Psychrophilic and mesophilic anaerobic digestion of brewery effluent: A comparative study

ARTICLE IN PRESS WAT E R R E S E A R C H

40 (2006) 2503 – 2510

Available at www.sciencedirect.com

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

Psychrophilic and mesophilic anaerobic digestion of brewery effluent: A comparative study Sean Connaughton, Gavin Collins, Vincent O’Flaherty Microbial Ecology Laboratory, Department of Microbiology and Environmental Change Institute (ECI), National University of Ireland, Galway (NUI, Galway), University Road, Galway, Ireland

art i cle info

A B S T R A C T

Article history:

Two expanded granular sludge bed-anaerobic filter (EGSB-AF) bioreactors (3.38 l active

Received 14 November 2005

volume) were used to directly compare psychrophilic (15 1C), anaerobic digestion (PAD) to

Received in revised form

mesophilic (37 1C) anaerobic digestion (MAD) for the treatment of a brewery wastewater

21 April 2006

(chemical oxygen demand (COD) concentration of 31367891 mg l
1). Bioreactor perfor-

Accepted 27 April 2006

mance was evaluated by COD removal efficiency and biogas yields at a range of hydraulic and organic loading rates. Specific methanogenic activity (SMA) assays were also employed

Keywords:

to investigate the activity of the biomass in the bioreactors. No significant difference in the

Biogas

COD removal efficiencies (which ranged from 85–93%) were recorded between PAD and

Brewery wastewater

MAD during the 194-d trial at maximum organic and hydraulic loading rates of 4.47 kg m
3

EGSB

day
1 and 1.33 m3 m
3 day
1, respectively. In addition, the methane content (%) of the

Granular sludge

biogas was very similar. The volumetric biogas yield from the PAD bioreactor was

Psychrophilic anaerobic digestion

approximately 50% of that from the MAD bioreactor at an organic loading rate of

Specific methanogenic activity

4.47 kg COD m
3 day
3 and an applied liquid up-flow velocity (Vup) of 2.5 m h
1. Increasing the Vup in the PAD bioreactor to 5 m h
1 resulted in a volumetric biogas production rate of approximately 4.1 l d
1 and a methane yield of 0.28 l CH4 g
1 COD d
1, which were very similar to the MAD bioreactor. Significant and negligible biomass washout was observed in the mesophilic and psychrophilic systems, respectively, thus increasing the sludge loading rate applied to the former and underlining the robustness of the latter, which appeared underloaded. A psychrotolerant mesophilic, but not truly psychrophilic, biomass developed in the PAD bioreactor biomass, with comparable maximum SMA values to the MAD bioreactor biomass. PAD, therefore, was shown to be favourably comparable to MAD for brewery wastewater treatment and biogas generation. & 2006 Elsevier Ltd. All rights reserved.

1.

Introduction

Anaerobic digestion (AD) is now an established and proven technology for the effective treatment of a multitude of industrial wastewater categories (Driessen and Yspeert, 1999; Macarie, 2000; Elmitwalli et al., 2001; Bouallagui et al., 2005; Rincon et al., 2006). However, the majority of full-scale applications and research effort, until recently, has been Corresponding author. Tel.: +353 0 91 493734; fax: +353 0 91 494598.

E-mail address: [email protected] (V. O’Flaherty). 0043-1354/$ - see front matter & 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2006.04.044

concentrated on AD within the mesophilic (25–45 1C) or thermophilic (45–65 1C) temperature ranges. This was largely due to the belief that sub-ambient or psychrophilic (o20 1C) AD (PAD) was not viable because of low microbial activity and biogas production rates under low-temperature conditions (Lin et al., 1987; Lettinga et al., 2001). Despite this, the majority of industrial effluents are discharged at low-ambient temperatures (Lettinga et al., 2001). As a consequence, one of the

