (4)

Suspended

(ss)

70 (5) 0.05

(0.01)

0.01

(0.005)

Colloidal (col) 84 (4) 0.10 (0.02) 0.04 (0.01)

Dissolved

(dis)

70 (5) 0.01

(0.005)

0.02

(0.01)

Standard deviations are in brackets.

WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 3 7 9 – 1 3 8 71382

2.4. Calculations

For ABmax batch experiments:

ABmax _t ¼ 100CODt_r_i � CODt_r_e

CODt_r_i

� �, (1)

ABmax _dis ¼ 100CODdis_r_i � CODt_m_e

CODdis_r_i

� �, (2)

ABmax _col ¼ 100CODcol_r_i � CODt_f_e þ CODt_m_e

CODcol_r_i

� �, (3)

ABmax _ss ¼ 100CODss_r_i � CODt_r_e þ CODt_f_e

CODss_r_i

� �, (4)

Yss_dis ¼CODss_m_e

CODdis_r_i, (5)

Ycol_dis ¼CODcol_m_e

CODdis_r_i, (6)

Ydis_col ¼CODdis_f_e � CODdis_m_e

CODcol_r_i, (7)

Yss_col ¼CODss_f_e � CODss_m_e

CODcol_r_i, (8)

Ydis_ss ¼CODdis_r_e � CODdis_f_e

CODss_r_i, (9)

Ycol_ss ¼CODcol_r_e � CODcol_f_e

CODss_r_i. (10)

For the UASB reactor:

Ht ¼ 100CH4 as COD þ effluent CODdis � influent CODdis

influent CODss þ influent CODcol

� �,

(11)

Hr ¼ 100

�CH4 as COD þ effluent CODdis � influent CODdis

ðinfluent CODss þ influent CODcolÞ � ðeffluent CODss þ effluent CODcol

� �,

ð12Þ

Mt ¼ 100CH4 as COD

influent CODt

� �, (13)

Mr ¼ 100CH4 as COD

influent CODt � effluent CODt

� �, (14)

SRTmax ¼V � X

Qw � Xw

� �, (15)

SRTmin ¼V � X

Qw � Xw þ Q � Xe

� �. (16)

For the UASB reactor, it was assumed that 75% of the

measured biogas was methane (based on the biogas composi-

tion in previous studies in the treatment of domestic waste-

water in the UASB reactor, Halalsheh, 2002; Mahmoud, 2002;

Elmitwalli et al., 2002b). The amount of dissolved methane in

the effluent was calculated according to Henry’s Law.

Calculating hydrolysis or methanogenesis based on influent

and removed COD aims at evaluating, respectively, overall

performance of the reactor and capacity of the biomass in the

reactor in anaerobic digestion of the wastewater, as explained

by Elmitwalli et al. (2002a). For calculating the SRTmax, it was

assumed that the effluent VS was a part of the influent VS

with a residence time equal to the HRT, while for calculating

the SRTmin, it was assumed that the effluent VS had the same

SRT as the excess sludge. Due to the presence of channels and

dead zones in the sludge bed of the anaerobic reactor, the

actual SRT in the UASB is between SRTmax and SRTmin.

3. Results and discussion

3.1. Maximum anaerobic biodegradability

The results demonstrated that the grey water had a high

value of ABmax (7674%), which is relatively higher than that

reported for Dutch domestic sewage (72%), Elmitwalli et al.,

2001. Table 2 shows the ABmax of the COD fractions in grey

water and the yield between the fractions. The results

showed that the CODcol had the highest ABmax (86%) and

the CODss and CODdis had similar ABmax of 70%. These values

are similar to those reported by Elmitwalli et al. (2001) for the

COD fractions of the Dutch domestic sewage, 77, 86 and 62%

for CODss, CODcol and CODdis, respectively. The higher value

of the ABmax_col is the reason for the higher ABmax of grey

water as compared to the domestic sewage. The grey water

contained 30–35% CODcol, while the domestic sewage has

25–30% (Elmitwalli et al., 2001). The results showed that the

anaerobic degradation of CODcol produced higher fraction of

CODss and CODdis. Moreover, the anaerobic degradation of

CODss and CODdis produced higher fraction of CODcol. The

higher production of CODss from the degradation of influent

CODcol might be due the production of anaerobic bacteria and

bioflocculation of the colloids and the produced bacteria. It is

well known that many living microorganisms have a floccu-

lent growth habit and produce extracellular polymers, usually

polysaccharides, polypeptides or peptidoglycans, which en-

hance biosorption of particles (Dugan, 1987).

