Bioresource Technology 100 (2009) 2534–2539

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/ locate/bior tech

Thermophilic treatment of bulk drug pharmaceutical industrial wastewaters byusing hybrid up flow anaerobic sludge blanket reactor

D. Sreekanth a, D. Sivaramakrishna a, V. Himabindu a,*, Y. Anjaneyulu b

a Centre for Environment, Institute of Science and Technology, Jawaharlal Nehru Technological University, Kukatpally, Hyderabad 500 085, Andhra Pradesh, Indiab TLGVRC, JSU Box 18739, JSU, Jackson, MS 32917-0939, USA

a r t i c l e i n f o

Article history:Received 14 August 2008Received in revised form 14 November 2008Accepted 14 November 2008Available online 8 January 2009

Keywords:Pharmaceutical wastewaterUASBOrganic loading rateHRTThermophilic treatment

0960-8524/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.biortech.2008.11.028

* Corresponding author. Tel./fax: +91 040 2315613E-mail addresses: vurimindihimabindu@yahoo.com

Himabindu).

a b s t r a c t

The hybrid up flow anaerobic sludge blanket reactor was evaluated for efficacy in reduction of chemicaloxygen demand (COD) and biochemical oxygen demand (BOD) of bulk drug pharmaceutical wastewaterunder different operational conditions. The start-up of the reactor feed came entirely with glucose,applied at an organic loading rate (OLR) 1 kg COD/m3 d. Then the reactor was studied at different OLRsranging from 2 to 11 kg COD/m3 d with pharmaceutical wastewater. The optimum OLR was found tobe 9 kg COD/m3 d, where we found 65–75% COD and 80–94% of BOD reduction with biogas productioncontaining 60–70% of methane and specific methanogenic activity was 320 ml CH4/g-VSS d. By the char-acterization studies of effluent using GC–MS, the hazardous compounds like phenol, l,2-methoxy phenol,2,4,6-trichloro phenol, dibutyl phthalate, 1-bromo naphthalene, carbamazepine and antipyrine werepresent. After the treatment, these compounds degraded almost completely except carbamazepine. Ther-mophilic methanothrix and methanosaetae like bacteria are present in the granular sludge.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

The pharmaceutical wastewaters contain a variety of organicand inorganic constituents including spent solvents, catalysts, addi-tives, reactants and small amounts of intermediates and products,and may therefore be high in chemical oxygen demand (COD) (Fentet al., 2006; Oktem et al., 2007). It is estimated that approximatelyhalf of the pharmaceutical wastewaters produced worldwide arediscarded without specific treatment (Enick and Moore, 2007).Pharmaceutical wastewater has traditionally been treated usingphysico-chemical (Kulik et al., 2008) and aerobic biological pro-cesses (Suman Raj and Anjaneyulu, 2005). Recently there are spe-cific studies reporting the application of anaerobic technology forthe treatment of Bulk drug pharmaceutical wastewaters (Chenet al., 2008; Gangagni Rao et al., 2005; Chelliapan et al., 2006). How-ever, the high COD concentration in such pharmaceutical wastewa-ters makes them potential candidates for anaerobic technology(Chelliapan et al., 2006). The anaerobic process is very much favor-able for high strength wastewater, where aerobic oxidation of or-ganic matter would result in high-energy consumption andproduction of huge quantities of sludge. The most important meritsof anaerobic treatment are the ability to treat high strength wastes,low energy input, low sludge yield, low nutrient requirement, lowoperating cost, low space requirement and net benefit of energy

ll rights reserved.

3., sri_jntures@yahoo.co.in (V.

