Luna 2007
Luna 2007
www.elsevier.com/locate/wasman
a
Departamento de Ingenierı́a Quı´mica y Ambiental, Escuela Superior de Ingenieros, Universidad de Sevilla, Camino de los Descubrimientos s/n,
41092 Sevilla, Spain
b
Instituto de Ciencias de la Tierra Jaume Almera (CSIC), c/Lluis Solé Sabarı́s s/n, 08028 Barcelona, Spain
Abstract
Treatment of municipal solid waste (MSW) landfill leachate generally results in low percentages of nutrient removal due to the high
concentration and accumulation of refractory compounds. For this reason, individual physical, chemical and biological processes have
been used for the treatment of raw landfill leachate and sometimes for the mixture of domestic wastewater and landfill leachate. In this
work, the possibility of treating landfill leachate was tested in a bench-scale pilot plant by a two-step method combining adsorption and
coagulation–flocculation. Zeolite synthesized from coal fly ash, a by-product of coal-fired power stations, was used in this study both as a
decantation aid reagent and as an adsorbent of COD and NH4–N. The coagulation–flocculation step was performed by the use of alu-
minium sulphate and a polyelectrolyte (ACTIPOL A-401). The leachate was collected directly from a storage unit of the organic fraction
of MSW, before it was composted. For this reason the raw leachate was diluted before treatment. The sludge was recirculated to enhance
the removal efficiency of nutrients as well as to optimize flocculant saving and to decrease sludge production.
The results showed that it is possible to remove 43%, 53% and 82% of COD, NH4–N, and suspended solids, respectively. Therefore,
this method may be an alternative for ammonium removal, as well as a suitable pre- or post-treatment step, in combination with other
processes in order to meet regulatory limits.
Ó 2006 Elsevier Ltd. All rights reserved.
0956-053X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.wasman.2006.10.016
1878 Y. Luna et al. / Waste Management 27 (2007) 1877–1883
removal of LL in continuous experiments with recircula- Although considerable laboratory research has been
tion of the effluent (lab and full scale). Contaminant developed in this field using batch experiments, there is a
removal from urban or industrial wastewater has been lack of experimentation at a continuous laboratory- or
investigated by the use of different adsorbent materials, pilot plant-scale in conditions closer to the ones existing
such as powdered activated carbon (PAC), powdered zeo- in practice. The main objective of this study was to investi-
lite and other low-cost sorbents obtained from industrial gate the effectiveness of CV-Z synthetic zeolite in the treat-
by-products or waste material (Aktas and Çeçen, 2001; ment of LL, in a continuous lab-scale plant. In addition,
Bailey et al., 1999; Çeçen et al., 2003; Kargi and Yonus this study analyses the effect of the use of zeolite on floccu-
Pomukoglu, 2004). In general, these low-cost sorbents are lant consumption, when effluent alone or effluent together
less effective than PAC in removing contaminants. How- with sludge were recirculated. The results were evaluated
ever, for this purpose, some of them may be convenient in terms of COD, NH4–N, suspended solids and PO4–P
to use because of their low cost. Additionally, the use of removal. COD and NH4–N removal was also quantified
these by-products may represent a way of increasing their as a function of the adsorbent concentrations.
added value since they are recycled into a treatment pro- This work intends to establish a technological basis, to
cess. Kargi and Yonus Pomukoglu (2004) compared the analyse the possible use of this easily synthesized zeolitised
efficiency of PAC and natural zeolites on the removal of fly ash in current MSW landfill leachate treatment plants,
COD and NH4–N by a combined adsorption-biological completing their treatment scheme, either as pre-treatment
treatment of LL and showed that PAC performed much of a biological process, or as a refining step for biologically
better than zeolite for the removal of COD, whereas treated leachate.