3.2. Maximum removal of COD fractions in the UASBreactor

Fig. 2 shows the results of the batch recirculation experi-

ments. The results demonstrated that the UASB reactor is an

efficient system for grey water treatment, as the maximum

ARTICLE IN PRESS

Run 1

0

200

400

600

800

0 2 4

Time (days)

CO

D (

mg/L

)

Run 2

0

200

400

600

800

CO

D (

mg/L

)

0

20

40

60

80

100

CO

D r

em

ova

l (%

)

1 3 5 0 2 4

Time (days)

1 3 5 0 2 4

Time (days)

1 3 5

Total COD

Suspended COD

Colloidal COD

Dissolved COD

a b c

Fig. 2 – Course of COD fractions concentration and removal efficiency in the batch recirculation experiments of grey water in

the UASB reactor (wastewater upflow velocity ¼ 0.33 m/h).

Table 3 – Average concentration and removal efficiency of COD fractions in grey water treatment in the UASB reactor atdifferent HRTs

Parameter Phase 1: HRT ¼ 16 h Phase 2: HRT ¼ 10 h Phase 3: HRT ¼ 6 h

Influent % Removal Influent % Removal Influent % Removal

CODt 618 (130) 64.0 (5.0) 647 (137) 52.3 (4.8) 682 (106) 52.0 (12.0)

CODss 308 (162) 83.5 (5.4) 353 (131) 79.4 (7.6) 310 (86) 67.6 (17.2)

CODcol 177 (114) 51.7 (19.0) 177 (81) 29.2 (19.8) 236 (90) 37.1 (17.5)

CODdis 133 (36) 50.9 (8.9) 117 (40) 30.3 (7.6) 136 (33) 34.8 (20.5)

Standard deviations are in brackets.

WAT E R R E S E A R C H 41 (2007) 1379– 1387 1383

removal efficiency for CODt, CODss, CODcol and CODdis were

84, 99.5, 79 and 74%, respectively. For achieving the maximum

removal, about 1–2 days of wastewater recirculation was

required. The maximum removal efficiency of COD fractions

of grey water is relatively higher than those reported by van

der Last and Lettinga (1992), Wang (1994) and Elmitwalli et al.

(2001) for pre-settled, pre-treated and raw domestic waste-

water, respectively. In the batch recirculation experiments,

the physical removal of CODss and CODcol by sedimentation

and filtration in the sludge bed of the UASB reactor without

anaerobic degradation was the reason for obtaining higher

value of maximum COD removal than that of ABmax. There-

fore, the maximum removal of CODss and CODcol were higher

than their ABmax. The results of the batch recirculation

experiments showed that the CODdis had a slightly higher

removal value (74%) as compared to its value of ABmax (70%),

because the fine particles (o0.45mm) might be removed by

physical entrapment after long period of wastewater recircu-

lation.

3.3. Performance of the UASB reactor

3.3.1. COD fractions removalTable 3 shows the concentration and removal efficiency of

COD fractions in grey water treatment in the UASB reactor.

The results demonstrated that the CODss followed by CODcol

represented the major part of the grey water CODt. In the

domestic wastewater and black water, the major part of

CODt is CODss followed by CODdis (Levine et al., 1985; Gaillard,

2002). The particulate (suspended+colloidal) COD represented

about 80% of grey water CODt, while it corresponded to

65–75% of domestic wastewater CODt (Wang, 1994). The

higher content of particulate in the grey water is due to the

increase of the CODcol percentage (30–35% of CODt), while it is

25–30% in the domestic wastewater CODt (Elmitwalli et al.,

2002b). The relatively high percentage of the CODcol in the

grey water might be due to the presence of surfactants and

lipids in the wastewater. The percentage of CODss was almost

similar in both grey water and domestic wastewater (40–50%

of CODt), while the percentage of CODdis was higher in

domestic wastewater (25–35%), as compared to grey water

(18–20%).

