See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225283191

Grey water characterization and treatment for reuse in an arid environment

Article  in  Water Science & Technology · June 2012

DOI: 10.2166/wst.2012.167 · Source: PubMed

CITATIONS READS

36 3,243

2 authors, including:

Khalid Bani-Melhem
Hashemite University
39 PUBLICATIONS   1,679 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Toward an innovative device for grey water treatment by integrating electrocoagulation technique with adsorptive materials View project

Environmental medicine: social, instrumental and medical aspects (i.e. Delta dilemma, One health concept, Zero waste) View project

All content following this page was uploaded by Khalid Bani-Melhem on 19 February 2015.

The user has requested enhancement of the downloaded file.

 72 © IWA Publishing 2012 Water Science & Technology | 66.1 | 2012

Grey water characterization and treatment for reuse

in an arid environment
E. Smith and K. Bani-Melhem

ABSTRACT
E. Smith (corresponding author)
Grey water from a university facilities building in Cairo, Egypt was analysed for basic wastewater
K. Bani-Melhem
parameters. Mean concentrations were calculated based on grab samples over a 16-month period. Environmental Engineering Program,
The American University in Cairo,
Values for chemical oxygen demand (COD) and nutrients exceeded values reported in a number of Cairo 11511,
Egypt
other studies of grey water, while coliform counts were also high. A submerged membrane E-mail: edsmith@aucegypt.edu
bioreactor (SMBR) system using a hollow fibre ultrafiltration membrane was used to treat the grey
water with the aim of producing effluent that meets reuse guidelines for agriculture. A test run for 50
days at constant transmembrane pressure resulted in very good removal for key parameters
including COD, total suspended solids (TSS), colour, turbidity, ammonia nitrogen, anionic surfactants,
and coliform bacteria. High standard deviations were observed for COD and coliform concentrations
for both monthly grab samples and influent values from the 50-day SMBR experiment. SMBR effluent
meets international and local guidelines for at least restricted irrigation, particularly as pertains to
COD, TSS, and faecal coliforms which were reduced to mean treated values of 50 mg/L, 0 mg/L
(i.e., not detected), and <50 cfu/100 mL, respectively.
Key words | grey water, reuse, submerged membrane bioreactor

INTRODUCTION

The reuse of treated wastewater has moved to the centre of pathogenic bacteria, together with aerobic biological treat-
policy discussions and activity in the arid countries of the ment of dissolved organic matter (Merz et al. ; Li et al.
Middle East and North Africa (MENA) in the face of stressed ; Huelgas & Funamizu ).
freshwater sources, ongoing development goals, and the grow- Few studies exist on characterization of grey water in arid
ing need for environmental protection (Bazza ; Al Salem environments. In this work, several physical, chemical and
& Abouzaid ). Less attention has been given, however, biological parameters were analysed for grey water collected
to the collection and treatment of grey water for reuse, despite from a facilities services building at the American University
the increase in commercial and residential grey water reuse of Cairo’s (AUC) new campus situated in the desert near
practices in other regions of the world (Al-Jayyousi ; Cairo, Egypt. Approximately half of the total water demand
March et al. ; Friedler et al. ; Gross et al. ; Roesner (∼3,000 m3/d) at the campus is associated with landscape
et al. ). Grey water constitutes 50–80% of the total waste- irrigation of gardens including nearly 7,000 trees. Water is
water generated in households (Li et al. ), with the value currently from New Cairo city water supply and divided
varying for commercial establishments. Grey water quality between separate domestic and irrigation storage and deliv-
will also be highly variable depending on the source as well ery systems with a view toward eventual reuse of
as numerous site-specific and social factors (Pidou et al. wastewater for the irrigation of vegetation on the campus.
). While wastewater reuse standards vary in strictness Three reuse options are being considered; namely: (1) on-
according to application, almost all cases necessitate treat- site treatment and reuse of the entire wastewater stream;
ment. Physical, chemical and biological processes have been (2) discharging wastewater to the public sewer system and
applied for grey water treatment. Recently, the submerged connecting to a New Cairo City treated wastewater line;
membrane bioreactor (SMBR) has been investigated as an and (3) segregating and reusing gray water from site waste-
attractive method for grey water recycling as it combines phys- water while discharging the remainder to the public sewer
ical–chemical separation of colloidal substances, including system. This study provides data for evaluation of the latter
doi: 10.2166/wst.2012.167
 73 E. Smith & K. Bani-Melhem | Grey water characterization and treatment for reuse Water Science & Technology | 66.1 | 2012

