- Email: [email protected]
Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration
Journal of Cleaner Production xxx (2016) 1e8
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration a, *, Federico Castagnola a, Maria Cristina Collivignarelli a, Alessandro Abba Giorgio Bertanza b a b
Department of Civil Engineering and Architecture, University of Pavia, Via Ferrata 1, 27100, Pavia, Italy Department of Civil, Environmental, Architectural Engineering and Mathematics, University of Brescia, Via Branze 43, 25123, Brescia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 September 2016 Received in revised form 19 December 2016 Accepted 19 December 2016 Available online xxx
The increase of sewage sludge production together with the high treatment and disposal costs in the last years has pushed to study different solutions aimed at sludge minimization. In this paper, the thermophilic membrane technology was evaluated as an alternative for municipal sewage sludge reduction. The experimentation (carried out by means of a pilot scale plant, 1 m3 volume) was divided into two steps: the first one was aimed at confirming the results obtained in a previous research focused on industrial sludge treatment; the second step was devoted to define the best process conditions (in particular the optimization of the aeration phases) and to the chemical permeate characterization. The results of the experiments highlight that the hydraulic retention time (HRT) and aeration conditions play a crucial role on the overall process performance. The volatile suspended solids removal efficiency was greater than 80% under the following conditions: HRT even lower than 15 d; 2 h of aeration - 6 h of non aeration cycles; and organic loading rate of 2.0 kgCOD m3 d1. The permeate showed a good biodegradability under mesophilic conditions thus being treatable by means of conventional biological processes. Moreover, ammonia (the permeate presenting high concentrations) could be recovered as a fertilizer (stripping and subsequent washing of the exhausted gas is an established technique). Finally, the ammonia-free permeate can be valorised as a carbon source in denitrification processes. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Intermittent aeration Membrane biological reactor Sludge minimization Thermophilic biological treatment Ultrafiltration
1. Introduction The sewage sludge produced by conventional activated sludge (CAS) wastewater treatment plants (WWTPs) is characterized by a high organic content and the presence of heavy metals, organic contaminants and/or pathogens, depending on the nature of the treated wastewater and on the process conditions. In Europe, the yearly production of sewage sludge is estimated at 10.13 million tons of dry matter (Collivignarelli et al., 2015a) and it is expected to reach the amount of 13 million tons (Kelessidis and Stasinakis, 2012) by 2020. Moreover, the sludge treatment and disposal costs can be equal to 50e60% of the total operating costs of a WWTP (Campos et al., 2009). Nowadays, several sewage sludge reuse options (e.g. for agricultural application, as substrate in constructed wetlands, in cement production, for ceramic making, ….) have been identified
* Corresponding author. ). E-mail address: [email protected] (A. Abba
and investigated in the scientific literature (Ahmad et al., 2016). However, the chemical-physical-microbiological properties of the sludge can restrict its chances of reuse. Thus, minimization is anyway an interesting solution. Many WWTPs are equipped with a specific biological treatment stage aimed at stabilizing the sludge and reducing its production. Among these treatments, anaerobic digestion is widely used rez-Elvira et al., 2006; Tyagi and Lo, 2013), with the valuable (Pe opportunity to recover energy from biogas and, possibly, nutrients (especially phosphate) from the supernatant by means of chemical precipitation (Huang et al., 2015). However, during the years, many alternative solutions for sludge minimization have been studied. Interesting technologies placed in the water line seem to be: chemical oxidation, especially with ozone (Zhang et al., 2009), for the promotion of lysis-cryptic growth of biomass; oxic-settlinganaerobic process (OSA) (Demir and Filibeli, 2016); membrane bioreactor (MBR), especially in thermophilic conditions (Collivignarelli et al., 2014; 2015b). Moreover, the most important technologies applied in the sludge line are physical and chemical
http://dx.doi.org/10.1016/j.jclepro.2016.12.101 0959-6526/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
2
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