ARTICLE IN PRESS 2504

WAT E R R E S E A R C H

40 (2006) 2503– 2510

main advantages of psychrophilic anaerobic wastewater treatment would be increased cost-efficiency, as the need to heat influent wastewaters or to direct AD-produced energy back into system maintenance (e.g. bioreactor heating) is reduced or eliminated. The use of new or modified bioreactor designs, such as various versions of the up-flow anaerobic sludge bed (UASB), internal circulation (IC), expanded granular sludge bed (EGSB) and EGSB-anaerobic filter (EGSB-AF) bioreactors, has, in part, facilitated the successful demonstration of PAD at laboratory scale for the treatment of a wide variety of wastewater categories (Rebac et al., 1995; Lettinga, 1999a; Collins et al., 2003; McHugh et al., 2004; Enright et al., 2005). The brewery industry consumes and produces significant volumes of process water and wastewater, respectively, resulting in water:beer:wastewater ratios ranging from 4–11:1:2–8 m3 for each m3 of beer produced (Driessen and Vereijken, 2003). To date, a number of laboratory and full-scale trials have been carried out with both synthetic and natural brewery effluents, which have concentrated on the applicability of AD to the biomethanation of brewery wastewater and on operational parameters, such as sludge-type employed, bioreactor configuration, hydraulic retention times and organic loading rates (Cronin and Lo, 1998; Ochieng et al., 2002; Driessen and Vereijken, 2003; Parawira et al., 2005). Some breweries that have set up in-house AD wastewater treatment processes have chosen either thermophilic or mesophilic operational temperatures (Harada et al., 1996; Leal et al., 1998; Parawira et al., 2005; Akarsubasi et al., 2006). However, some laboratory-scale work has also been carried out at ambient temperatures, which has illustrated that low-temperature AD of brewery effluents is feasible and can now be considered as an alternative to thermophilic or mesophilic AD (Yu and GU, 1996; Cronin and Lo, 1998). Despite the abundance of research into the stabilisation of brewery effluents, no data have yet been reported from a direct comparison between mesophilic and psychrophilic AD of brewery wastewater. Recently, Enright et al. (2005) recommended that research be carried out to directly compare mesophilic and psychrophilic AD treatment and, in particular, to evaluate the potential of PAD for bioenergy production. This research is imperative if PAD is to be established as a viable treatment alternative within the wider field of industrial wastewater treatment. The study and data presented in this paper offer a comparison between mesophilic AD (MAD) and PAD. The aim of this study was to assess the process performance of two EGSB-AF bioreactors inoculated from the same seed sludge source and to treat a brewery wastewater. One of the bioreactors was operated at 37 1C, while the second was maintained at 15 1C.

2.

Materials and methods

2.1.

Source of biomass

A mesophilic anaerobic sludge was obtained from a full-scale, granular biomass nursery plant operated at 37 1C in the

Netherlands (Paques B.V.). The sludge consisted of wellsettling black granules ranging in size from 1–3 mm in diameter and had a volatile suspended solids (VSS) content of 73 g l
1.

2.2.

Bioreactor design and operation

Two identical 3.73 l (active liquid volume, 3.38 l) glass laboratory scale expanded granular sludge bed-anaerobic filter (EGSB-AF) bioreactors, B1 and B2, designed as described by Collins et al. (2005a), were each inoculated with 1.1 l of the granular sludge. This volume of seed sludge provided each bioreactor with 23.8 g VSS l
1. The influent feed supplied to both bioreactors was procured from Beamish & Crawford, Cork, Ireland. This brewery has a production capacity of over 50,000 m3 per annum, a beer to wastewater ratio of 1:6 and generates between 300,000–420,000 m3 of process effluent annually. The effluent chemical oxygen demand (COD) discharged from this brewery ranges between 1000–6000 mgl
1 depending on product type and volume produced. However, for the period covered by this trial, the average COD concentration was 31367890.9 mg l
1 and the mean pH was 7.270.45 (Figs. 1 and 2; Table 1). The treatment trial was divided into four different operational periods, P1–P4. Each period was characterised by a change in either the hydraulic retention time (HRT) or applied liquid up-flow velocity (Vup; Table 1). B1 and B2 were operated for a total trial period of 194 d at 37 1C and 15 1C70.5 1C, respectively.

2.3.

Specific methanogenic activity (SMA) assays

SMA assays were performed as described by Colleran et al. (1992) and Coates et al. (1996) using the seed inoculum and granular and fixed-film biomass samples recovered from the bioreactors at the conclusion of the trial (Table 2). The substrates tested, and the concentrations used, were acetate (30 mM), butyrate (15 mM), propionate (30 mM), ethanol (30 mM) and H2/CO2 (80:20 v/v) as described in greater detail by Collins et al. (2003).

2.4.

Analytical techniques

Samples of bioreactor influent/effluent and biogas were routinely sampled for COD/pH and methane determinations, respectively, according to Standard Methods American Public Health Association (APHA, 1995). Biogas production volumes (biogas yield) were recorded using a wetgas meter designed and manufactured by Centre Point Electronics, Galway, Ireland. Methane yield coefficient (MYC) values (expressed as l [CH4 produced] g [COD removed]
1) were calculated according to methods reported by Borja et al. (2004) and Rincon et al. (2006). Methane yield efficiency values were derived from the established stoichiometric value of 0.35 l [CH4 produced] g [COD removed]
1 equalling 100% efficiency as reported by Lawrence and McCarty (1969).

ARTICLE IN PRESS WAT E R R E S E A R C H

4 0 (200 6) 250 3 – 251 0

2505

Fig. 1 – COD removal and COD concentrations of influent and effluents from B1 (A, the mesophilic bioreactor) and B2 (B, the psychrophilic bioreactor): % COD removal (-m-); influent COD concentration (mg l
1; -E-); effluent COD concentration (mg l
1; -K-).