generation in the form of biogas (Acharya et al., 2008; Lefebvreet al., 2006; Mahmoud, 2008). Modern anaerobic processes usedfor high rate reactors such as the UASB or AF have been applied tothe treatment of a wide variety of industrial wastewaters with ahigh soluble COD content, including paper-pulp liquors (Ahn andForster, 2002; Elliott and Mahmood, 2007), spent sulphide liquors(Jantsch et al., 2002), and those from the food industry (Berardinoet al., 2000; Stabnikova et al., 2008). Hybrid UASB reactors havebeen successfully applied to the partial treatment of fiberboardmanufacturing wastewaters (Fernandez et al., 2001) and slaughter-house effluent (Torkian et al., 2003). The hybrid UASB reactorscould also become a preferred option for certain bulk drug pharma-ceutical wastewaters due to several operational advantages overother reactor configurations. Reported studies in anaerobic treat-ment of pharmaceutical wastewater refer mostly to the use of up-flow anaerobic stage reactor and anaerobic suspended filmreactors (Chelliapan et al., 2006; Mohan et al., 2001). Publicationson the application of hybrid reactors to the treatment of pharma-ceutical wastewater containing organic solvents are limited (Oktemet al., 2007). No study was carried out on the thermophilic treat-ment of bulk drug pharmaceutical wastewater by using hybridUASB reactor, therefore the suitability of using this type of reactorconfiguration for treatment of bulk drug pharmaceutical wastewa-ters justifies further investigation under thermophilic condition.Thermophilic process offers several merits, such as an increaseddegradation rate for organic solids, a high gas production rate, im-proved solid liquid separation, increased disinfection of pathogenicorganisms and eliminating the need of cooling for effluent of high

D. Sreekanth et al. / Bioresource Technology 100 (2009) 2534–2539 2535

temperature (Park et al., 2008). Thus, it is of practical interest toevaluate the performance of anaerobic treatment of bulk drug phar-maceutical wastewater under thermophilic conditions.

The specific objectives include:

� Investigating the performance of the reactor at different organicloading rates.

� Investigating the performance of the reactor under shockloading.

� Morphological examination of granular sludge treating bulkdrug pharmaceutical wastewater.

2. Methods

2.1. Experimental setup

Lab scale experiments were conducted in lab scale HUASB reac-tor of 17 l capacity. The reactor is a circular column with a length of90 cm, internal diameter of 10 cm and wall thickness of 2 mm. Thereactor was provided with hopper bottom. Four sampling ports areprovided along its length at equal distance. Inlet end opens to-wards the bottom of the reactor, so the feed strikes at the bottom.An outlet was provided at the top, which is connected to the efflu-ent tank. On the top of the reactor a gas solid separator is providedto separate gas and solid raised due to the upward movement ofthe feed. The gas outlet was connected through rubber tubing tothe liquid displacement system to measure the gas production.The amount of gas produced is directly proportional to the amountof liquid displaced and hence gas produced can be measured atregular intervals of time. A filter media made up of PVC ringswas provided at the middle of the reactor. The reactors were oper-ated at thermophilic temperature (55 ± 3 �C) by a water-jacket.Heated water was pumped from a recirculation water bath throughthe constant temperature jacket surrounding the reactor. Syntheticwastewater comprising substrates, glucose, plus balanced nutri-ents and alkalinity was used to feed the reactor using a peristalticpump.

2.2. Wastewater source and characterization

The wastewater for this study was obtained from a local bulkdrug pharmaceutical unit in Hyderabad, from which the mainproduct was terbinafine hydrochloride and characterized usingstandard methods (APHA, 2000). The characteristics of bulk drugpharmaceutical wastewater used for the study were (mg/l): colour(orange), total dissolved solids (8500–9000), total suspended solids(2800–3000), COD (13,000–15,000), BOD (7000–7500), volatilefatty acids (600–750), alkalinity as CaCO3 (2500–3000), chlorides(200–250), nitrates (120–170), sulphates (300–450), phosphates(100–120), phenol (25–30), 2-methoxy phenol (20–25), 2,4,6-tri-chloro phenol (20–25), dibutyl phthalate (30–40), 1-bromo naph-thalene (5–10), antipyrene (5–10) and carbamazepine (10–15).The pH of the bulk drug pharmaceutical wastewater was 7.0–7.5.The BOD: COD ratio of the wastewaters was in the range 0.45–0.6, which is amenable to biological treatment. All the chemicalsused were of analytical reagent grade. Water used in all the exper-iments was laboratory distilled water with pH (7.2–8.0), Alkalinity(40–120 mg/l), chlorides (20–30 mg/l). Tap water from a groundwater source is used for the dilution of raw effluent.

2.3. Inoculum

The inoculum for seeding the reactor was brought from a full-scale UASB reactor, which was in use for treatment of wastewater

from a slaughterhouse. The sludge was initially passed through a0.2 mm screen to remove the foreign material. Sludge (6.0 l) wasadded to the reactor. The volatile suspended solids (VSSs) of theinoculum was 28.86 g/l and the methanogenic activity was foundto be 120.8 ml-CH4/g-VSS d. The sludge was acclimatized for threeweeks using pharmaceutical waste water from bulk drug pharma-ceutical industry under anaerobic conditions. After acclimatization,the VSS of the sludge increased to 35.69 g/l and methanogenicactivity raised to 132.0 ml-CH4/g-VSS d, and same was inoculated.