re et al., pre-treatments for improving sludge hydrolysis (Carre 2010) and thus anaerobic degradability. Among sludge final treatment alternatives, the following can be mentioned: wet oxidation (Menoni and Bertanza, 2016), and sludge-to-energy technologies, such as incineration (Panepinto et al., 2016) with the possibility of phosphorus recovery from ashes (Li et al., 2017), gasification and pyrolysis (Samolada and Zabaniotou, 2014); furthermore, a promising technology seems to be sewage sludge drying (in a specific solar greenhouse) with the use of heat from OFMSW (organic fraction from municipal solid waste) composting (Rada et al., 2014). In a previous experimentation (Collivignarelli et al., 2015c), the authors evaluated the treatability of thickened sewage sludge (coming from an industrial WWTP) by means of a thermophilic membrane bioreactor process, operating with intermittent aeration cycles. In this process, the reduction of volatile solids (VS) in the sludge is obtained through cellular lysis, promoted by the thermophilic condition and enhanced by the intermittent aeration; the alternation of aerobic and anaerobic conditions involves the activation of an uncoupled metabolism (Wei et al., 2003). The hydrolization of the fed sludge, leading to an increase of soluble COD, produces fresh organic substrate available for the thermophilic biomass. Moreover, the exothermic oxidation of this substrate releases an amount of heat which ensures autothermal conditions. In short, the results obtained by Collivignarelli et al. (2015c) demonstrate the possibility to obtain a strong sludge stabilization (the VS/ TS ratio was reduced from 70% down to 45%), the reduction of VS and COD being 64% and 57%, respectively, with hydraulic retention time (HRT) of 20 d. In the present paper, a municipal sewage sludge, instead of an industrial one, was submitted to the same treatment, in order to confirm the results obtained in the previous research. Moreover, experimental tests were finalized at the optimization of process conditions (in particular efforts were focused on the optimization of the aeration phases). Finally, the characteristics and possible recovery/disposal options of the liquid residue were investigated. The experimentation was divided into two periods. During the first one (duration 125 d) the same aeration pattern adopted in Collivignarelli et al. (2015c) was followed. In the second period (duration 300 d) the optimization of process conditions and the characterization of the permeate, in view of its final destination, were investigated.
modifications, respectively. 2.2. Experimental plan and management strategies of the pilot plant The pilot plant was fed with a thickened sludge taken from a municipal WWTP (10,000 people equivalent e PE e predenitrification configuration, no separate sludge stabilization). The average characteristics of the thickened sludge are summarized in Table 1. The experimentation was divided into two periods. During the first one (Period 1) the pilot plant was conducted adopting the same aeration strategy described in Collivignarelli et al. (2015c): the aeration/non aeration phases were alternated every 4 h. This period was subdivided into three sub-periods (A, B and C) during which the feeding flow rate (Qin) was varied in order to investigate the influence of the HRT. The extraction of excess sludge (Qpurge) was regulated with a flow rate equal to 0.1 Qin. The feeding was provided at the beginning of the first non aeration phase of the morning, while the permeate was extracted the subsequent day, just before the feeding. Between Periods 1 and 2, the pilot plant was modified as shown in Fig. 1. During Period 2, the aeration/non aeration cycles were varied so as to assess the effect on process performance. In particular, we focused on the following aspects: VS and COD removal yields, biodegradability and distribution of nitrogen forms in the effluent (permeate). Period 2 was divided into four subperiods (D, E, F and G), each one characterized by a specific aeration pattern: a progressive increase of the duration of the non aeration phase was adopted. Moreover, in this period, Qpurge was reduced with respect to Period 1. The extraction of mixed liquor occurred essentially during the cleaning of the pre-filters; losses were also due to sludge foaming. Finally, during sub-period E and subsequent periods, the feeding was provided at the beginning of each non aeration phase, the permeate being continuously extracted. For both the experimental periods, the operative temperature (55 C), the oxygen flow rate (0.5 m3 h1), the inflow oxygen pressure (2 bar), and the dissolved oxygen (DO) thresholds for the automatic supply were kept constant. The detailed operative conditions of each experimental period and sub-period are summarized in Table 2.