Fig. 2 – Percentage methane in biogas produced by B1 (-E- the mesophilic bioreactor) and B2 (-m- the psychrophilic bioreactor).

3.

Results

3.1.

Bioreactor performance

A rapid start-up was observed in both bioreactors during the first operational period (P1). This was most apparent in B2

(15 1C), which achieved higher COD removal rates and a higher biogas methane content (Figs. 1 and 3). At the beginning of P2, a decline was observed in the performance of both bioreactors. This period, combined with the applied process changes, was further characterised by the most variable influent COD concentration (1700–5700 mg l
1; Fig. 1). This influent COD variability coincided with erratic

ARTICLE IN PRESS 2506

WAT E R R E S E A R C H

40 (2006) 2503– 2510

Table 1 – Operational parameters, conditions and results for both bioreactors during the four experimental periods. Operational periods Periods days Bioreactor-temp (1C) Influent CODa (mg I
1) OLRb (kg COD m
3 d
1) OLRb (kg COD kg VSS
1 d
1) HRIc (h)

P1

P2

P3

P4

P1

P2

P3

P4

0–53 B1–37 1C 32317692.9 1.62 0.07

54–96 B1–37 1C 291471226 2.91 NR

97–133 B1–37 1C 33627819.3 4.47 NR

134–194 B1–37 1C 30377785.5 4.04 2.7

0–53 B2–15 1C 32317692.9 1.62 0.07

54–96 B2–15 1C 291471226 2.91 NR

97–133 B2–15 1C 33627819.3 4.47 NR

134–194 B2–15 1C 30377785.5 4.04 0.17

48

24

18

18

48

24

18

18

VLRd (m3 Ww m
3 d
1) SLRe (m3 Ww kg VSS
1 d
1) Up-flow velocity (m h
1) COD removal (%) Influent pH

0.5 0.002

1 NR

1.33 NR

1.33 0.89

0.5 0.002

1 NR

1.33 NR

1.33 0.06

2.5 87.175.9 7.170.4

2.5 85.5712.7 7.270.8

2.5 89.875.1 7.270.3

3 90.476.6 7.270.3

2.5 92.674.6 7.170.4

2.5 84.179.7 7.270.8

2.5 85.677.3 7.270.3

5 88.777.2 7.270.4

Effluent pH Biogas flow rates (ml h
1) CH4 in biogas (%)

8.470.3 NR 57.8718.1

8.170.3 NR 68.1716.6

7.870.5 203.4737.1 70.876.9

8.170.2 184.8726.8 71.275

8.370.3 NR 72.3711.2

7.870.3 NR 73.4713.1

7.970.3 101714.4 74.475.6

870.2 170.7725.3 73.875.7

All values are the period mean7period standard deviation; NR ¼ Not recorded; Ww ¼ Wastewater. a Chemical oxygen demand. b Organic loading rate. c Hydraulic retention time. d Volumetric loading rate. e Sludge loading rate.

Table 2 – Specific methanogenic activity of the seed inoculum and biomass samples from B1 and B2 during the trial period (ml CH4(g VSS
1 d
1)) Bioreactor OT (1C)

Biomass

Test temp (1C)

Test day

Ethanol

Acetate

Propionate

Butyrate

H2/CO2

— — 37 37 37 37

Granular inoculum Granular inoculum Granular sludge Granular sludge fixed-film fixed-film

37 15 37 15 37 15

0 0 194 194 194 194

158.8 (14.4) 12 (0.5) 461.9 (8.5) 41.2 (0.6) 261.8 (8.4) 9 (0.4)

78.2 (14.4) 18.3 (2.5) 185.9 (3.3) 23.2 (2.5) 63.4 (3.3) 8 (0.4)

114.2 (18.6) 4.3 (0.2) 91 (9.3) 5.4 (0.3) 28.9 (1.4) 1 (0.1)

30.8 (12.6) 7.7 (0.1) 132.3 (10.3) 15.9 (0.3) 49.4 (2.1) 2.4 (0.1)

57.3 (13.6) 8.9 (2.2) 422.2 (2.8) 61.6 (0.6) 311.9 (3.9) 44.2 (0.2)

15 15 15 15

Granular sludge Granular sludge fixed-film fixed-film

37 15 37 15

194 194 194 194

587.3 (10.9) 78.5 (1.5) 386.4 (9.2) 3.4 (0.6)

378.8 (6.6) 75.1 (2.4) 183.4 (1.4) 31.5 (5.1)

42.6 (1.3) 6.1 (0.3) 23.3 (4.8) 2.8 (0.8)