option; specifically to have a picture of grey water quality and tank was performed to establish the influent concentration of
variation generated on the campus, and to investigate the key parameters during the 50-day period of operation.
potential for grey water treatment for reuse in irrigation Regardless of the sampling scheme, the composition of the
using a laboratory-scale MBR. raw grey water varied according to the distribution of activities
of building employees. The primary contributions to grey
water at the site were from cleaning, sinks, and kitchen activi-
MATERIALS AND METHODS ties. Samples were analysed by Hach methods (Hach, DR
2000, USA) for chemical oxygen demand (COD), chloride, sul-
Grey water collection and analysis phate, ammonia nitrogen (NH3-N), nitrate nitrogen (NO3-N),
phosphate (PO4-P), anionic surfactants, total dissolved solids
Grey water was collected from a facility services building on (TDS), conductivity, colour, total suspended solids (TSS),
the AUC campus located in New Cairo, Egypt. The plumb- and turbidity. Anionic surfactants were determined as methyl-
ing was modified for separation/collection of grey water ene blue active substances using dodecyl sodium sulphonate
during July 2009, including installation of a three way divert- as standard. Total and faecal coliforms were determined by
ing valve at the final grey water outlet to allow flow of grey the membrane filtration procedure according to Standard
water to either a 200 L collection tank at the site or directly Methods (). pH and temperature were measured using a
to the sewer. Monthly grab samples were assessed over a 15- pH meter model CG 842 (SCHOTT, Germany). Calibration
month period beginning in summer 2009 in order to obtain of the pH meter was conducted once a week before use. The
a general picture of grey water composition and variation average values presented in this study were calculated as an
with time. To obtain monthly grab samples, the diverting arithmetic mean of the collected data.
valve was opened to the empty collection tank for a 2–4 h
period. On the same day, a 20 L polypropylene container Grey water treatment
was filled from the collection tank, representing a 2–4 h
composite sample. The 20 L container was transported A laboratory scale SMBR with a working volume of 3.63 L
immediately to the laboratory (∼20 min) for analysis, with was used in this study (Figure 1). A hollow fibre, ultrafiltration
W
any required storage in a refrigeration unit at 4 C. (UF) membrane module, ZeeWeed-1 (GE/Zenon Membrane
During summer 2010, samples were collected daily from Solutions, Canada) was placed in the bioreactor. The device
the same grey water system described above to give an idea consisted of 80 fibres with 0.2 m length and pore size of
of day-to-day variations and provide raw grey water for the 0.04 μm and a total surface area of 0.047 m2. The MBR was
laboratory MBR experiments. These samples were more operated in flow-through mode with effluent from the mem-
instantaneous in that they were taken directly from the brane module withdrawn via a peristaltic pump operated at
outlet pipe into a 20 L polypropylene container beginning constant transmembrane (suction) pressure of 7.12 kPa.
1 min after opening the diverter valve, and then transported Due to the decrease in permeate flux with time for constant
to the laboratory. The raw grey water was poured into the con- transmembrane pressure, a level sensor was connected with
tinuously mixed, grey water (influent) tank for the MBR setup the feed pump via a level controller system to maintain con-
(see Figure 1). Frequent sampling and analysis from the mixed stant volume in the bioreactor. Compressed air was supplied

Figure 1 | Experimental setup used for grey water treatment by SMBR.

 74 E. Smith & K. Bani-Melhem | Grey water characterization and treatment for reuse Water Science & Technology | 66.1 | 2012