2. Materials and methods 2.3. Monitoring plan and analytical methods 2.1. Pilot plant description The experimentation was carried out in a thermophilic MBR pilot plant (volume ¼ 1 m3) that works with alternate aeration (with pure oxygen)/non aeration cycles. The plant configuration and the geometrical characteristics of the pilot plant are the same as those described in Collivignarelli et al. (2015c): in particular, the ultrafiltration section consists of a vessel with 7 ceramic membranes (23 channels each, cut-off 300 kDa and pore size 10 nm). During the experimentation (between Period 1 and Period 2), the pilot plant was modified. In effect, the oxygen supply system (4 porous plate placed on the floor of the tank) was replaced with a Venturi-type device for the direct injection of pure oxygen in the recirculation line of the mixed liquor. This modification was set up in order to reduce the oxygen consumption (the mass transfer efficiency was improved by adopting this system, which is also installed in the real plant described by Collivignarelli et al., 2015d). Finally, in order to limit the foam accumulation on the water surface, a mechanical device was installed: the factors affecting foam formation have been investigated by Collivignarelli et al. (2016). Fig. 1 shows the pilot plant scheme before and after adopting these
The influent sludge was sampled weekly, all along the experimentation, while the mixed liquor and the permeate (effluent) were drawn daily. On these samples, the concentrations of COD, total nitrogen (TN), total solids (TS) and VS, along with pH, were measured. In addition, the distribution of nitrogen forms (NHþ 4 -N, NO 3 -N, NO2 -N and organic nitrogen) was determined in the permeate. All analyses were performed according to the official methods (APAT and IRSA-CNR, 2003; IRSA-CNR, 1984; APHA, 2012). Respirometric tests (Oxygen Uptake Rate e OUR and Nitrate Utilization Rate e NUR), under mesophilic conditions (temperature between 20 and 25 C), were also carried out. The aim of these tests was to evaluate the capacity of a mesophilic biomass to degrade the carbonaceous residue of the permeate, in view of a subsequent treatment in the WWTP that originated the sludge, or in view of the possible reuse as external carbon source in a denitrification process. OUR tests were performed according to ISO 8192 (2007) standard. These tests were carried out under both endogenous (respiration due to cell maintenance) and exogenous (respiration due to degradation of a substrate) conditions. As for exogenous tests, they were always operated at the same food to mass (F/M) ratio: the
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
3
Fig. 1. Pilot plant scheme before and after the upgrade.
Table 1 Characteristics of the fed substrate (thickened sludge). Period 1
TS [g L1] VS [g L1] VS/TS [%] COD [mg L1] TN [mg L1] TP [mg L1] a b
Period 2
Average
CVa
nb
Average
CVa
nb
31 18 60 28,700 1750 680
0.07 0.07 0.05 0.10 0.20 0.16
17 17 17 17 17 17
33 19 57 30,000 1630 630
0.17 0.17 0.09 0.16 0.19 0.35
35 35 35 35 35 27
Coefficient of variation. Number of samples.
NUR batch tests were carried out according to Kristensen et al. (1992): the “permeate þ nitrate” substrate was dosed into a labscale temperature-controlled reactor, filled with mesophilic biomass. The depletion of NO 3 -N concentration was monitored over time for 4 h. 2.4. Data processing For each experimental sub-period, the removal yields of VS, COD and TN were determined by means of Eq. (1):
h½% ¼ ðXin Xreactor;f Xreactor;i Xout Þ=Xin *100
permeate was diluted with distilled water in such a way to obtain the typical COD concentration of the wastewater treated in the WWTP where the sludge was taken.