36 (2.5) 9.7 (0.8) 59.1 (1.6) 7.7 (0.8)

250.2 (16.4) 79.8 (4) 472.4 (5.7 28.4 (11)

OT—Bioreactor operational temperature. All values are the mean of triplicates with standard errors (standard deviation/On, where n is 3) in the parentheses.

process performance displayed by both bioreactors. Specifically, COD removal during P2 was greater in B1 than B2 (Table 1; Fig. 1); however, the treatment efficiency in B1 was more variable (variance (s2); 152.7) than in B2 (89.6) . P3 commenced with the final HRT reduction of the trial (24–18 h) and, initially, both bioreactors responded positively to the applied pressures, which was evident by a steady increase in process performance. However, towards the end of P3 a reduction in COD removal and biogas production was observed in both bioreactors (Figs. 1 and 3). This coincided with a gradual decrease in the COD concentration of the influent feed (Fig. 1). For the final period (P4), the HRT was maintained at 18 h but the Vup was increased to 3 and 5 m h
1 for B1 and B2, respectively. During the early stages of P4 both bioreactors demonstrated a recovery in process performance (Fig. 1; Table 1). Overall process stability was observed by the

conclusion of the trial in both bioreactors (Figs. 1 and 3) when only a marginal difference was observed in COD removal and biogas yields between both bioreactors (Figs. 1 and 3; Table 1).

3.2.

Biogas production and yields

Periods 1 and 2 were characterised by erratic process performance, illustrated by fluctuating COD removal and biogas methane concentrations (Figs. 1 and 3). However, from the start of P3 the methane content of the biogas from both bioreactors stabilised considerably (Fig. 2; Table 1). This may have been due to ongoing acclimatisation independent of the changes in operating conditions. Nevertheless, it was observed that the biogas methane concentration was higher from the psychrophilic bioreactor throughout the entire trial (Fig. 1; Table 1). Furthermore, from the commencement of P3,

ARTICLE IN PRESS WAT E R R E S E A R C H

2507

4 0 (200 6) 250 3 – 251 0

Fig. 3 – Biogas production and yields from B1 (-E- the mesophilic bioreactor) and B2 (-m- the psychrophilic bioreactor).

Table 3 – Biogas production volumes and methane yields during periods 3 and 4 for both bioreactors Bioreactor Operational periods Periods days Operational temp (1C) Biogas yield (1 d
1) CH4 yield (l d
1) VLRa (l Ww l
1 d
1) Influent CODb (g l
1) COD loading rate (g l
1 d
1) COD removal rate (g l
1 d
1) MYCc (lCH4 g [COD]
1 d
1) % efficiency (100% ¼ 0.35d)

B1

B1

B2

B2

P3 97–133 37 1C 4.8870.89 3.4670.63 1.33 3.470.82 4.5271.1 3.7170.9 0.28 80

P4 134–194 37 1C 4.4470.64 3.1670.46 1.33 3.0470.77 4.0471 3.4870.9 0.27 77.1

P3 97–133 15 1C 2.4270.35 1.870.26 1.33 3.470.82 4.5271.1 3.570.85 0.15 43

P4 134–194 15 1C 4.170.61 3.0370.45 1.33 3.0470.77 4.0471 3.270.79 0.28 80

All values are the period mean7standard deviation. Ww—Waste water. a Volumetric loading rate. b Chemical oxygen demand. c Methane yield coefficient. d The stoichiometric value at STP of 0.35 l CH4 produced g
1 COD removed.

and for the remainder of the trial, biogas volumes were also recorded (Fig. 3; Table 3). During P3, the mesophilic bioreactor produced almost twice as much biogas per day as the psychrophilic bioreactor (4.88 versus 2.42 l d
1; Fig. 3 and Table 3), thus offsetting the lower biogas quality from B1. However, at the beginning of P4 a clear improvement in biogas production was observed in the psychrophilic bioreactor with an initial 100% increase recorded and an overall increase of 70% by the conclusion of the trial (Fig. 3; Table 3). Conversely, the mesophilic bioreactor displayed a decrease in biogas production towards the end of P3 and, overall, displayed a 10% decrease in biogas production by the conclusion of P4 (Fig. 3; Table 3). The improvement in B2 coincided with the applied increase to the Vup from 2.5–5 m h
1. A lesser increase was applied to B1 (2.5–3 m h
1), due to concerns over granule disintegration and potential biomass wash-out. Prior to the Vup amendment, the MYC during P3 was almost twice as high for the

mesophilic bioreactor as the psychrophilic bioreactor (0.28 and 0.15 l CH4 g [COD]
1 d
1 for B1 and B2, respectively; Fig. 3 and Table 3). However, although the operating conditions of B1 and B2 were different, following the applied change to the Vup, the MYC of the psychrophilic bioreactor surpassed that of the mesophilic bioreactor (Table 3). This was also reflected in the methane yield efficiency values for both bioreactors by the conclusion of the trial (Table 3).