NH4–N removal performance of zeolite was significantly
superior to that of PAC. 2. Materials and methods
Synthetic zeolites may be obtained from fly ash, a by-
product of coal power plants. There are several hydrother- 2.1. Landfill leachate
mal activation methods used to synthesize zeolites from
coal fly ash. All of them are based on the dissolution of The landfill leachate was collected from the liquid waste
Al–Si-bearing phases with alkaline solution (mainly NaOH generated by the organic fraction of MSW in a storage
or KOH solutions) and the subsequent precipitation of unit, before its treatment by composting, in the integrated
zeolitic material (Querol et al., 2002). The use of synthetic municipal solid waste plant Monte Marta located in Seville
zeolites, with a high cationic exchange capacity, a large sur- (Spain).
face area and a high residual carbon content, has been Given the high content of contaminants in the original
investigated in the removal of heavy metals from wastewa- leachate sample, the sample was diluted up to 13% using
ter (Lee, 1997; Moreno et al., 2001a), NH4–N (Chang et al., tap water for both the laboratory coagulation–flocculation
2001; Jung et al., 2004; Lee, 1997; Otal et al., 2005), as well tests and the pilot-plant study without recirculation. How-
as in the immobilization of heavy metals from polluted ever, for the continuous operation with recirculation, the
soils (Moreno et al., 2001b). leachate was diluted with the recirculated effluent.
The synthetic zeolite used in our laboratory (CV-Z) was
obtained after the alkaline treatment of fly ash coming 2.2. Experimental set-up
from a Spanish power plant (Narcea), by using NaOH
solutions at 125 °C during 8 h. CV-Z is mainly composed Initial coagulation–flocculation tests, carried out follow-
of NaP1, low levels of analcime and chabazite zeolites ing ASTM Standard (1989) were performed in order to
and the fly ash remaining after the alkaline activation. This estimate the extent of nutrient removal. Zeolitic material
material has a cationic exchange capacity of 2.7 meq g 1, suspension (10 g L 1), aluminium sulphate (30 g L 1) as
equivalent to 60% zeolite content (Moreno et al., 2001a; the coagulant, and ACTIPOL A-401 (Brenntag Spain
Moreno et al., 2001b). In previous studies (Otal et al., WT) (0.25–0.5 g L 1) as flocculation agent were used as
2002; Otal et al., 2005), experiments were conducted in reagents for the treatment.
batch reactors (stirred or column reactors), in order to test
the adsorption capacity of CV-Z zeolite for the removal of 2.2.1. Laboratory-scale experiments
COD, total nitrogen and heavy metals from LL by combin- Nutrient removal from leachate was carried out using a
ing coagulation–flocculation and adsorption processes. The bench-scale plant that was designed to carry out different
elimination efficiencies achieved for COD and total nitro- operations (adsorption, coagulation–flocculation and
gen were, respectively, 56% and 64.5% by the treatment sedimentation).
in a stirred tank reactor and 10.3% and 50.4% by the best The adsorption tank used has an 18 cm inside diameter
in-column treatment (50 mL min 1 flow rate). Moreover, and a height of 30 cm, and is made of borosilicate glass.
the zeolite CV-Z can be considered a low cost adsorbent, The outflow may be located at different heights, so the
as reported previously (Otal et al., 2005), comparing the hydraulic retention time (HRT) can be regulated. The
cost effectiveness of treatment of CV-Z and a commercial volume of the adsorption tank (zeolite tank) was variable
adsorbent (Zeolite Type A). along the assay ranging from 0.76 L to 5 L. The suspension
Y. Luna et al. / Waste Management 27 (2007) 1877–1883 1879
was agitated with a 2-blade centrifugal stirrer at 556 rpm. tor was conducted in a discontinuous mode using a syringe.
Zeolitic suspension was fed to the coagulation–flocculation All tubing was made of silicone. The settling tank was
tank by means of a peristaltic pump. made out of 20 mm-thick methacrylate, and its dimensions
The coagulation–flocculation tank, also made out of are: 44 cm high, 68 cm long and 36 cm wide. If needed, the
borosilicate glass, had an inside diameter of 20 cm and tank had the possibility of using two sedimentation zones
height of 35 cm. The volume used for the experiments of variable volume in series.