at a constant rate of 4.1 L/min through the air diffuser located effluent samples for the SMBR were evaluated for many of
at the bottom of the membrane module to provide good the same parameters as the monthly grab samples.
mixing of the sludge suspension and to create a shear stress
for effective scouring of the membrane surface.
The sludge for inoculation was taken from the secondary RESULTS AND DISCUSSION
clarifier in El-Jabal El Asfar municipal wastewater treatment
plant in Cairo. The sludge was acclimated to a sample of grey Grey water analysis
water for 45 days prior to membrane filtration experiments
using a fill-and-draw technique described by Chang et al. Summary statistics for the monthly grab, grey water samples are
(). The system was operated for 50 days (during summer presented in Table 1. Time variations of nine key wastewater
2010) comprising eight consecutive filtration-cleaning parameters over the study period are illustrated in Figures 2
cycles and operated at room temperature without control. through 5. The highest variation in parametric values was for
The fouling behaviour was evaluated phenomenologically COD and coliform counts. Several other parameters had stan-
by measuring the decline of permeate flux with time; thus dard deviations of more than half the mean value; namely,
no backwashing of the membrane was done during an oper- turbidity, anionic surfactants, colour, and nitrogen and phos-
ation cycle. Each cycle continued until the permeate flux phorus compounds. For this reason, median values are also
dropped to 10% of its initial value of ∼28 L/m2 h. Before start- included in Table 1 to minimize the weight of statistical outliers.
ing a new cycle, and in order to restore most of the There is not an obvious correlation between these variations
membrane’s permeability, the membrane module was and temperature/season, although a pronounced COD peak
removed from the bioreactor and physical and chemical occurs in late summer in both 2009 and 2010 (Figure 2).
cleaning were applied according to the protocol described COD and biochemical oxygen demand (BOD) values also cor-
by Meng et al. (). The MBR was operated at complete relate fairly well. Dissolved (TDS) and suspended solids (TSS)
sludge retention (SRT); i.e., no sludge was removed during concentrations are relatively steady over the study period
the period of the study except for a small amount every two (Figure 3), with TSS values generally less than those in munici-
days to determine the MLSS in the bioreactor. Influent and pal wastewater owing to the non-inclusion of black water.

Table 1 | Monthly grey water characteristics – summer 09 to autumn 10

Parameter Units Mean St Dev Median Min Max

pH pH 7.43 0.44 7.48 6.70 8.34

W
Temperature C 23.5 2.1 23.6 18.6 27.0
TDS mg/L 253 46 252 166 327
Conductivity μs/cm 496 105 450 360 655


Chloride (Cl ) mg/L 52.2 13.8 51.0 24.4 69.0
Sulphate (SO2

4 ) mg/L 58 11 60 38 80
TSS mg/L 74 31 80 15 116
Turbidity FTU 90 50 86 38 158
Apparent colour PtCo 299 202 286 102 625
COD mg/L 587 618 300 38 1,843
BOD5 mg/L 191 152 186 5 431
Ammonia (NH3) mg/L 4.48 3.69 3.90 0.17 12.75
Ammonium (NHþ
4) mg/L 3.38 3.24 4.00 0.08 11.30
Nitrate (NO

3) mg/L 3.96 2.70 2.64 0.12 7.48
Phosphate (PO3

4 ) mg/L 1.16 1.00 0.89 0.23 4.11
Anionic surfactants mg/L 8.55 5.16 9.94 0.27 11.12
6 6
Faecal coliforms CFU/100 mL 1.05 × 10 3.30 × 10 30,000 400 1.20 × 107
6 6
Total coliforms CFU/100 mL 1.30 × 10 3.84 × 10 38,000 2,500 1.40 × 107
 75 E. Smith & K. Bani-Melhem | Grey water characterization and treatment for reuse Water Science & Technology | 66.1 | 2012

Figure 2 | Time variation of COD and BOD for monthly grey water samples. Figure 4 | Time variation of nutrients for monthly grey water samples.

Figure 3 | Time variation of TDS and TSS for monthly grey water samples. Figure 5 | Time variation of coliforms for monthly grey water samples.

Ammonia and nitrate trended up and down throughout samples is higher for the MBR study. Exceptions to these
the study period, while phosphate peaked in early winter trends are the COD, nutrient, and coliform values.
and was relatively low the remainder of the year (Figure 4). Mean COD for the monthly grab samples was 70%
Figure 5 illustrates the large range in coliform concentrations higher than the MBR influent mean, and the standard devi-
(up to 5 orders of magnitude), and that time variable trends in ation was also higher; however the median value was
total and faecal coliform counts follow each other closely. essentially equivalent. Ammonia (NH3-N) in the MBR study
Coliform levels also appear to be highest in late spring and is more than double the mean for the grab samples, while
summer. COD and nitrogen values are generally higher the NO
3 -N mean in the grab samples is five times greater