3.3.

SMA profiles

SMA assays carried out with the seed sludge confirmed its mesophilic nature i.e. higher values were obtained from the mesophilic (37 1C) than from the psychrophilic (15 1C) assays (Table 2). Nevertheless, the development of methanogenic activity was evident during the trial under psychrophilic conditions; an exception to this was the propionate-degrading activity of the sludge which decreased during the trial

ARTICLE IN PRESS 2508

WAT E R R E S E A R C H

40 (2006) 2503– 2510

period (Table 2). The final sludge sample taken from B2 and assayed at 37 1C generated higher SMA values for ethanol and acetate degradation, than the corresponding B1 sample. Conversely, higher SMA values at 37 1C for propionate, butyrate and H2/CO2 degradation were recorded from the sample taken from B1. Interestingly, however, all of the 15 1C assays indicated higher SMA values from the B2 than from the B1 samples, with the exception of butyrate degraders (Table 2). The fixed-film sample from B2 generally displayed higher activity than the B1 sample (Table 2). Specifically, SMA values measured at 37 1C were higher for the B2 sample, except for the activity measured against propionate, than the B1 sample (Table 2). This was also true for the B2 sample tested at 15 1C, except for ethanol- and H2/CO2-mediated values (Table 2).

4.

Discussion

The results presented in this study indicate that similar levels of wastewater treatment efficiency can be achieved in cold (o20 1C) anaerobic bioreactors to that already achieved by warm (420 1C) bioreactors, under the conditions applied. In this trial, both bioreactors displayed a rapid start-up and compared well to previous mesophilic trials (e.g. O’Flaherty et al., 1998), including brewery wastewater trials (e.g. Parawira et al., 2005). Interestingly, the start-up time required for the colder bioreactor (B2) was shorter than that for B1. Rapid bioreactor start-up periods at low temperatures have been reported previously using various bioreactor configurations and treating a wide variety of wastewaters (Rebac et al., 1999; Lettinga et al., 1999b; McHugh et al., 2004; Collins et al., 2005a). Despite the difference in start-up times, an overall similarity in process performance between B1 and B2 was observed throughout the trial, emphasising, in this case, the comparability of results generated from PAD to those from MAD (Table 1). Indeed, toward the end of the trial, the analogous performance of B1 and B2 was particularly evident i.e. with respect to values for COD removal, biogas volumes and methane content (Figs. 1–3). Optimum process stability, in terms of influent treatment and methane production, was apparent in both bioreactors at this time. Nonetheless, it was evident, from effluent COD concentrations determined during the trial, that B1 performed marginally better than B2. This difference was more pronounced during P3, when both bioreactors were exposed to the highest influent COD concentrations of the trial and when B1 produced approximately twice the daily volume of biogas produced by the psychrophilic bioreactor. The superior productivity of the mesophilic bioreactor was not unexpected, as the SMA of the seed sludge used to inoculate both bioreactors—which was sourced from a full-scale mesophilic bioreactor—was much higher at 37 1C than 15 1C. The observation that B2 SMA was higher at 37 1C than 15 1C, even after 194 d of operation at 15 1C (Table 2), is consistent with other well-documented data reported on psychrophilic bioreactor trials (Rebac et al., 1995; Collins et al., 2005b; Enright et al., 2005) and we speculate that a longer trial period would have been required for the development of psychrophiles in B2. Despite the absence of psychrophilic populations, the SMA profile of B2 was enhanced under low-