was 1.8 L. Two baffles (3 cm width and 8 mm thick) of
methacrylate were stuck along the wall to improve the agi- 2.2.1.1. Continuous operation without recirculation. Adsorp-
tation of the suspension. An additional methacrylate baffle tion and coagulation–flocculation treatments were com-
protected the tank outflow. A 3-blade overhead propeller bined and performed in the same stirred tank reactor
stirrer at 233 rpm agitated the reactor content. Landfill (Fig. 1a). The bench-scale plant used LL diluted up to
leachate was continuously pumped from a feeding tank 13% with tap water that was the influent to the reactor
by a peristaltic pump. (F0). Addition of the zeolite suspension resulted in an extra
Peristaltic pumps were used to introduce the required dilution of the original leachate up to 7.8%. To start up the
dosages of coagulant and flocculant. Peristaltic pumps plant, 6.6 L of F0 was mixed with the reagents previously
were also used to continuously recirculate the effluent. estimated: 4.35 L of zeolite suspension (10 g L 1), 154 mL
Sludge recirculation into the coagulation–flocculation reac- of coagulant (30 g L 1) and 88 mL of flocculant
Fo = 2.11 L/h
7.85 % leachate
HRT= 2.5 h
HRT= 13 min
HRT= 2.5 h
HRT=13 min
QRECIRCULATION=0–35mL/h
Fig. 1. Layout of the experimental conditions used in the course of the continuous operation tests: without recirculation (a) and with recirculation (b).
1880 Y. Luna et al. / Waste Management 27 (2007) 1877–1883
(0.25 g L 1). The F0 flow rate employed for this study was yses. Chemical oxygen demand (COD), total Kjeldhal
2.11 mL h 1, while continuous coagulation–flocculation– nitrogen (TKN), NH4–N and suspended solids (SS) analy-
adsorption experiments were performed at a residence time ses were based on Standard Methods for the Examination
of 13 min. The dosages for the zeolite suspension, coagu- of Water and Wastewater (1989). COD was carried out by
lant and flocculant remained constant throughout the the dichromate reflux method, whereas the total nitrogen
experiment, their respective flow rates being 1390 mL h 1 content was determined using the macro Kjeldahl method.
(13.9 g h 1), 55.2 mL h 1 (1.66 g h 1) and 31.2 mL h 1 The measurement of SS was performed by drying the resi-
(7.8 mg h 1). The treated leachate was clarified in the set- due on glass fibre Whatman GF-C filter for 1 h at 105 °C
tling tank with an HRT of 2.5 h. For appropriate opera- until reaching constant weight. NH4–N concentration
tion, sludge was withdrawn once a day. The calculation was measured by using an ammonium ion selective elec-
of nutrient removal efficiency was based on a dilution of trode (Crison Instruments). Ortho phosphate was deter-
the landfill leachate of 7.8%. mined according to the molybdate spectrophotometric
method (UNE-EN 1189, 1996), which involves the prior fil-
2.2.1.2. Continuous operation with recirculation. To begin tration of the samples through 0.45 lm cellulose acetate fil-
this mode of operation, the sample to be treated was ter. All of the analyses were conducted in duplicate, giving
diluted by mixing 1100 mL of raw leachate and 9900 mL the average of two values.
of clarified effluent, previously obtained from the continu-
ous operation without recirculation. Reagents were added 3. Results and discussion
to this sample at the following optimum dosages, which
were estimated previously: 35.2 g of zeolite, 99.7 mL of 3.1. Continuous operation without recirculation
coagulant (30 g L 1) and 49.9 mL of flocculant
(0.25 g L 1). This pre-treated leachate was placed in the As indicated in the previous section, in the first phase of
coagulation–flocculation reactor and was used to fill the this study a LL continuous treatment without recirculation
sedimentation tank. was carried out. With this aim, raw landfill leachate was
During the recirculation mode (see experimental set-up diluted up to 13% of its original concentration in order
in Fig. 1b), the coagulation–flocculation reactor was fed to favour its treatability. The sample obtained (F0) was
with the raw leachate, mixed with recirculated effluent as analyzed and characterized in the laboratory (see Table 2
well as with the aforementioned reagents. Considering that for composition). The loading rates used during these
the zeolitic suspension was prepared with clarified recircu- experiments can be seen in Table 3. Since the composition
lated effluent, it is evident in this case that the tank for the and characteristics of the raw LL changed for each sample,
zeolitic suspension was working as an adsorption tank. the coagulant, flocculant and CV-Z dosage for the F0 had
Feed (F0), recirculation (F01) and zeolitic suspension (F02) to be optimized, by jar-tests, before starting up the pilot
flow rates were maintained constant at 350 mL h 1, plant tests. The results obtained in this case were:
2030 mL h 1 and 1120 mL h 1, respectively. As shown in 700 mg L 1, 3.3 mg L 1and 6.6 g L 1 of coagulant, floccu-
Table 1, coagulant and flocculant doses varied throughout lant and CV-Z, respectively.