than those reported in other grey water studies. For instance, than in the MBR influent. This may be influenced by the
average COD is ∼500 mg/L versus the range of 77–240 slight differences in sampling procedure noted previously;
reported by Eriksson et al. (). Ammonia-N and nitrate namely that monthly grab samples may have included grey
averages in this study are near 4.0 and 4.5 mg/L, respectively, water that had 2–4 h residence time in the collection tank,
while most previously reported values for both parameters resulting in some conversion of NH3 to NO
3 prior to analy-

are less than 1.0 (Roesner et al. ). The grey water as is sis. Higher mean concentrations of coliforms in the monthly
does not meet reuse standards for even restricted irrigation grab samples are due to the spike in Month 11 of >107 CFU/
according to either international or local guidelines, 100 mL; the highest influent concentration measured in the
especially as pertains to TSS, COD, and coliform bacteria MBR experiment was 106 CFU/100 mL.
(Ali-Badawy ; UN-WHO guidelines ).
Comparing grab sample values (Table 1) with influent Grey water treatment
values for the MBR experiment in summer 2010 (Table 2)
indicates a close consistency for the mean and even standard SMBR removal efficiencies for key wastewater parameters
deviation for most parameters tested. Mean values are just are given in Table 2. Excellent removal efficiencies (in
slightly higher for the MBR study for pH, TDS, TSS, turbidity, excess of 90%) were achieved for physical impurities such
colour, and surfactants. Interestingly, the standard deviation as colour and turbidity, and TSS was removed completely.
for these parameters is also higher for the MBR influent Faecal and total coliforms decreased by more than 4 log
(except for colour), despite the fact that the number of units to less than 50 CFU/100 mL as a result of rejection by
 76 E. Smith & K. Bani-Melhem | Grey water characterization and treatment for reuse Water Science & Technology | 66.1 | 2012

Table 2 | Mean influent/effluent characteristics and removal efficiencies for SMBR treatment of grey water (50-day experiment)

Percent removal
Water quality indexes N Influent Effluent This work Merz et al. (2007) Friedler et al. (2005)

pH 29 7.91 ± 1.02 7.89 ± 0.18 – – –

23.6 ± 0.8 23.9 ± 0.9 – – –
W
Temperature ( C) 29
DO (mg/L) 27 2.09 ± 1.17 5.48 ± 0.41 – – –
TDS (mg/L) 10 286 ± 111 297 ± 90 – – –
Conductivity (μs/cm) 25 573 ± 222 598 ± 179 – – –
TSS (mg/L) 25 78 ± 52 ND 100 – 87
Turbidity (FTU) 29 96.6 ± 87.8 5.0 ± 3.1 94.8 98 100
Colour (PtCo) 27 360.5 ± 170.3 26.0 ± 17.0 92.5 – –
COD (mg/L) 20 339 ± 316 50 ± 45 85.2 85 77
NH3-N (mg/L) 20 9.33 ± 3.9 0.33 ± 0.22 96.5 72 –
NO3-N (mg/L) 18 0.74 ± 0.74 9.4 ± 5.8 – – –
PO3

4 (mg/L) 31 0.519 ± 0.223 0.226 ± 0.103 56.6 4 –
Anionic surfactants (mg/L) 14 9.55 ± 6.79 0.48 ± 0.50 95.0 97 –
Faecal coliform (CFU/100 mL) 8 2.3 × 105 33 ± 30 99.99 99 (68)a >99.99 (2)a
± 1.0 × 105
Total coliform (CFU/100 mL) 8 4.2 × 105 39 ± 31 99.99 – –
± 2.3 × 105

N: number of samples, ND: not detected.

a
Number in parentheses is CFU/100 mL in effluent.