temperature conditions, and activity data (assayed at 15 and 37 1C) of B2 biomass were generally greater than those recorded from the seed sludge. Interestingly, however, the fixed-film samples retrieved from B2 generally exhibited higher SMA across the substrate range at 37 and 15 1C than the corresponding B1 samples. Although we have no direct evidence to support this, we suggest that the upper, fixed-film section of the bioreactor may be important for the complete mineralisation of the brewery wastewater, and that the sludge bed in B2 may, thus, have been primarily involved in hydrolysis and acidification. Indeed, we have previously identified the importance of this fixed-film chamber in low-temperature 2,4,6-trichlorophenol (TCP)degrading EGSB-AF bioreactors, whereby the fixed biofilm contributed to the degradation of intermediate compounds produced by the sludge bed (Collins et al., 2005a). More research will be required to establish the optimum bioreactor configuration required for PAD, but potential applications of our design include the treatment of wastewaters with higher suspended solids concentrations, such as municipal wastewaters. Despite the higher methane concentrations recorded from B2, reported values in the literature regarding methane solubility imply that methane is more soluble in low temperature effluent, thus making it more difficult to harvest from psychrophilic than either mesophilic or thermophilic bioreactors (Lettinga et al., 2001). In any event, and at an identical Vup of 2.5 m h
1 (P3), the mesophilic bioreactor produced approximately twice the volume of biogas as the psychrophilic bioreactor. However, during P4, after the applied Vup increase, biogas production from the psychrophilic bioreactor improved substantially, while only a marginal increase was observed in the mesophilic bioreactor. Under psychrophilic conditions increased mixing within the sludge bed is recommended to facilitate greater substrate-microbe contact (Rebac et al., 1998; Lettinga et al., 2001). We attribute the increased biogas yield in our psychrophilic bioreactor to the stripping effect of the extra mixing associated with the increased Vup. Any improvement in methane yield has obvious economic relevance as it demonstrates that PAD can compete favourably with MAD in terms of bioenergy production. Despite this, and as the same Vup increase was not applied to the mesophilic bioreactor (due to concern over granule disintegration and subsequent biomass washout), a direct comparison cannot be made. Notwithstanding this, the fact that a greater Vup increase could not be applied to B1 highlights a relevant process disadvantage with our MAD system. The applied OLRs, expressed as a function of bioreactor volume, are not adequately informative of the pressures applied to B1 and B2 during the trial. In fact, by the conclusion of the trial, and due to biomass washout, 0.25 l and 1 l of biomass were retained in B1 and B2, respectively, which correlated to a total mass of 5 g VSS and 80 g VSS, respectively. In light of this, the applied OLR was 2.7 and 0.17 kg COD kg [VSS]
1 d
1 in B1 and B2, respectively (Table 1). Thus, the sludge loading rate (SLR) applied to B1 and B2 was 0.89 and 0.06 m3 kg [VSS]
1 d
1, respectively. It is apparent from these data that a far higher load was applied to the mesophilic bioreactor; however, the increased and continued

ARTICLE IN PRESS WAT E R R E S E A R C H

4 0 (200 6) 250 3 – 251 0

washout of biomass from B1 raises questions about the sustainability of the mesophilic system. Importantly, the specific methane production rates (Table 3) for the mesophilic bioreactor throughout the trial and the psychrophilic bioreactor during P4 are comparable to those reported from previous mesophilic trials (e.g. Borja et al., 2002; Rincon et al., 2006). Furthermore, based on the calorific value of methane (55 kJ g
1; Dale et al., 2002), the potential bioenergy harvest from each bioreactor during P4 can be calculated to be 116.6717.1 kJ g [COD]
1 d
1 for B1 and 111.7716.5 kJ g
1 d
1 for B2. These data also compare favourably with other mesophilic trials (e.g. Yu and Gu, 1996) and thus, underline the potential for bioenergy harvesting from full-scale PAD systems. Nevertheless, it is apparent that the potential methane productivity of B1 and B2 as measured in SMA assays, appeared to be much higher than that obtained in both bioreactors; in other words, the data suggest that both bioreactors operated below their maximal capacities which could have accounted for the lack of an obvious difference in operating performance between B1 and B2. However, and importantly, the VSS (biomass) concentration of the bioreactors at the conclusion of the trial should also be considered in assessing bioreactor performance The methane productivity of B1 and B2, at the conclusion of the trial, was 622 and 37.8 ml CH4 g [VSS]
1 d
1, respectively. Thus, it is clear that B1 did achieve a performance close to the theoretical methanogenic activity suggested by SMA assays, while B2 productivity was far lower than the maximum predicted by the batch tests. Thus, although the greater biomass loss in B1, due to granule disintegration, resulted in higher organic and sludge loading, coupled with higher methane productivity, this bioreactor was less physically sustainable than the psychrophilic bioreactor. Furthermore, it is apparent that B2 was underloaded and, given the lower level of biomass loss under similar hydraulic conditions as in B1, we posit that a significantly higher OLR could be applied to the colder bioreactor. Indeed, our previous work has suggested that loading rates of between 25 and 37 kg COD m
3 d
1 can be applied under low-temperature conditions (Connaughton et al., 2006). It is clear that, notwithstanding the underloaded nature of B2, a satisfactory MYC was achieved by the cold bioreactor compared to B1. In addition, however, we propose that in the absence of the biomass disintegration and washout from B1, a significantly higher organic load would be applicable under mesophilic operating conditions. Additionally, our SMA data suggest that the biomass in B2 remained mesophilic (psychrotolerant), and it is unclear whether a higher organic load could be applied if a psychrophilic community developed in B2. The performance of B2 is very interesting, especially in the context of industrial wastewater management, such as in the brewery sector. Management concerns, when considering AD as a wastewater treatment option, include production effluent discharge requirements, effluent quality exiting the treatment system, operating and maintenance costs of the system, energy conservation and potential bioenergy generation from the system (Driessen and Vereijken, 2003; Fillaudeau et al., 2006). We propose, although much work remains

2509

to be done, that this study indicates the suitability of PAD to meet these requirements.