The results demonstrated that the highest CODt removal

(64%) was achieved at the longest HRT (16 h). However,

decreasing the HRT to 6 h reduced the CODt removal only to

52%. The CODt removal of the grey water is slightly lower

than that achieved in the treatment of domestic sewage at

similar conditions in the UASB reactor. Normally, the

UASB reactor in tropical regions (at temperature 425 1C) and

HRT 4–8 h removes 60–70% of CODt (Leitao, 2004). The

slightly lower CODt removal of the grey water as compared

to domestic wastewater might be due to the high content

of CODcol and surfactant in the grey water. The CODt

removal in the UASB reactor at HRT of 6 h is significantly

ARTICLE IN PRESS

WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 3 7 9 – 1 3 8 71384

higher than that reported by Elmitwalli et al. (2006a)

for a septic tank and an UASB reactor operated at lower

temperatures. It was found that the UASB reactor removed

33% of CODt of grey water at HRT of 20 h and low tempe-

rature of 18 1C and the septic tank removed 11–14% of CODt

at long HRT of 2–3 days and 20 1C (Elmitwalli et al., 2006a).

The septic tank had the lowest CODt removal, as it removes

only CODss by settling, while in the UASB reactor all COD

fractions can be removed by both physical (sedimen-

tation and filtration) and biological (anaerobic digestion)

mechanisms.

3.3.2. Nutrients removalTable 4 illustrates the average concentration and removal

efficiency of nutrients (N and P) in grey water treatment in the

UASB reactor. The results indicated that the grey water had a

limited amount of nitrogen, which was mainly in particulate

form (80–90%), while in the black water and domestic waste-

water most of the nitrogen is soluble as NH4. The concentra-

tion of phosphorous in grey water was almost similar to that

in the domestic wastewater (6.8–8.6 mg total-PO4-P/L for

Dutch municipal wastewater, Elmitwalli et al., 2002a, b). The

UASB reactor removed only the particulate nutrients by

sedimentation and filtration and, therefore, it had relatively

low removal of nutrients (Aiyuk et al., 2004). At the longest

HRT (16 h), the NH4 and ortho-P had a negative removal due to

hydrolysis of particulate N and P, respectively.

3.3.3. Sludge characteristics, hydrolysis and methanogenesisFig. 3 and Table 5 show, respectively, the sludge profile and the

characteristics of the sludge in the UASB reactor at the end of

each operational phase. The decreasing of the HRT resulted in

lowering the sludge concentration. The average sludge

concentration in the UASB reactor (12.6–14.1 g VS/L) was

slightly lower than that reported in the treatment of

municipal wastewater (415 g VS/L, Leitao, 2004). The rela-

tively low sludge concentration in the UASB reactor treating

grey water is mainly due to the lower VS/TS ratio, as

compared to that in the treatment of municipal wastewater

(60–85%, Elmitwalli et al., 2002a, b; Mahmoud, 2002). The

UASB reactor treating grey water had a relatively low VS/TS

ratio, because the reactor was treating raw grey water,

without grit removal. Therefore, the grit in the raw grey

water, like fine sand and inorganic material, precipitated in

Table 4 – Average concentration and removal efficiency of nutr

Parameter Phase 1: HRT ¼ 16 h

Influent % Removal Influ

Total PO4–P 9.9 (0.3) 15.2 (3.6) 9.7

Ortho PO4–P 6.6 (1) �5.5 (11.3) 8.7

Particulate PO4–P 3.3 (0.7) 53.0 (11.2) 1.0

TKj–N 27.1 (3.5) 29.8 (4.8) 27.3

NH4–N 5.5 (0.8) �70.0 (44.0) 3.9

Particulate N–N 21.6 (3.3) 52.8 (10.5) 23.4

Standard deviations are in brackets.

the reactor, resulting in low VS/TS ratio. The sludge profile,

visual observations and COD/VS ratio confirm the previous

assumption. The COD/VS ratio is similar to that reported for

the municipal wastewater treatment in the UASB reactor

(Mahmoud, 2002). The low values of sludge digestibility in the

three operational phases demonstrated that the sludge in the

reactor was almost stabilised. These values are similar to that

obtained in the treatment of the domestic wastewater in the

UASB reactor at tropical conditions (Leitao, 2004).

The results showed that the SMAmax increased at decreas-

ing the HRT from 16 to 6 h. This also was obtained by Leitao

(2004) in the domestic wastewater treatment in the UASB

reactor at tropical conditions. However, this phenomenon

cannot be generalised. In the domestic wastewater treatment

in a hybrid UASB reactor at low temperature, the SMAmax

decreased at lowering the HRT (Elmitwalli et al., 2002a, b).

Although the SMAmax increased at decreasing the HRT, the

CODt removal decreased. Therefore, it seemed that the

decrease in particulate removal and hydrolysis at lowering

the HRT might be the reason. The results of the SRT showed

that the SRTmin at HRT of 6 h is relatively low (27 days).