(1)
where, for the parameter X: Xin ¼ input load to the biological reactor all along the sub-period
Table 2 Detailed operative conditions of each experimental period and sub-period. Period
Sub-period
First e last day
Duration [d]
Qin [L d1]
Qpurge [% Qin]
Number per day and duration [h] of:
Period 1
A B C D E F G
0e61 62e89 90e125 134e212 228e283 284e369 370e430
62 28 36 79 56 86 61
50 75 50 50 100 100 100
10 10 10 1.0 1.8 1.7 4.6
3 3 3 3 3 3 3
Period 2
Aeration
4 4 4 4 4 3 2
¼ ¼ ¼ ¼ ¼ ¼ ¼
12 12 12 12 12 9 6
Non aeration 3 3 3 3 3 3 3
4 4 4 4 4 5 6
¼ ¼ ¼ ¼ ¼ ¼ ¼
12 12 12 12 12 15 18
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
4
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
Xreactor,i ¼ amount present in the reactor at the beginning of the sub-period Xreactor,f ¼ amount present in the reactor at the end of the subperiod Xout ¼ output load (due to permeate and excess sludge) all along the sub-period. 3. Results and discussion 3.1. Operative conditions The actual operative conditions of each experimental period and sub-period are reported in Table 3. The average concentration of DO in the reactor, during the aeration phases, was higher than 4 mg L1 during Period 1 while, from Period 2, after the replacement of the oxygen supply device, a slight decrease was observed (range: 2.63e3.50 mg L1). A very low DO concentration was also measured (<0.40 mg L1, except for an anomalous concentration of 1.5 mg L1 which was recorded during the sub-period A of Period 1) during the non aeration phases: injection of oxygen at regular time intervals was needed to prevent the occlusion of the oxygen supply membrane diffusers/nozzle (a too intensive injection of oxygen during sub-period A led to the anomalous value recorded). Autothermal conditions were maintained; the average mixed liquor temperature was slightly lower than 55 C. The organic loading rate (OLR) was around 1 kgCOD m3 d1 during Period 1; since sub-period E, the OLR was doubled by means of Qin increase. 3.2. VS, COD and TN concentration trend The VS, COD and TN concentration pattern in the feed, mixed liquor and permeate, respectively, is illustrated in Fig. 2. The variability of the influent stream characteristics is related to the way we adopted for feeding the plant: fresh sludge was withdrawn weekly by the WWTP. The VS and COD concentrations in the mixed liquor showed a higher variability in Period 2. This was due to the not continuous mode of extraction of excess sludge, in contrast with the procedure adopted in Period 1 (as previously specified). As regards the permeate, the concentrations of VS and COD were considerably lower than those measured in the mixed liquor. This resulted from the combined effect of the biological degradation of the organic substrate and the effective solid/liquid separation provided by the membrane unit. Finally, TN concentrations in the permeate and feed were similar. This shows that influent nitrogen was solubilized and low or negligible removal occurred. 3.3. Mass balance The removal yields of VS and COD, calculated for each sub-
period, are reported in Fig. 3: the thermophilic membrane reactor (TMR) exerted a very good removal efficiency, with a progressive increase, for VS, all along the experimentation. The VS removal yields (Fig. 3a) seem to be influenced by both HRT and aeration conditions, VS solubilisation and soluble matter oxidation occurring during the non aerated and aerated phases, respectively. During Period 1, the best performance (80% removal efficiency) was achieved in sub-period C, characterized by high HRT (27 d), equal duration of the aerated/non aerated phases, effective DO depletion during the non aerated phases (unlike what was recorded during sub-period A: see Table 3). The same VS removal yield (80%) was obtained in sub-period D, when the plant worked under similar conditions. The subsequent reduction of HRT in subperiod E, down to 12 d, resulted in a slight decrease of the VS removal yield, thus confirming the role of HRT, as already observed in sub-period B. Starting from sub-period F, the increase of the duration of non aerated phases led to an increase of VS removal up to values close to the maximum theoretically achievable, corresponding to the biodegradable fraction of VS (which can be estimated as reported in Eckenfelder, 1999). Operating conditions of sub-periods F and G were also optimal as concerns COD removal (Fig. 3b), which was higher than 85%. The Venturi-type oxygen supply system proved to be more efficient with respect to the one previously installed (membrane diffusers). A reduction of consumed oxygen per kg of COD removed was in effect observed: the average value for the membrane diffusers was equal to 7.6 kgO2 kgCOD1 removed, while, in case of direct oxygen injection into the recycling stream, oxygen consumption dropped down to 1.25 kgO2 kgCOD1 removed. Finally, the mass balance of TN (data not shown) has highlighted an average loss of about 30%. This may be due to gasification, presumably ammonia stripping, biological nitrification being inhibited under thermophilic conditions (Juteau, 2006; Abeynayaka and Visvanathan, 2011; Collivignarelli et al., 2015d) and denitrification being negligible for the practical absence of nitrite and nitrate in the inlet substrate. Ammonia stripping was in effect likely favoured by the high temperature (about 55 C) and pH (average 8.3, with peaks over 9) measured in the reactor.