the UF membrane. Percent removals reported for two other

investigations of MBR treatment of grey water are included
in Table 2 for comparison. Organic matter (as COD) was
reduced by 85%, with time-variable removal given in Figure 6.
Figure 6 illustrates that the COD concentration in the grey
water during the first two membrane cycles (12–14 days)
was higher (average of ∼1,020 mg/L) than the subsequent
cycles (average of 300 mg/L). Despite the initial COD
removal being less than 80%, however, the average removal
during this early period was still 86%, or about the average Figure 6 | COD in influent and effluent of the MBR (left axis) and removal of COD (right
axis).
removal for the entire study period. In other words, MBR per-
formance was relatively unaffected by an order of magnitude
fluctuation in influent COD concentration.
This is an important issue that requires further research
with respect to optimizing SMBR performance as it is
known that certain dissolved and colloidal organic fractions
can enhance fouling of UF membranes and reduce permeate
flux (Judd ; Yamato et al. ). Fouling of the mem-
brane was manifested in the reduction in permeate flux
from the MBR with time as illustrated in Figure 7. During
the first hours of each cycle the flux declined very rapidly,
approaching steady state conditions after 15 h of operation.
The associated increase in hydraulic retention time (HRT)
during each cycle is shown in Figure 8. Figure 7 | Evolution of membrane permeate flux in the MBR.
 77 E. Smith & K. Bani-Melhem | Grey water characterization and treatment for reuse Water Science & Technology | 66.1 | 2012

The organic loading is also related to development of

biomass in the MBR which was monitored by determination
of MLSS and MLVSS as shown in Figure 9. Following the
acclimation period, the MLSS was 3,500 mg/L and
increased nearly linearly until day 12 when it levelled off
and remained at 6,200–6,400 mg/L until day 24, despite a
60–70% drop in influent COD around day 12 (Figure 6).
At day 24 there was a notable drop in MLSS associated in
part with the reduction in influent COD, but due mostly to
a failure in the level controller system resulting in a con- Figure 10 | NH3 in influent and effluent of the MBR (left axis) and removal of NH3 (right axis).

siderable MLSS loss in the overflow of the bioreactor. As

illustrated in Figure 9, the MLVSS/MLSS ratio fluctuated
back and forth within the range 0.65–0.85, indicating no
obvious accumulation of inorganic matter.
Ammonia nitrogen (NH3-N) reduction through biologi-
cal nitrification in the MBR was excellent with conversion
of nearly 97% on average and constant throughout the
period of operation (Figure 10). Phosphorus removal (phos-
phate) was less than 60% (Figure 11), while anionic
surfactants were reduced by 95% for an average influent of
nearly 10 mg/L (Figure 12).
Figure 11 | PO3
3

4 in influent and effluent of the MBR (left axis) and removal of PO4
Similar to COD, influent values of most raw grey water (right axis).
parameters were high during the first 10–15 days of the
SMBR experiment as illustrated in Figures 10–12. However,

Figure 12 | Anionic surfactants in influent and effluent of the MBR (left axis) and removal
of anionic surfactants (right axis).

Figure 8 | Hydraulic residence time (HRT) in MBR.

the percent removal of these constituents was still relatively
high during this period. In fact, Figures 6 and 10–12 indicate
that ‘valleys’ in the percent removal are normally associated
with temporal minima in influent concentrations. This
derives mostly from the percent calculation itself; i.e., as
effluent values are already very low, a drop in influent will
reduce the relative removal.
Most significant is that the SMBR treatment produces efflu-
ent that satisfies international and local guidelines for at least
restricted irrigation reuse. Table 3 presents mean MBR effluent
values from Table 2 versus Egyptian guidelines for wastewater
reuse in irrigation (Ali-Badawy ). In every case, even actual
effluent values satisfied these limits except for COD in the
Figure 9 | MLSS and MLVSS (left axis), and the ratio MLSS/MLVSS (right axis) in MBR. initial 10 days of the experiment when influent COD was
 78 E. Smith & K. Bani-Melhem | Grey water characterization and treatment for reuse Water Science & Technology | 66.1 | 2012