However, the system had a sufficient CODt removal and

methanogenesis (Table 6). Accordingly, the real SRT should be

between SRTmin and SRTmax.

Table 6 shows the percentage of methanogenesis and

hydrolysis in the UASB reactor. The Mt, Ht and Hr decreased

at lowering the HRT, while the Mr was almost constant.

Therefore, lower CODt removal and the higher SMAmax at

shorter HRT is due to the decrease in the particulate

removal and hydrolysis and the increase of the organic

loading rate at lower HRT. The difference between the

removed CODt and Mt is limited (20–24% of the removed

CODt) and is due to the growth of anaerobic bacteria and the

accumulation of slowly biodegradable and non-biodegradable

particulate in the reactor. However, this difference in the

batch experiments (ABmax and maximum removal in the

batch recirculation experiments) is 10% of the maximum

CODt removal. Accordingly, 10–14% of the removed CODt in

the UASB reactor is slowly biodegradable COD, which was not

converted in the grey water treatment in the UASB reactor.

The results of digestibility tests confirmed the previous

conclusion, as 7–12% of the CODt of the sludge in the UASB

reactor was digested after batch digestion for a period of

76–84 days.

ients (N and P) in grey water treatment in the UASB reactor

Phase 2: HRT ¼ 10 h Phase 3: HRT ¼ 6 h

ent % Removal Influent % Removal

(0.7) 17.4 (5.1) 9.9 (0.8) 20.6 (7.1)

(1.2) 14.5 (9.3) 8.4 (0.1) 18.7 (1.3)

(0.5) 43.2 (33.3) 1.5 (0.3) 30.2 (4.0)

(4.5) 21.7 (5.2) — —

(1.0) 15.0 (35.5) 3.5 (1.6) 47.2 (53.6)

(4.2) 31.2 (13.1) — —

ARTICLE IN PRESS

Table 5 – Characteristics of the sludge in the UASB reactor at the end of each phase

Parameter Phase 1: HRT ¼ 16 h Phase 2: HRT ¼ 10 h Phase 3: HRT ¼ 6 h

Average VS (g/L) 14.1 13.2 12.6

Average VS/TS (%) 55 (5) 48 (2) 45 (3)

COD/VS (mg/mg) 2.1 (0.1) 2.0 (0.1) 2.0 (0.1)

Digestibility (% kg COD/kg COD) 12 (3) 11 (2) 7 (2)

SRTmin (d) 93 64 27

SRTmax (d) 481 377 338

SMAmax (kg COD/(kg VS d)) 0.18 (0.03) 0.27 (0.04) 0.28 (0.04)

Standard deviations are in brackets.

Table 6 – Calculated methanogenesis and hydrolysis inthe UASB reactor treating grey water

Phase HRT % Methanogenesis % Hydrolysis

Mt Mr Ht Hr

1 16 51 (3) 80 (5) 52 (3) 78 (4)

2 10 40 (4) 77 (4) 32 (2) 56 (4)

3 6 38 (3) 76 (5) 30 (3) 54 (2)

Standard deviations are in brackets.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 10 20 30 40 50 60 70

TS (g/L)

0 10 20 30 40 50 60 70

VS (g/L)

Re

acto

r h

eig

ht

(m)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

Re

acto

r h

eig

ht

(m)

HRT = 16 h

HRT = 10 h

HRT = 6 h

Fig. 3 – Sludge profile in the UASB reactor at the end of each operational phase.

WAT E R R E S E A R C H 41 (2007) 1379– 1387 1385

3.4. Potential of applying anaerobic digestion in greywater treatment

The obtained high values for ABmax (76%) and maximum COD

removal in the UASB reactor (84%) showed the potential of the

UASB reactor in grey water treatment. Moreover, the results

demonstrated that a high CODt removal (52–64%) was obtained

in the UASB reactor at HRT between 6 and 16 h and wastewater

temperature of 30 1C. Accordingly, the UASB reactor has a

significantly higher CODt removal at a short HRT as compared

to the septic tank, the common system of grey water pre-

treatment. The temperature of grey water ranges between 18

and 38 1C (Eriksson et al., 2002) and it can be considered similar

to that applied in this research, if short connection between

the grey water sources and the UASB reactor, and proper

isolation of this connection are applied. Consequently, instal-

ling of the UASB reactor treating grey water in the cellar of the

buildings will be a suitable option. Thus, the losses in the

wastewater temperatures can be minimised and the energy

losses from warm water in the buildings can be utilised for

extra heating of the reactor, when it is required. For keeping

the temperature of the reactor at about 30 1C and minimising

the heat loss, a good insulation is required. Accordingly, the

thermal resistance of the insulator should be X5 K m2/W. For

each insulator, the thermal resistance is known and, therefore,

its thickness can be determined.