3.4. Nitrogen forms in the permeate The liquid effluent of the TMR (permeate) was rich in TN (Fig. 2). A specific campaign aimed to determine the distribution of nitrogen forms in the permeate was carried out. In fact, in case NHþ 4 -N was the prevailing specie, N recovery as ammonia salt solution (e.g. sulphate or nitrate) by means of stripping and subsequent acidic absorption could be a valuable alternative. By the way, the relatively high permeate temperature would favour this process. In effect, high ammonium concentrations were expected, as the result of the degradation of protein substrate (Yenigün and Demirel,
Table 3 Operative conditions for each sub-period. Period
Sub-period
DO (Aerated/Non aerated phase) [mg L1]
Qin [L d1]
T [ C]
HRT [d]
OLR [kgCOD m3 d1]
F/M [kgCOD kgTS1 d1]
Period 1
A B C D E F G
4.38/1.53 4.35/0.36 4.12/0.31 3.28/0.34 2.71/0.31 3.50/0.34 2.63/0.13
36 54 38 32 85 77 71
54.5 54.2 54.6 54.1 54.2 54.2 54.9
28 19 27 31 12 13 14
0.81 1.13 0.75 1.00 2.15 1.98 2.09
0.011 0.016 0.011 0.014 0.033 0.045 0.050
Period 2
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
5
Fig. 2. Concentration trends of VS, COD and TN in the feed, mixed liquor and permeate (*: interruption for the upgrading of the oxygen supply system; **: interruption due to mechanical failure of the membrane filtration unit; ***: interruption due to the pre-feeding and recirculation pump failure; data sets used for the calculation of mass balances are included within the grey areas).
2013). The results of this investigation are summarized in Fig. 4: the percentage distribution of the different N forms with respect to TN was calculated. The predominant component was always represented by NHþ 4 -N and only a small part of TN consisted of organic nitrogen; NO 3 -N and NO2 -N presence was negligible. This observation supports the hypothesis that the measured loss of 30% of TN was due to the stripping of NHþ 4 -N. The percentage of ammonia in the permeate (i.e. the hydrolysis efficiency), as expected, is in agreement with the VS and COD removal efficiency (see Fig. 3). 3.5. Respirometric tests on permeate The permeate showed a COD concentration in the range
700e8000 mg L1, depending on the considered sub-period. Starting from sub-period D, OUR tests were carried out. The mesophilic biomass taken from the same WWTP of the treated thickened sludge was used. Tests were conducted under both endogenous (without substrate addition) and exogenous (with dosage of the permeate as external substrate) conditions. In this case, the permeate was previously diluted as described in the Materials and methods section. In this way the biodegradability of the permeate (and possible inhibitory effects) was investigated, in view of a following treatment of the permeate in a CAS WWTP or in view of a recovery as external carbon source in denitrification processes. In Fig. 5 the results of the respirometric tests are reported as ratio between the exogenous OUR and the endogenous OUR. Measured values were always greater than 1, thus indicating the
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
6
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
Fig. 3. Mass balance calculation: VS and COD removal yields in each experimental sub-period.
Fig. 4. Percentage distribution of nitrogen forms in the outgoing permeate (mean value).
absence of acute toxicity of the substrate. Moreover, the increase of the ratio all along the experimentation shows that operating conditions kept during sub-periods F and, in particular, G are of major interest also from this perspective. It is worth noting that a higher respiration rate was measured, with respect to the one obtainable with municipal wastewater dosed as a substrate at the same COD concentration (Fig. 5 - dashed line). Concerning the possible use of the TMR permeate as external carbon source in denitrification units, the results of NUR tests (Table 4) show that the observed specific denitrification rate is higher than the one obtained by dosing municipal sewage as a carbon source; moreover, the denitrification rate obtained with the TMR permeate is comparable with the one attainable by dosing readily biodegradable COD.