Table 3 | Mean MBR effluent values versus Egyptian guidelines for wastewater reuse in Al-Jayyousi, O. R.  Greywater reuse: towards sustainable
irrigation water management. Desalination 156, 181–192.
Al Salem, S. & Abouzaid, H.  Wastewater reuse for
MBR Non-food
agriculture: regional health perspective. East. Med. Health J.
Parameter Units effluent cropsa Unrestricted
12 (3/4), 446–458.
COD mg/L 50 80 40 Bazza, M.  Wastewater recycling and reuse in the Near East
TSS mg/L 0 40 20 region: experience and issues. Water Sci. Technol. 3 (4),
33–50.
TDS mg/L 297 2,000 2,000
Chang, I. S., Lee, C. H. & Ahn, K. H.  Membrane filtration
Faecal CFU/100- 33 1,000 100 characteristics in membrane-coupled activated sludge
coliforms mL system: the effect of floc structure on membrane fouling. Sep.
a
Including palm trees, cotton flax, jute, cereals, forage crops, and flowers.
Sci. Technol. 34, 1743–1758.
Eriksson, E., Auffarth, M. H. & Ledin, A.  Characteristics of
grey wastewater. Urban Water 4, 85–104.
especially high. Ongoing investigations aim at optimizing Friedler, E., Kovalio, R. & Galil, N. I.  On-site greywater
treatment and reuse in multi-storey buildings. Water Sci.
organic removal to potentially achieve unrestricted use in irri-
Technol. 51 (10), 187–194.
gation and reducing membrane fouling in order to sustain Gross, A., Azulai, N., Oron, G., Arnold, M., Nejidat, A. & Ronen,
higher permeate flux and thereby reduce the cost of treatment. Z.  Environmental impact and health risks associated
with greywater irrigation: a case study. Water Sci. Technol. 52
(8), 161–169.
CONCLUSIONS Huelgas, A. & Funamizu, N.  Flat-plate membrane bioreactor
for the treatment of higher-load graywater. Desalination 250,
162–166.
1. In one of the few reported studies of grey water analysis Judd, S.  Fouling control in submerged membrane bioreactors.
in the MENA region, mean COD and coliform concen- Water Sci. Technol. 51 (6–7), 27–34.
trations had high standard deviations for both monthly Li, F., Wichmann, K. & Otterpohl, R.  Review of the
grab samples and a focused 50-day study. In general, technological approaches for grey water treatment and
reuses. Sci. Total Environ. 407 (11), 3439–3449.
mean values were consistent between the two sampling
March, J. G., Gual, M. & Orozco, F.  Experiences on
cases for grey water from the same source.
greywater re-use for toilet flushing in a hotel (Mallorca
2. SMBR treatment of the grey water resulted in excellent Island, Spain). Desalination 164 (3), 241–247.
removal of nearly all wastewater parameters studied Meng, F., Zhang, H., Yang, F., Zhang, S., Li, Y. & Zhang, X. 
with effluent meeting guidelines for restricted irrigation Identification of activated sludge properties affecting
reuse, at a minimum. membrane fouling in submerged membrane bioreactors. Sep.
Purif. Technol. 51, 95–103.
Continuing efforts to optimize SMBR treatment may render Merz, C., Scheumann, R., El Hamouri, B. & Kraume, M. 
the reuse of grey water as a cost-effective asset in the overall Membrane bioreactor technology for the treatment of
water budget of many arid communities in the MENA region. greywater from a sports and leisure club. Desalination 215 (1–
3), 37–43.
Pidou, M., Avery, L., Stephenson, T., Jeffrey, P., Parsons, S. A., Liu,
S., Memon, F. A. & Jefferson, B.  Chemical solutions for
ACKNOWLEDGEMENT greywater recycling. Chemosphere 7(1), 147–155.
Roesner, L., Qian, Y., Criswell, M., Stromberger, M. & Klein, S.
This publication is based on work supported by Award No.  Long-term effects of landscape irrigation using
UK-C0015, made by King Abdullah University of Science household graywater. WERF and SDA, 03-CTS-18CO.
Standard Methods  Standard Methods for the Examination of
and Technology (KAUST).
Water and Wastewater, 19th edition. APHA/AWWA/WPCF,
Washington, DC.
UN-WHO, UNEP, and FAO  WHO Guidelines for the Safe
REFERENCES Use of Wastewater, Excreta and Greywater – Vol. 2:
Wastewater Use in Agriculture. WHO, Geneva.
Ali-Badawy, A.  Transport of E. coli in Egyptian agricultural Yamato, N., Kimura, K., Miyoshi, T. & Watanabe, Y.  Difference
soils as a result of reclaimed wastewater reuse for irrigation. in membrane fouling in membrane bioreactors caused by
MS Thesis, American University in Cairo, Egypt. membrane polymer materials. J. Membr. Sci. 280, 911–919.

First received 12 August 2011; accepted in revised form 20 February 2012

View publication stats