The height of the UASB reactor should be designed to

maintain the upflow velocity less than 0.5 m/h as recom-

mended by Mahmoud (2002) and therefore a height of 2 m, as

applied in this research, will be suitable to be applied in the

cellar of the buildings.

As the grey water has a high peak flow factor of 6.2 (Imura

et al., 1995), the UASB reactor treating grey water should have

a relatively longer HRT, as compared to that treating domestic

wastewater. Accordingly and based on the performance of

UASB reactor, a HRT between 6 and 8 h can be considered the

suitable HRT for the UASB reactor treating grey water.

ARTICLE IN PRESS

WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 3 7 9 – 1 3 8 71386

It is assumed within the ecological sanitation, when the

anaerobic digestion is applied, that the produced biogas will

be utilised for electricity production and/or heating. The

biogas loss as dissolved methane in the effluent of the UASB

reactor will be a limited value at 30 1C (o10% of the influent

COD as mentioned by Elmitwalli et al., 2006b). At application

of an UASB reactor for the pre-treatment of the grey water,

the produced biogas can be utilised for heating and produc-

tion of electricity, if it is combined with the produced biogas

from the anaerobic digester of the black water and kitchen

organic wastes. For example, in the settlement of ‘Flinten-

breite’ in Luebeck, Germany, the produced biogas from

anaerobic digester of black water and kitchen organic wastes

is used for heating and electricity production (Otterpohl et al.,

1999, 2003).

4. Conclusions

The following conclusions can be withdrawn from the

research:

1.

The grey water had high values of maximum anaerobic

biodegradability (7674%) and maximum COD removal in

the UASB reactor (8474%) at 30 1C.

2.

The CODcol had the highest maximum anaerobic biode-

gradability (8674%) and the CODss and CODdis had similar

maximum anaerobic biodegradability of 7075%.

3.

In the treatment of grey water in the UASB reactor at 30 1C,

CODt removal of 52–64% was achieved at HRT between

6–16 h.

4.

The UASB reactor treating grey water at HRT between

6–16 h and 30 1C removed 22–30% and 15–21% of total

nitrogen and total phosphorous in the grey water, respec-

tively, mainly due to the removal of particulate nutrients

by physical sedimentation, filtration and by incorporation

into biomass.

5.

The sludge in the UASB reactor treating grey water

at HRT between 6–16 h and 30 1C was stable and had

a sufficient maximum specific methanogenic activity

(0.18–0.28 kg COD/(kg VS d)) and sludge residence time

(minimum and maximum sludge residence time ranged

between 27–93 and 338–481 days, respectively).

Acknowledgements

The first author acknowledges Alexander von Humboldt

foundation for giving him a fellowship to carry out this

research. The authors are grateful to Jens Nielsen, Stefan

Deegener, Claudia Wendland and Moataz Shalabi for techni-

cal support.

R E F E R E N C E S

Aiyuk, S., Amoako, J., Raskin, L., van Haandel, A., Verstraete, W.,2004. Removal of carbon and nutrients from domestic waste-water using a low investment, integrated treatment concept.Water Res. 38, 3031–3042.

APHA, AWWA, WEF, 1998. Standard Methods for the Examinationof Water and Wastewater, 20th ed. Washington, DC, USA.

Dugan, P.R., 1987. The function of microbial polysaccharides inbioflocculation and biosorption of mineral ions. In: Attia, Y.A.(Ed.), Proceeding of Flocculation in Biotechnology and Se-paration Systems. Elsevier Science Publishers B.V., The Neth-erlands, pp. 337–350.

Elmitwalli, T.A., Soellner, J., de Keizer, A., Zeeman, G., Bruning, H.,Lettinga, G., 2001. Biodegradability and change of physicalcharacteristics of particles during anaerobic digestion ofdomestic sewage. Water Res. 35 (5), 1311–1317.

Elmitwalli, T.A., Sklyar, V., Zeeman, G., Lettinga, G., 2002a. Lowtemperature pre-treatment of domestic sewage in anaerobichybrid and anaerobic filter reactor. Biores. Technol. 82 (3),233–239.