2.4. Analytical methods

Analysis of alkalinity, phosphates, sulphates, BOD, ash contentof the sludge, sludge volume index (SVI), chlorides, total suspendedsolids (TSS), nitrates and COD were conducted in accordance withStandard Method (APHA, 2000). Gas production was measured bythe displacement of 0.1 N H2SO4 solution. Oxidation-reduction po-tential was monitored by using ORP meter (Model – 108, Orion,USA). The biogas composition was estimated using a gas chromato-graph (AIMIL-NUCON Series 7500, New Delhi, India) fitted with athermal conductivity detector (TCD) and Poropak Q stainless steelcolumn. The oven, injector, and detector temperatures were set as40, 60 and 60 �C, respectively and nitrogen was used as the carriergas. Volatile fatty acid (VFA) concentration was measured aftercentrifuging the samples to remove the suspended solids. A gaschromatography (AIMIL-NUCON, India, Series 7500) equipped witha flame ionization detector (FID) and Chromasorb 101 column wasused for the analysis of VFA. The detector, injector and oven tem-perature were 200, 195 and 180 �C, respectively. The carrier gasused was nitrogen, and a mixture of hydrogen and air was usedto sustain the flame in the detector. Image analysis system (Leica,Germany) was employed to determine the size of the granules.Scanning electron microscope (SEM) (Model: JOEL-JSM 5600) wasemployed to study the morphology of the granules. Specimenpreparation for SEM included fixation with 5% glutaraldehydeand 1% osmium tetroxide, followed by dehydration with 50–100% acetone before drying. Finally, a thin layer of gold metalwas applied over the sample using an automated sputter coater(JEOL JFC-1600) for about 3 min. The samples were finally scannedat various magnifications. The hazardous compounds were identi-fied by acidifying aqueous samples to pH < 2, extracting the com-pounds with dichloromethane, filtering using a MDI Teflonsyringe filters (SY25NN, 0.2 lm pore size) and injecting 1 ll ofthe filtrate into a gas chromatograph (6890 N, Agilent) with a massselective detector (5973). A HP 5 MS, capillary column, (0.25 mm,30 m, 0.25 lm), was used for sample elution. All the experimentswere carried out in triplicates and the values presented in this pa-per are a mean of three readings for reproducibility of the results.

2.5. Specific methanogenic activity of granules

Specific methanogenic activity (SMA) of granules was carriedout in 250 ml flasks. The granules were incubated three days to re-move residual substrates, then washed with phosphate buffer solu-tion (1.28 g Na2HPO4/l and 0.42 g Na2HPO4/l) in order to removeadditional residual substrates and, used for the activity test. Thegranule samples of 2.2 g in volatile suspended solids were trans-ferred to flasks containing nutrient medium and incubated withoutagitation further 12 h at 35 ± 3 �C in a dark room. Then, glucosesolution was added (3000 mg/l as COD) and flushed with N2 for5 min and incubated at 35 ± 3 �C. The gas produced was measuredat 1 h intervals by liquid displacement method. After each gasmeasurement, the flasks were manually shaken. Specific methano-genic activity of granules was determined by taking the slope ofthe graph drawn by methane production (cumulative) against timeand expressed as CH4 l/g-VSS d (Somasiri et al., 2008).

Fig. 1. Performance of the hybrid UASB reactor for the removal of COD and BODduring the period of operation.

2536 D. Sreekanth et al. / Bioresource Technology 100 (2009) 2534–2539

2.6. Start-up of hybrid UASB reactor

The start-up period of an anaerobic hybrid UASB reactor is di-rectly proportional to the concentration of the microbial popula-tion. Rate of start-up depends on the type of inoculum, the typeand strength of waste, level of volatile acids maintained. (Nandyand Kaul, 2001). Normally reactors are started by acclimatizingthe biomass with glucose (Kasapgil et al., 2002). Glucose is a read-ily degradable, soluble carbohydrate that does not, itself, limit therate of anaerobic biodegradation. It produces readily measurableintermediary metabolites in anaerobic digestion, and is commonlyused as a carbonaceous substrate in many experimental studies(Oktem et al., 2007). Glucose was, therefore, used as a substrateduring the initial acclimation phase of this study, and during thelatter stages of which it was gradually replaced with pharmaceuti-cal wastewater. Anaerobic seed culture (6.0 l) collected from the(UASB) anaerobic digester of slaughter house (ALKABEER Industry,Hyderabad) was used for the inoculation in UASB reactor. The con-centration of volatile suspended solid (VSS) and suspended solid(SS) in the sludge blanket of UASB reactor was 19.93 kg VSS/m3