5.

Conclusions

The following conclusions can now be drawn: (1) this investigation has demonstrated that PAD of brewery wastewater is highly effective for both COD removal and bioenergy generation; (2) biogas yields from the PAD bioreactor were comparable with those from the MAD bioreactor with similar COD removal; (3) lower levels of biomass loss indicated that the psychrophilic bioreactor tested was more robust than the warmer bioreactor; thus, at low temperatures, it may be possible to apply higher organic loads than to equivalent mesophilic bioreactor volumes; (4) the potential energy conservation and bioenergy generation achievable through PAD appears very promising for the future. All of the above are particularly relevant as industrial energy prices steadily increase and a greater emphasis on environmental protection is continually being demanded both by policy makers and the general public.

Acknowledgements The authors wish to sincerely thank the Beamish and Crawford Brewery, Cork, Ireland and gratefully acknowledge Mr. Dick Ryan for his technical assistance. The receipt of financial support from the Irish Environmental Protection Agency, Enterprise Ireland, and a research scholarship from Sligo County Council to S.C., are also gratefully acknowledged. R E F E R E N C E S

American Public Health Association, 1995. Standard Methods for the Examination of Water and Wastewater. 19th ed., APHA/ American Water Works Association/Water Environment Federation, Washington, DC, USA. Akarsubasi, A.T., Ince, O., Oz, N.A., Kırdar, B., Ince, B.K., 2006. Evaluation of performance, acetoclastic methanogenic activity and archaeal composition of full-scale UASB reactors treating alcohol distillery wastewaters. Process Biochem. 41 (1), 28–35. Bouallagui, H., Touhami, Y., Cheikh, B., Hamdi, M., 2005. Bioreactor performance in anaerobic digestion of fruit and vegetable wastes. Process Biochem. 40, 989–995. Borja, R., Rincon, B., Raposo, F., Alba, J., Martin, A., 2002. A study of anaerobic digestion of two-phase olive mill solid waste (OMSW) at mesophilic temperatures. Process Biochem. 38, 733–742. Borja, R., Rincon, B., Raposo, F., Dominguez, J.R., Millan, F., Martin, A., 2004. Mesophilic anaerobic digestion in a fluidised-bed reactor of wasterwater from the production of protein isolates from chickpea flour. Process Biochem. 39, 1913–1921. Coates, J.D., Coughlan, M.F., Colleran, E., 1996. Simple method for the measurement of the hydrogenotrophic methanogenic activity of anaerobic sludges. J. Microbiol. Meth. 26, 237–246. Colleran, E., Concannon, F., Golden, T., Geoghan, F., Crumlish, B., Killilea, J.E., Henry, M., Coates, J., 1992. Use of methanogenic activity tests to characterize anaerobic sludges, screen for anaerobic biodegradability and determine toxicity thresholds

ARTICLE IN PRESS 2510

WAT E R R E S E A R C H

40 (2006) 2503– 2510

against individual anaerobic trophic groups and species. Water Sci. Technol. 25, 31–40. Collins, G., Woods, A., McHugh, S., Carton, M., O’Flaherty, V., 2003. Microbial community structure and methanogenic activity during start-up of psychrophilic anaerobic digesters treating synthetic industrial wastewaters. FEMS Microbiol. Ecol. 46, 159–170. Collins, G., Foy, C., McHugh, S., O’Flaherty, V., 2005a. Anaerobic treatment of 2,4,6-trichlorophenol in an expanded granular sludge bed-anaerobic filter (EGSB-AF) reactor at 15 1C FEMS Microbiol. Ecol. 53, 167–178. Collins, G., Foy, C., McHugh, S., Mahony, T., O’Flaherty, V., 2005b. Anaerobic biological treatment of phenolic wastewater at 15–18 1C. Water Res. 39, 1620–1640. Connaughton, S., Collins, G., O’Flaherty, V., 2006. Development of microbial community structure and activity in a high-rate anaerobic bioreactor at 18 1C. Water Res. 40, 1009–1017. Cronin, C., Lo, K.V., 1998. Anaerobic treatment of brewery wastewater using UASB reactors seeded with activated sludge. Biores. Technol. 64, 33–38. Dale, A., Lythall, C., Aucoot, J., Sayer, C., 2002. High precision calorimetry to determine the enthalpy of combustion of methane. Thermochim. Acta 382, 47–54. Driessen, W., Yspeert, P., 1999. Anaerobic treatment of low, medium and high strength effluent in the agro-industry. Water Sci. Technol. 40 (8), 221–228. Driessen, W., Vereijken, T., 2003. Recent developments in biological treatment of brewery effluent. The Institute and Guild of Brewing Convention, Zambia. Elmitwalli, T.A., Soellner, J., de Keizer, A., Bruning, H., Zeeman, G., Lettinga, G., 2001. Biodegradability and change of physical characteristica of particles during anaerobic digestion of domestic sewage. Water Res. 35, 1311–1317. Enright, A.M., McHugh, S., Collins, G., O’Flaherty, V., 2005. Low-temperature anaerobic biological treatment of solventcontaining pharmaceutical wastewater. Water Res. 39, 4587–4596. Fillaudeau, L., Blanpain-Avet, P., Daufin, G., 2006. Water, wastewater and waste management in brewery industries. J. Cleaner Prod. 14 (5), 463–471. Harada, H., Uemura, S., Chen, A.C., Jayadevan, J., 1996. Anaerobic treatment of a recalcitrant distillery wastewater by a thermophilic UASB reactor. Biores. Technol. 55, 215–221. Lawrence, A.W., McCarty, P.L., 1969. Kinetics of methane fermentation in anaerobic treatment. J. Water Pollut. Contr. Fed. 41, R1–R16. Leal, K., Chacin, E., Behling, E., Gutierrez, E., Fernandez, N., Forster, C.F., 1998. A mesophilic digestion of brewery wastewater in an unheated anaerobic filter. Biores. Technol. 65, 51–55.