3.6. Economic and industrial potential The data collected during this experimentation show that a semi-industrial scale plant (mixed liquor volume of 1 m3) is able to treat 30 kg d1 (or even more as suggest by recent findings of ongoing research activities) of thickened sewage sludge (4% dry matter content, 40,000 mgCOD L1), or 6 kg d1 of dewatered sludge (20% dry matter content and 200,000 mgCOD L1). Assuming to provide a full scale industrial plant of 1000 m3 (according to the size of a real plant for the treatment of industrial aqueous waste, which was studied by the authors e Collivignarelli et al., 2015d), this could treat about 11,000 t y1 of thickened sewage sludge (4% dry matter content and 40,000 mgCOD L1), or 2200 t y1 of dewatered sludge (20% dry matter content and
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
7
Fig. 5. Results of the OUR tests performed during sub-periods from D to G (**: interruption due to mechanical failure of the membrane filtration unit; ***: interruption due to the pre-feeding and recirculation pump failure; data sets used for the calculation of mass balances are included within the grey areas): results are expressed as the ratio between exogenous and endogenous OUR.
Table 4 Specific denitrification rate of activated sludge obtained by dosing different substrate as carbon source. Carbon source
Sludge denitrification rate 1 1 [mgNO h ] 3 eN gVSS
TMR permeate Municipal sewage Readily biodegradable COD Endogenous respiration
1.9 ± 0.3 0.6e1.0a 1.0e3.0a 0.2e0.6a
a
Source: Foladori et al., 2010.
200,000 mgCOD L1). The running costs of this plant, based on the experience acquired throughout the 11 years operation of a full scale facility (Collivignarelli et al., 2015d), can be estimated as follows: (i) power consumption, around 10 kWh m3 of MBR permeate (essentially for mixing and pumping); (ii) pure oxygen supply, around 1.2 kgO2 kgCOD1 removed; (iii) costs of membrane maintenance and depreciation between 0.5 and 1.5 V m3 of MBR permeate (this value is strongly influenced by the type of biomass, the presence of polymers, precipitates …). Based on these figures and on the sludge disposal cost (strictly depending on local conditions) the evaluation of economic sustainability can be performed, case by case. 4. Conclusions The MBR thermophilic process with intermittent aeration proved to be a promising option for the reduction of sewage sludge in municipal WWTPs. The main results of the experimentation can be summarized as follows: HRT and aeration conditions are crucial process parameters: with HRT <15 d and 2 h aeration - 6 h non aeration cycles the removal efficiency of VS and COD was appreciably greater than
80%; these performances are even better than those obtained in a previous research, where an industrial, instead of a municipal, sludge was treated (Collivignarelli et al., 2015c). The direct injection of pure oxygen into the sludge recirculation line allows to limit oxygen consumption to around 1.25 kgO2 kgCOD1 removed. The liquid effluent of the process (permeate) shows a better biodegradability under mesophilic conditions than municipal wastewater, thus allowing its final treatment in a conventional municipal WWTP; moreover, the permeate can be valorised (after ammonia removal) as carbon source in a denitrification unit. The TMR process converts the organic nitrogen present in the fed sludge into ammonia: in one of the experiments the conversion was complete. The ammonia can be extracted both from the reactor