Elmitwalli, T.A., Zeeman, G., Oahn, K.L.T., Lettinga, G., 2002b.Treatment of domestic sewage in a two-step system anaerobicfilter/anaerobic hybrid reactor at low temperature. Water Res.36 (9), 2225–2232.

Elmitwalli, T.A., van Lier, J., Zeeman, G., Lettinga, G., 2003.Treatment of domestic sewage at low temperature in a two-anaerobic step system followed by a trickling filter. Water Sci.Technol. 48 (11/12), 199–206.

Elmitwalli, T.A., Mahmoud, N., Soons, J., Zeeman, G., 2004.Characteristics of grey water: Polderdrift, The Netherlands.In: Proceeding of Second International Symposium on Ecolo-gical Sanitation ‘‘Ecosan—Closing the Loop’’, Lubeck, Ger-many.

Elmitwalli, T.A., Feng, Y., Behrendt, J., Otterpohl, R., 2006a.Anaerobic digestion potential for ecological and decentra-lised sanitation in urban areas. Water Sci. Technol. 53 (9),45–54.

Elmitwalli, T.A., Shalabi, M., Wendland, C., Otterpohl, R., 2006b.Grey water treatment in UASB reactor at ambient temperature.In: Proceeding of Seventh IWA Specialised Conference onSmall Water and Wastewater Systems to be held in Merida,Mexico, from March 7–10, 2006.

Eriksson, E., Auffarth, K., Henze, M., Ledin, A., 2002. Character-istics of grey wastewater. Urban Water 4 (1), 85–104.

Gaillard, A., 2002. Waste(water) characterisation and estimationof digestion kinetics. M.Sc. Thesis, Wageningen University,The Netherlands.

Halalsheh, M., 2002. Anaerobic pre-treatment of strong sewage: aproper solution for Jordan. Ph.D. Thesis, Wageningen Uni-versity, The Netherlands.

Imura, M., Sato, Y., Inamori, Y., Sudo, R., 1995. Development of ahigh-efficiency household biofilm reactor. Water Sci. Technol.31 (9), 163–171.

Jefferson, B., Laine, A., Parsons, S., Stephenson, T., Judd, S., 1999.Technologies for domestic wastewater recycling. Urban Water1, 285–292.

Jefferson, B., Laine, A.L., Judd, S.J., Stephenson, T., 2000. Mem-brane bioreactors and their role in wastewater reuse. WaterSci. Tech. 41 (1), 197–204.

Leitao, R.C., 2004. Robustness of UASB reactors treating sewageunder tropical conditions. Ph.D. Thesis, Wageningen Univer-sity, The Netherlands.

Levine, D., Tchobanaglous, G., Asano, T., 1985. Characterization ofthe size distribution of contaminants in wastewater: treat-ment and reuse implications. J. Water Pollut. Control Fed. 57(7), 805–816.

Mahmoud, N., 2002. Anaerobic pre-treatment of sewage underlow temperature (15 1C) conditions in an integrated UASB–di-gester system. Ph.D. Thesis, Wageningen University, TheNetherlands.

Nolde, E., 1999. Greywater reuse systems for toilet flushing inmulti-storey building-over ten years experience in Berlin.Urban Water 1, 275–284.

ARTICLE IN PRESS

WAT E R R E S E A R C H 41 (2007) 1379– 1387 1387

Otterpohl, R., Albold, A., Oldenburg, M., 1999. Source control inurban sanitation and waste management: ten systems withreuse of resources. Water Sci. Technol. 39 (5), 153–160.

Otterpohl, R., Braun, U., Oldenburg, M., 2003. Innovative technol-ogies for decentralised water, wastewater and biowastemanagement in urban and peri-urban areas. Water Sci.Technol. 48 (11/12), 23–32.

van der Last, A.R.M., Lettinga, G., 1992. Anaerobic treatmentdomestic sewage under moderate climatic (Dutch) conditions

using upflow reactors at increased superficial velocities. WaterSci. Technol. 25 (7), 167–178.

Wang, K., 1994. Integrated anaerobic and aerobic treatment ofsewage. Ph.D. Thesis, Wageningen University, The Nether-lands.

Zeeman, G., Lettinga, G., 1999. The role of anaerobic diges-tion of domestic sewage in closing the water and nutri-ents cycle at community level. Water Sci. Technol. 39 (5),187–194.