the experiment. Sludge recirculation started after 105 h of The pH of the effluent remained almost constant and
operation; 35 mL h 1 of sludge produced on the settling ranged from 7.2 to 7.5 throughout the plant operation time
tank was withdrawn and manually added to the coagula- (25 h). In this experiment, the PO4–P concentration of F0
tion–flocculation reactor to adjust the flocculant concen-
trations to the desired level. The calculation of nutrient
Table 2
removal efficiency, for this mode of operation, was based Mean characteristics of diluted and raw landfill leachate, as influents of
on the raw LL. the continuous operation without recirculation and with recirculation
Variable Diluted LL concentration Raw LL concentration
2.3. Analytical methods (mg L 1) (mg L 1)
pH 7 6.8
Samples of influent and effluent were periodically col- COD 6811 48,110–52,310
lected throughout the time of the pilot plant tests for anal- NH4–N 270 2070–2540
PO4–P 0.088 0.251–0.757
Table 1 SS 183 1402–2570
Recirculation experiment
Time of operation (h)
Table 3
0–43.3 43.3–105.2 105.2–117
No recirculation experiments
F (mL h 1) F (mL h 1) F (mL h 1)
COD NH4–N PO4–P SS
Coagulant 36 116 108
(g h 1) (mg h 1) (lg h 1) (mg h 1)
Flocculant 15.8 78 31.7
Loading 14.4 569 185 385.3
Dosage of coagulation and flocculation aids, and flow rates (F) maintained
rate
during the test.
Y. Luna et al. / Waste Management 27 (2007) 1877–1883 1881
was 0.088 mg L 1, while the value after mixing F0 and the of view, it seemed interesting to evaluate the possibility of
zeolite suspension was 0.053 mg L 1. Phosphate concentra- diluting the raw LL using water from the sedimentation
tion in all of the effluent samples analysed was below the outlet, thus testing the recirculation capacity of the plant.
detection limit, which was 0.040 mg L 1. Percentages of Therefore, in this phase, the pilot plant was monitored
COD, NH4–N and suspended solids removal over the time for 120 h in a continuous mode with either effluent or efflu-
are shown in Fig. 2. The results clearly allow the distinction ent plus sludge recirculation. During operation of the plant
of three phases throughout the experiment; one, from time in this mode, dilution of the LL was obtained by mixing,
0 h to 7 h, which was the starting up phase, a second one, into the coagulation–flocculation reactor, the raw feed,
from 7 h to 18 h, that was a transition period, and the third the zeolite suspension and a portion of the returned effluent
one, from 18 h to the end of the experiment, in which the of the plant.
plant was performing in steady state. During this last The pH of the LL remained almost constant from 6.4 to
phase, the percentages of removal of all of the parameters 6.8, while the effluent pH increased slightly as compared to
studied remained constant. The average percentages of that of the influent, the average being 7.0 (6.4–7.5). Results
removal during the steady state phase, based on the leach- of COD, NH4–N and SS removal from the start-up of the
ate dilution of 7.8% of raw LL, were 28%, 63% and 71% for plant until a 120 h operation period was reached are pre-
COD, NH4–N and SS, respectively. With these results, the sented in Fig. 3. The percentage removal of SS decreased
amounts of COD and NH4–N adsorbed on CV-Z zeolite from the beginning until 54% removal efficiency was
were 290 mg g 1 and 25.7 mg g 1, respectively. achieved. At this moment the COD and NH4–N removal
These results are in agreement with those obtained in a efficiencies were 49% and 68%, respectively. When the plant
previous study (Otal et al., 2002), using discontinuous had been working for 45 h, the coagulant and flocculant
physico-chemical treatments for the treatment of a LL, flow rates applied to the reactor were increased. This pro-
diluted up to a 10%. It was found that after adding duced a progressive increase in the SS removal efficiency
1500 mg L 1 of coagulant [Al2(SO4)3], 22.6 mg L 1 of floc- of the system (Fig. 3). On the other hand, it can be seen
culant (ACTIPOL A-401) and 660 mg L 1 of CV-Z to the that the increase of reagents did not result in any improve-
leachate samples, a precipitate was rapidly formed, which ment in COD or NH4–N removal. The operating condi-
settled once the stirring was stopped, producing a clear tions of the plant were changed again after 80 h, since
supernatant. Contaminant removals achieved were 56% 35 mL h 1 of sludge were returned to the coagulation–floc-
and 64.5% for COD and total nitrogen (TKN), respec- culation reactor. An arrow in Fig. 3 indicates this change in
tively. Those results were quite significant and showed the system configuration. Results after this change reflect
the potential of the CV-Z zeolite for the treatment of land- that a large percentage of SS removal was achieved
fill leachate. In this work, the high removal efficiencies (80%). During the final operating period (from 103 to
obtained indicate that a continuous combined coagula- 120 h), the flocculant flow rate was reduced from 19.5 to
tion–flocculation plus adsorption process could treat the 7 mg h 1 to save this reagent and, interestingly, the SS,
diluted LL. COD and NH4–N removal remained almost constant. This
seems to indicate that the improvement in the SS removed
3.2. Continuous operation with recirculation could preferably be due to the increase in the coagulant
dose. In addition, the slight decrease of the NH4–N
In spite of the positive results obtained after continuous removal, in response to the increase of coagulant and floc-
operation without recirculation, from an engineering point culant doses at 45 h, could be associated with a change in
the pH brought about by an increase in the dosage of alu-
minium sulphate.