itself (about 30% being stripped out from the mixed liquor at the operative temperature and pH conditions) and by means of the permeate stripping (70%) and subsequently recovered as ammonium sulphate. Acknowledgements The authors wish to thank Idroclean S.p.A., ASMortara S.p.A., Aquagest S.r.l. and ASMia S.r.l. for their technical and financial support to the research. References Abeynayaka, A., Visvanathan, C., 2011. Mesophilic and thermophilic aerobic batch biodegradation, utilization of carbon and nitrogen sources in high-strength wastewater. Bioresour. Technol. 102 (3), 2358e2366. Ahmad, T., Ahmad, S., Alam, M., 2016. Sustainable management of water treatment sludge through 3‘R’ concept. J. Clean. Prod. 124, 1e13. American Public Health Association (APHA), 2012. Standard Methods for the Examination of Water and Wastewater, twenty-second ed. Washington DC, USA. APAT, IRSA-CNR, 2003. Analytical Methods for the Waters (In Italian). Materials and
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
8
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
Guidelines 29/2003. Campos, J.l., Otero, L., Franco, A., Mosquera-Corral, A., Roca, E., 2009. Ozonation strategies to reduce sludge production of seafood industry WWTP. Bioresour. Technol. 100, 1069e1073. re, H., Dumas, C., Battimelli, A., Batstone, D.J., Delgene s, J.P., Steyer, J.P., Carre Ferrer, I., 2010. Pretreatment methods to improve sludge anaerobic degradability: a review. J. Hazard. Mater. 183, 1e15. , A., Bertanza, G., 2014. Treatment of high strength pharCollivignarelli, M.C., Abba maceutical wastewater in a thermophilic aerobic membrane reactor (TAMR). Water Res. 63, 190e198. , A., Padovani, S., Frascarolo, M., Sciunnach, D., Turconi, M., Collivignarelli, M.C., Abba Orlando, M., 2015a. Recovery of sewage sludge on agricultural land in Lombardy: current issues and regulatory scenarios. Environ. Eng. Manag. J. 14 (7), 1477e1486. Collivignarelli, M.C., Abb a, A., Bertanza, G., 2015b. Why use a Thermophilic Aerobic Membrane Reactor for the treatment of industrial wastewater/liquid waste? Environ. Technol. 36 (16), 2115e2124. Collivignarelli, M.C., Castagnola, F., Sordi, M., Bertanza, G., 2015c. Treatment of sewage sludge in a thermophilic membrane reactor (TMR) with alternate aeration cycles. J. Environ. Manag. 162, 132e138. Collivignarelli, M.C., Bertanza, G., Sordi, M., Pedrazzani, R., 2015d. High-strength wastewater treatment in a pure oxygen thermophilic process: 11-year operation and monitoring of different plant configurations. Water Sci. Technol. 71 (4), 588e596. Collivignarelli, M.C., Castagnola, F., Sordi, M., Bertanza, G., 2016. Sewage sludge treatment in a thermophilic membrane reactor (TMR): factors affecting foam formations. Environ. Sci. Pollut. Res. http://dx.doi.org/10.1007/s11356-0167983-4. Available online 4 November 2016, (in press). Demir, O., Filibeli, A., 2016. The investigation of the sludge reduction efficiency and mechanisms in oxic-settling-anaerobic (OSA) process. Water Sci. Technol. 73 (10), 2331e2323. Eckenfelder Jr., W.W., 1999. Industrial Water Pollution Control, third ed. McGraw Hill International. ISBN: 978e0070393646. Foladori, P., Andreottola, G., Ziglio, G., 2010. Sludge Reduction Technologies in Wastewater Treatment Plants. IWA Publishing, London, UK. ISBN: 9781843392781. Huang, H., Liu, J., Ding, L., 2015. Recovery of phosphate and ammonia nitrogen from the anaerobic digestion supernatant of activated sludge by chemical precipitation. J. Clean. Prod. 102, 437e446.