and 28 kg VSS/m3, respectively during inoculation. The reactorwas operated in a continuous mode of operation. The feed compo-sition of the HUASB reactor was as follows: glucose 1.0 g/l, NH4Cl800 mg/l, KH2PO4 200 mg/l, CaCl2 � H2O 48 mg/l, FeSO4 � 7H2O40 mg/l, H3BO3 0.1 mg/l, ZnCl2 0.1 mg/l, CuCl2 0.1 mg/l,MnSO4 � 4H2O 0.1 mg/l, AlCl3 0.1 mg/l, NiCl2 0.1 mg/l, NaHCO3

3000 mg/l, trace metal solution 1 ml/l. The composition of tracemetal solution was as follows: 5 g Mg SO4 � 7H2O, 6 g FeCl2 � 4H2O,0.88 g CoCl2 � 4H2O, 0.1 g H3BO3, 0.1 g ZnSO4 � 7H2O, 0.05 g Cu-SO4 � 5H2O, 1 g NiSO4 � 8H2O, 5 g MnCl2. 4H2O and 0.64 g (NH4)6Mo7O24 � 4H2O (Chelliapan et al., 2006). One millilitre of the stocktrace metal solution was added per litre of feed. COD:N:P ratio wasmaintained around 300:5:1 (Sreekanth et al., 2008). The COD andBOD reduction is found to be in the range of 35–60% and 56–80%, respectively, during the start-up phase. The pH of the reactorfeed is always maintained neutral by adding necessary amount oforthophosphoric acid. The outlet pH is found to be in the rangeof 7.5–7.6 indicated an active metabolism of the methanogens.The best operation of anaerobic reactors can be expected whenthe pH is maintained near neutrality (Rajeswari et al., 2000). Thenative anaerobic biomass responded very well by yielding biogasof 0.25 m3/kg COD removed which consisted of 58–62% methane.This indicates satisfactory start-up of the reactor. High bulk liquidorganic loading and the start-up with high inoculum seed (VSS of35.69 g/l and methanogenic activity of 132.0 ml-CH4/g-VSS d) re-sulted in good microbial growth. During this period the HRT was1.5 days.

Fig. 2. Performance of hybrid UASB reactor at various organic loading rates.

3. Results and discussion

3.1. Performance of the HUASB reactor at different OLRs

The reactor operated with higher organic loading rates afterattaining a consistent stable carbon removal condition at OLR of1 kg COD/m3 d to assess the optimum loading rate during processoptimization. The OLR was gradually improved to 2 kg COD/m3 dat which point the feed to the UASB reactor was progressively re-placed with pharmaceutical wastewater. The reactor performancewas evaluated up to 12 kg COD/m3 d at an increment of1.0 kg COD/m3 d OLR. At each loading rate, the reactor was oper-ated until a steady state reached as indicated by a constant gas pro-duction rate (±5%) and effluent COD level (±8%). The performanceof the reactor with varying loading rates, function of COD andBOD removal at were explained in Fig. 1. It was evident from thefigure that every incremental step in OLR marked an inhibition in

the reactor performance for a short period after initial loading. Thismay be an indication to the increase in the substrate concentration,which requires sufficient acclimatization period for native microflora to nurture to the changed environmental conditions of thesystem. From the data, we can assess that the period required foracclimatization is directly related to the OLR. The performance ofthe reactor at various OLR levels were consolidated and presentedin Fig. 2. During the reactor operation there was a reduction in CODand BOD levels up to 47–66% and 83–86%, respectively. It is evi-dent from the table that increment in loading rate has shown grad-ual increment in substrate removal rate up to 10 kg COD/m3 d andsubsequently substrate removal rate dropped, indicating the inhi-bition of the biomass activity at that concentration. It explains thatthe designed reactor can be operated up to 10 kg COD/m3 d with-out system inhibition and subsequent increase in OLR lead toreduction in overall process efficiency. Throughout the reactoroperation biogas production and a consistent value in the rangeof 0.3–0.5 ml/mg COD reduction was observed irrespective of theoperated OLR. In the present study, ORP values are varied between�35 and �61. The anaerobic bacteria function best between ORPvalues of +30 mV and –400 mV (Reddy et al., 1995). On analysisof biogas, we observed gases in varying composition among themmost abundant are in range of 65–70% methane and 30–35% CO2.Hydraulic retention time (HRT) considerably affects biogas compo-sition, and in the present study a HRT of 1.7 days is found to beoptimum to retard the CO2 toxicity with relatively uniform CO2

content (30–35%) at all the studied OLR. These values are similarto that of Gangagni Rao et al. (2005), where they treated the bulk