COD SS NH4-N
Altogether, these results indicate that in the experimental
100
system used, with addition of 1 g CV-Z, 0.31 g of coagu-
90
lant, 0.63 mg of flocculant and 3.13 mL of returned sludge,
80 it was possible to remove 644 mg of COD, 36.6 mg of NH4–
70 N and 59.4 mg of SS. Thus, the efficiency of COD, NH4–N
% Removal
physical–chemical treatment methods. Water Poll. Res. J. Canada 20 of zeolitic material synthesized from coal fly ash. In: 2001 Fly Ash
(2), 329–340. Utilization Symposium, Lexington, KY, USA.
Jung, Y.J., Chung, Y.C., Shin, H.S., Son, D.H., 2004. Enhanced ammonia Otal, E., Pereira, C.F., Vilches, L.F., Querol, X., 2002. Application of
nitrogen removal using existent biological regeneration and ammo- synthetic zeolites to the depuration of a waste landfill leachate. Waste
nium exchange of zeolite in modified SBR process. Water Res. 38, 347– Management and the Environment. WIT Press, Southampton, Boston,
354. p. 141–150.
Kargi, F., Yonus Pomukoglu, M., 2004. Adsorbent supplemented Otal, E., Vilches, L.F., Moreno, N., Querol, X., Vale, J., Fernández
biological treatment of pre-treated landfill leachate by bed-batch Pereira, C., 2005. Application of zeolitised coal fly ashes to the
operation. Bioresource Technol. 94, 285–291. depuration of liquid wastes. Fuel 84, 1440–1446.
Kurniawan, T.A., Lo, W.H., Chan, G.Y.S., 2006. Physico-chemical Querol, X., Moreno, N., Umaña, J.C., Alastuey, A., Hernández, E.,
treatments for removal of recalcitrant contaminants from landfill López-Soler, A., Plana, F., 2002. Synthesis of zeolites from coal fly
leachate. J. Hazard. Mater. B129, 80–100. ash: an overview. Int. J. Coal Geol. 50, 413–423.
Lee, J.H., 1997. Treatment of municipal landfill leachates using artificial UNE Spanish Standards. UNE-EN 1189, 1996. Phosphorus Determina-
zeolite. Fuel Energ. Abstr. 38 (6), 445. tion. Ammonium Molybdate Spectrophotometric Method.
Moreno, N., Querol, X., Ayora, C., Fernández Pereira, C., Janssen- Uygur, A., Kargi, E., 2004. Biological nutrient removal from pre-treated
Jurkovicova, M., 2001a. Utilization of zeolites synthesized from coal landfill leachate in a sequencing batch reactor. J. Environ. Manage. 71,
fly ash for the purification of acid mine waters. Environ. Sci. Technol. 9–14.
35, 3526–3534. Viviani, G., Torregrosa, M., 1997. Co-treatment of domestic sewage and
Moreno, N., Querol, X., Alastuey, A., Garcı´a, A., López, A., C. Ayora, landfill leachate. In: Proceedings of Sardinia 97, Sixth International
2001b. Immobilization of heavy metals in polluted soils by the addition Landfill Symposium, pp. 235–242.