Lettinga, G., 1999a. Psychrophilic anaerobic treatment of low strength wastewaters. Water Sci. Technol. 39, 203–210. Lettinga, G., Rebac, S., Parshina, S., Nozhevnikova, A., van Lier, J.B., Stams, A.J.M., 1999b. High-rate anaerobic treatment of wastewater at low temperatures. Appl. Environ. Microbiol. 65, 1696–1702. Lettinga, G., Rebac, S., Zeeman, G., 2001. Challenge of psychrophilic anaerobic wasterwater treatment. Trends Biotechnol. 19 (9), 363–370. Lin, C.Y., Noike, T., Sato, K., Matsumoto, J., 1987. Temperature characteristics of the methanogenesis process in anaerobic digestion. Water Sci. Technol. 19, 299–310. Macarie, H., 2000. Overview of the application of anaerobic treatment to chemical and petrochemical wastewaters. Water Sci. Technol. 42 (5–6), 201–214. McHugh, S., Carton, M., Collins, G., O’Flaherty, V., 2004. Reactor performance and microbial community dynamics during anaerobic biological treatment of wastewater at 16–37 1C. FEMS Microbiol. Ecol. 48, 369–378. Ochieng, A., Ogada, T., Sisenda, W., Wambua, P., 2002. Brewery wastewater treatment in a fluidised bed bioreactor. J. Hazard. Mater. 311–321. O’Flaherty, V., Lens, P., Leahy, B., Colleran, E., 1998. Long-term competition between sulphate-reducing and methane-producing bacteria during full-scale anaerobic treatment of citric acid production wastewater. Water Res. 32, 815–825. Parawira, W., Kudita, I., Nyandoroh, M.G., Zvauya, R., 2005. A study of industrial anaerobic treatment of opaque beer brewery wastewater in a tropical climate using a full-scale UASB reactor seeded with activated sludge. Process Biochem. 40, 593–599. Rebac, S., Ruskova, J., Gerbens, S., van Lier, J.B., Stams, A.J.M., Lettinga, G., 1995. High-rate anaerobic treatment of wastewater under psychrophilic conditions. J. Ferm. Bioeng. 80, 499–506. Rebac, S., van Lier, J.B., Lens, P., van Cappellen, J., Vermuelen, M., Stams, A.J.M., Dekkers, F., Swinkels, K.T.M., Lettinga, G., 1998. Psychrophilic (6–15 1C) high-rate anaerobic treatment of malting wastewater in a two module EGSB system. Biotechnol. Prog. 14, 856–864. Rebac, S., van Lier, J.B., Lens, P., Stams, A.J.M., Dekker, F., Swinkels, K. Th. M., Lettinga, G., 1999. Psychrophilic anaerobic treatment of low strength wastewater. Water Sci. Technol. 39, 203–210. Rincon, B., Raposo, F., Borja, R., Gonzalez, J.M., Portillo, M.C., SaizJimenez, C., 2006. Performance and microbial communities of a continuous stirred tank anaerobic reactor treating twophase olive mill solid wastes at low organic loading rates. J. Biotechnol. 121 (4), 534–543. Yu, H., Gu, G., 1996. Biomethanation of brewery wastewater using an anaerobic upflow blanket filter. J. Cleaner Prod. 4, 219–223.