D. Sreekanth et al. / Bioresource Technology 100 (2009) 2534–2539 2537

drug industry wastewater by using fixed film reactor (AFFR). Anoptimum OLR of 9 kg COD/m3 d was achieved (Fig. 3) in the pres-ent study which is comparable to the results obtained by Kasapgilet al. (2002). The VFA concentration in effluent was explained. TheVFA concentration was varied from 150 to 400 mg/l as acetic acidat OLRs ranging from 1 to 11 kg COD/m3 d, respectively, indicatingno loss of methanogenic potential and a satisfactory balance be-tween acidogenic, aetogenic and methanogenic microorganismsin the mixed biomass. When the OLR was increased suddenly from2 to 5 kg COD/m3 d, the VFA concentration also increased from 400to 2500 mg/l as acetic acid, indicating methanogenic inhibition to-wards toxicity of bulk drug industrial effluents. At this stage, thefall in the pH to 6.5 was observed, which might be due to the in-crease of the VFA concentration. When the reactor was switchedback to 1 kg COD/m3 d, the VFA concentration decreased signifi-cantly to 395 mg/l. Fig. 4 indicates that the same condition was re-peated when the OLR was increased to 9–13 kg COD/m3 d. Similarvalues reported by Oktem et al. (2007). The specific methanogenicactivity (SMA) values at different organic loading rates are shownin Fig. 4. The SMA values were varied from 280 to 356 ml-CH4/g-VSS d at OLRs ranging from 1 to 11 kg COD/m3 d, respectively.However, when OLR was increased suddenly from 2 to 5 kg COD/m3 d, the SMA drastically decreased from 356 to 90 ml-CH4/g-VSS d and again increased to 250 ml CH4/g-VSS d, when the reactorwas switched to 2 kg COD/m3 d. Fig. 4 indicates that the same con-dition was repeated when the OLR was increased to 9–13 kg COD/m3 d and again the reactor reached to steady state conditions,when the OLR was decreased to 9 kg COD/m3 d.

In the present study slight sulphate reduction was observedthrough out the study .The initial sulphate concentration is

Fig. 3. Optimization of organic loading rate in hybrid UASB reactor.

Fig. 4. Variation of VFA concentration and variation of SMA during the period ofoperation.

450 mg/l present in the effluent, after treatment the sulphate con-centration was reduced to 250 mg/l. The effluent was characterizedfor hazardous compounds before and after treatment using GC–MS. From the characterization studies using GC–MS, the hazardouscompounds identified in the effluent were phenol, 2-methoxy phe-nol, 2,4,6-trichloro phenol, dibutyl phthalate, 1-bromo naphtha-lene and antipyrine. After the treatment these hazardouscompounds degraded almost completely except carbamazepinewas not degraded (Clara et al., 2004; Carballa et al., 2006).

3.2. Performance of the reactor under shock loading

In the present study, organic shock load tests were carried outto assess the reactor stability at two stable OLR (Table 1). The reac-tor is stabilized at 1.0 COD/m3 d and the OLR is increased to5 kg COD m3 d to observe the reactor performance for organicshock loading. It can be observed from the table that due to organicshock loading, the reactor performance significantly dropped withrespect to carbon removal and specific methanogenic activity.Therefore, again the OLR is brought back to 1.0 kg COD m3 d wherein the reactor performance is restored to normal stage. The reactoris again subjected to organic shock loading of 13 kg COD m3 dwhen the reactor is operating at stable OLR of 9 kg COD m3 d withconsistently steady state performance. In this case also reactor per-formance dropped with respect to carbon removal and is restoredwhen the organic loading is again decreased. It can be deducedfrom the above results that reactor can withstand shock loadsand can be brought back to normalcy quickly during the operatingphase. This result is in agreement with the experiment undertakenby Gangagni Rao et al. (2005), who exposed an anaerobic fixed filmreactor treating pharmaceutical wastewater to organic shock loadsand showed that the reactor performance dropped with respect tocarbon removal and was restored when the organic loading wasagain decreased to normal stabilized OLR.

3.3. Characteristics of granular thermophilic sludge

At the end of the 200 days operation the granular sludge was ta-ken from the bottom of the reactor for the measurement of itscharacteristics using scanning electron microscope. It showed thatover all surface of the granules were rough and uneven. The visualexamination of granular biomass revealed a black colour with aspherical shape. Large cavities were present on the surface. Thephysico-chemical characteristics of granules were studied duringthe start-up and steady state operation at various organic loadingrates. SVI of the sludge in the reactors decreased from a value of16–20 to 11–12 mL/g SS in the entire reactor. The plausible reasonfor the decrease in the SVI is the increased settleability of thesludge due to the pelletization of sludge occurring in the reactor.Increase in organic loading rate resulted in an increase in thesludge bed height. The sludge bed height increased from 14% to18% at 1 kg COD m3 d to about 35–36% at 12 kg COD m3 d in thereactor. The average size of the sludge granule before treatmentwas found to be 0.20 mm in diameter and after the treatmentwas found to be 1.8 mm. Ash content of the granules decreasedwith increase in the organic loading. Ash content of the granulesdecreased from 15–20% at 1 kg COD m3 d to 11–13% at12 kg COD m3 d. The observed ash fraction in the sludge is inagreement with the values reported earlier (Sreekanth et al.,2008). There are different groups of bacteria belonging to differentshapes in anaerobic granules. According to Michael (2003), metha-nogens could be present as rod, curved, spiral and coccus or spher-ical forms and their growth pattern may be irregular clusters orfilamentous chains. In the present study SEM observations demon-strated that Methanosarcina like coccoides were present in abun-dance, where as methanothrix like bamboo shaped rods existed

Table 1Performance of UASB at different organic shock loadings.

Organic shockloads

Inlet OLR(kg COD/m3 d)

Outlet OLR(kg COD/m3 d)

No. of days of operationat each OLR

CODreduction(%)

Gas production per unit CODreduced (ml/mg)

Specific methanogenic activity(ml-CH4/g-VSS d)

Shock load at5 OLR

2.0 0.7 9 65 0.4 356

5.0 4.0 5 20 0.2 902.0 0.8 7 60 0.4 350

Shock load at9 OLR

9.0 2.6 7 71 0.4 345

13.0 10.0 5 23 0.2 859.0 2.5 5 72 0.4 350

2538 D. Sreekanth et al. / Bioresource Technology 100 (2009) 2534–2539

to a far smaller extent. It is well known that the predominance ofmethanothrix in granular sludge is most essential for the establish-ment of a high performance UASB process (Hulshoff Pol et al.,1983). Although we attempted to induce the prevalence of met-hanothrix according to the start-up guideline proposed by Weigantand Deman (1986), the bacterium did not become a predominantmethanogen in our thermophilic granules. Probably thermophilicmethanothrix is more sensitive to the refractory or toxic com-pounds in the bulk drug pharmaceutical wastewater than the ther-mophilic Methanosarcina (Rintala et al., 1993). The temperatureoptimum for the growth of thermophilic Methanosarcina was re-ported to be in the range of 50–58 �C, while that for thermophilicmethanothrix ranged from 60–65 �C. Methanosaetae like bacteriawere also present in biomass.

3.4. Cost estimation

Cost estimation was done for a UASB reactor on the basis ofwastewater generation. The overall costs were represented bythe sum of the capital costs, the operating and maintenance costs.For a full-scale system these costs strongly depend on the natureand the concentrations of the pollutants, the flow rate of the efflu-ent and the configuration of the reactor. An estimation of costs hasbeen made regarding the operating costs for the treatment processused for the treatment of bulk drug pharmaceutical industrialwastewaters. The operational costs were (US$/day): glucose (0.4),electricity (1.8) and nutrients (0.4). The total estimated cost was2.6 US$ for treating 50 m3/day bulk drug pharmaceutical industrialwastewaters. This cost was very competitive when considering thetreated water for some potential reuse applications. This value wasnearly similar to that of our previous findings for treating syntheticwastewater containing phenolic compounds (Sreekanth et al.,2008).

4. Conclusion

The anaerobic hybrid UASB reactor is an appropriate option forthe treatment of wastewater with high organic concentration suchas bulk drug pharmaceutical wastewater under thermophilic con-ditions. It provides efficient organic removal efficiencies, evenwhen operated at high organic loading rates and under intermit-tent operation. Moreover, under stressed operating conditions,such as organic overloads, the reactor shows quite stable perfor-mance. The capacity to handle overloading is an added advantagefor such an industrial application where both quantity and qualityof the wastewater vary widely due to continuous operations. Effi-cient performance of the reactor up to an OLR of 9 kg COD m3 dis demonstrated, where 65–75% COD and 80–90% of BOD reductionwas observed. The reactor is found to withstand shock loads and isreverted to normal performance with in 4–5 days. Biogas produc-tion is found to be close to the theoretical value and high CODand BOD reductions could be achieved keeping the pH of the reac-

tor within the range. Scanning electron micrographs show that thegranules were composed of thermophilic Methanosarcina andthermophilic methanothrix like bacteria.

References

Acharya, B.K., Mohana, S., Madamwar, D., 2008. Anaerobic treatment of distilleryspent wash – A study on upflow anaerobic fixed film bioreactor. Bioresour.Technol. 99, 4621–4626.

Ahn, J.H., Forster, C.F., 2002. A comparison of mesophilic and thermophilicanaerobic up flow filter treating paper-pulp liquors. Process Biochem. 38,256–261.

APHA, 2000. Standard Methods for the Examination of Water and Wastewater, 20thed. American Public Health Association, Washington, DC.

Berardino, S.D., Costa, S., Converti, A., 2000. Semi-continuous anaerobic digestion ofa food industry wastewater in an anaerobic filter. Bioresour. Technol. 71, 261–266.

Carballa, M., Omil, F., Alder, A.C., Lema, J.M., 2006. Comparison between theconventional anaerobic digestion of sewage sludge and its combination with achemical or thermal pre-treatment concerning the removal of pharmaceuticalsand personal care products. Water Sci. Technol. 53, 109–117.

Chelliapan, S., Wilby, T., Sallis, P.J., 2006. Performance of an up flow anaerobic stagereactor (UASR) in the treatment of pharmaceutical wastewater containingmacrolide antibiotics. Water Res. 40, 507–516.

Chen, Z., Ren, N., Wang, A., Zhang, Z.P., Shi, Y., 2008. A novel application of TPAD–MBR system to the pilot treatment of chemical synthesis-based pharmaceuticalwastewater. Water Res. 42, 3385–3392.

Clara, M., Strenn, B., Kreuzinger, N., 2004. Carbamazepine as a possibleanthropogenic marker in the aquatic environment: investigations on thebehaviour of Carbamazepine in wastewater treatment and duringgroundwater infiltration. Water Res. 38, 947–954.

Elliott, A., Mahmood, T., 2007. Pretreatment technologies for advancing anaerobicdigestion of pulp and paper biotreatment residues. Water Res. 41, 4273–4286.

Enick, O.V., Moore, M.M., 2007. Assessing the assessments: pharmaceuticals in theenvironment. Environ. Impact Assess. Rev. 27, 707–729.

Fent, K., Weston, A.A., Caminada, D., 2006. Ecotoxicology of humanpharmaceuticals. Aquat. Toxicol. 76, 122–159.

Fernandez, J.M., Omil, F., Mendez, R., Lema, J., 2001. Anaerobic treatment offireboard manufacturing wastewaters in a pilot scale hybrid USBF reactor.Water Res. 35, 4150–4158.

Gangagni Rao, A., Naidu, G.V., Prasad, K.K., Rao, N.C., Mohan, S.V., Jetty, A., Sarma,P.N., 2005. Anaerobic treatment of wastewater with high suspended solids froma bulk drug industry using fixed film reactor (AFFR). Bioresour. Technol. 96, 87–93, 246.

Hulshoff Pol, L.W., de Zeeuw, W.J., Velzeboer, C.T.M., Lettinga, G., 1983. Granulationin UASB reactors. Water Sci. Technol. 15, 291–304.

Jantsch, T.G., Angelidaki, I., Schmidt, J.E., Branade, H., Ahring, B.K., 2002. Anaerobicbiodegradation of spent sulphide liquor in an UASB reactor. Bioresour. Technol.84, 15–20.

Kasapgil, I.B., Selcuk, A., Ince, O., 2002. Effect of chemical synthesis basedpharmaceutical wastewater on performance, acetoclastic methanogenicactivity and microbial population in an up flow anaerobic filter. J. Chem.Technol. Biotechnol. 77, 711–719.

Kulik, N., Trapido, M., Goi, A., Veressinina, Y., Munter, R., 2008. Combined chemicaltreatment of pharmaceutical effluents from medical ointment production.Chemosphere 70, 1525–1531.

Lefebvre, O., Vasudevan, N., Torrijos, M., Thanasekaran, K., Moletta, R., 2006.Anaerobic digestion of tannery soak liquor with an aerobic post-treatment.Water Res. 40, 1492–1500.

Mahmoud, N., 2008. High strength sewage treatment in a UASB reactor and anintegrated UASB-digester system. Bioresour. Technol. 99, 7531–7538.

Michael, H.G., 2003. The Microbiology of Anaerobic Digesters. John Wiley and Sons,Inc., Hoboken, New Jersey.

Mohan, S.V., Prakasham, R.S., Satyavathi, B., Annapurna, J., Ramakrishna, S.V., 2001.Biotreatability studies of pharmaceutical wastewater using an anaerobicsuspended film contact reactor. Wat. Sci. Tech. 43 (2), 271–276.

D. Sreekanth et al. / Bioresource Technology 100 (2009) 2534–2539 2539

Nandy, T., Kaul, S.N., 2001. Anaerobic pre-treatment of herbal based pharmaceuticalwastewater using fixed film reactor with recourse to energy recovery. WaterRes. 135 (2), 351–362.

Oktem, Y.A., Ince, O., Sallis, P., Donnelly, T., Ince, B.K., 2007. Anaerobic treatment of achemical synthesis-based pharmaceutical wastewater in a hybrid upflowanaerobic sludge blanket reactor. Bioresour. Technol. 99, 1089–1096.

Park, Y., Hong, F., Cheon, J., Hidaka, T., Tsuno, H., 2008. Comparison of thermophilicanaerobic digestion characteristics between single-phase and two-phasesystems for kitchen garbage treatment. J Biosci. Bioeng. 105, 48–54.

Rajeswari, K.V., Balakrishnan, M., Kansal, A., Lata, K., Kishore, U.V.N., 2000. State ofthe art of anaerobic digestion technology for industrial waste water treatment.Renew. Sust. Energ. Rev. 4, 135–156.

Reddy, V.V., Swamy, N.K., Kataiah, B., Ramalingaiah, B., 1995. Third InternationalConference, Appropriate Waste Management Technologies for DevelopingCountries. National Environmental Engineering Research Institute, Nagpur, pp.25–26.

Rintala, J., Lepisto, S., Ahring, B.K., 1993. Acetate degradation at 70 �C in up flowanaerobic sludge blanket reactors and temperature response of granules grownat 70 �C. Appl. Environ. Microbiol. 59, 1742–1746.

Somasiri, W., Li, X.F., Quan Ruan, W., Jian, C., 2008. Evaluation of the efficacy ofupflow anaerobic sludge blanket reactor in removal of colour and reduction ofCOD in real textile wastewater. Bioresour. Technol. 99, 3692–3699.

Sreekanth, D., Sivaramakrishna, D., Himabindu, V., Anjaneyulu, Y., 2008.Thermophilic degradation of phenolic compounds in lab scale hybrid up flowanaerobic sludge blanket reactors. J. Hazard. Mater.. doi:10.1016/J.Jhazmat.2008.09.070.

Stabnikova, O., Liu, X.Y., Wang, J.Y., 2008. Digestion of frozen/thawed food waste inthe hybrid anaerobic solid–liquid system. Waste Manage. 28, 1654–1659.

Suman Raj, D.S., Anjaneyulu, Y., 2005. Evaluation of biokinetic parameters forpharmaceutical wastewaters using aerobic oxidation integrated with chemicaltreatment. Process Biochem. 40 (1), 165–175.

Torkian, A., Eqbali, A., Hashemian, S.J., 2003. The effect of organic loading rate on theperformance of UASB reactor treating slaughterhouse effluent. Resour. Conserv.Recy. 40, 1–11.

Weigant, W.M., De Man, A.W.A., 1986. Granulation of biomass in thermophilicupflow anaerobic sludge blanket reactors treating acidified wastewaters.Biotechnol. Bioeng. 28, 718–727.