International Organization for Standardization (ISO), 2007. ISO 8192:2007: Water Quality e Test of Inhibition of Oxygen Consumption by Activated Sludge for Carbonaceous and Ammonium Oxidation. ISO, Geneva. IRSA-CNR, 1984. Analytical Methods for Sludge. Technological Parameters (In Italian). Polygraph Institute of National Mint, Rome. IRSA-CNR, No. 64, ed. Juteau, P., 2006. Review of the use of aerobic thermophilic bioprocesses for the treatment of swine waste. Livest. Sci. 102 (3), 187e196. Kelessidis, A., Stasinakis, A.S., 2012. Comparative study of the methods used for treatment and final disposal of sewage sludge in European countries. Waste Manag. 32, 1186e1195. Kristensen, G., Jørgensen, P.E., Henze, M., 1992. Characterization of functional micro-organism groups and substrate in activated sludge and wastewater by AUR, NUR and OUR. Water Sci. Technol. 25 (6), 43e57. Li, R., Teng, W., Li, Y., Wang, W., Cui, R., Yang, T., 2017. Potential recovery of phosphorus during the fluidized bed incineration of sewage sludge. J. Clean. Prod. 140, 964e970. http://dx.doi.org/10.1016/j.jclepro.2016.06.177. Menoni, L., Bertanza, G., 2016. Wet Oxidation of sewage sludge: a mathematical model for estimating the performance based on the VSS/TSS ratio. Chem. Eng. J. 306, 685e692. Panepinto, D., Fiore, S., Genon, G., Acri, M., 2016. Thermal valorization of sewer sludge: perspectives for large wastewater treatment plants. J. Clean. Prod. 137, 1323e1329. rez-Elvira, S.I., Nieto Diez, P., Fdz-Polanco, F., 2006. Sludge minimisation techPe nologies. Rev. Environ. Sci. Bio/Technol. 5 (4), 375e398. Rada, E.C., Ragazzi, M., Villotti, S., Torretta, V., 2014. Sewage sludge drying by energy recovery from OFMSW composting: preliminary feasibility evaluation. Waste Manag. 34 (5), 895e866. Samolada, M.C., Zabaniotou, A.A., 2014. Comparative assessment of municipal sewage sludge incineration, gasification and pyrolysis for a sustainable sludgeto-energy management in Greece. Waste Manag. 34 (2), 411e420. Tyagi, V.K., Lo, S.-L., 2013. Sludge: a waste or renewable source for energy and resources recovery? Renew. Sustain Energy Rev. 25, 708e728. Wei, Y., Van Houten, R.T., Borger, A.R., Eikelboom, D.H., Fan, Y., 2003. Minimization of excess sludge production for biological wastewater treatment. Water Res. 37, 4453e4467. Yenigün, O., Demirel, B., 2013. Ammonia inhibition in anaerobic digestion: a review. Process Biochem. 48, 901e911. Zhang, G., Yang, J., Liu, H., Zhang, J., 2009. Sludge ozonation: disintegration, supernatant changes and mechanisms. Bioresour. Technol. 100, 1505e1509.
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration a, *, Federico Castagnola a, Maria Cristina Collivignarelli a, Alessandro Abba Giorgio Bertanza b a b
Department of Civil Engineering and Architecture, University of Pavia, Via Ferrata 1, 27100, Pavia, Italy Department of Civil, Environmental, Architectural Engineering and Mathematics, University of Brescia, Via Branze 43, 25123, Brescia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 September 2016 Received in revised form 19 December 2016 Accepted 19 December 2016 Available online xxx
The increase of sewage sludge production together with the high treatment and disposal costs in the last years has pushed to study different solutions aimed at sludge minimization. In this paper, the thermophilic membrane technology was evaluated as an alternative for municipal sewage sludge reduction. The experimentation (carried out by means of a pilot scale plant, 1 m3 volume) was divided into two steps: the first one was aimed at confirming the results obtained in a previous research focused on industrial sludge treatment; the second step was devoted to define the best process conditions (in particular the optimization of the aeration phases) and to the chemical permeate characterization. The results of the experiments highlight that the hydraulic retention time (HRT) and aeration conditions play a crucial role on the overall process performance. The volatile suspended solids removal efficiency was greater than 80% under the following conditions: HRT even lower than 15 d; 2 h of aeration - 6 h of non aeration cycles; and organic loading rate of 2.0 kgCOD m3 d1. The permeate showed a good biodegradability under mesophilic conditions thus being treatable by means of conventional biological processes. Moreover, ammonia (the permeate presenting high concentrations) could be recovered as a fertilizer (stripping and subsequent washing of the exhausted gas is an established technique). Finally, the ammonia-free permeate can be valorised as a carbon source in denitrification processes. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Intermittent aeration Membrane biological reactor Sludge minimization Thermophilic biological treatment Ultrafiltration
1. Introduction The sewage sludge produced by conventional activated sludge (CAS) wastewater treatment plants (WWTPs) is characterized by a high organic content and the presence of heavy metals, organic contaminants and/or pathogens, depending on the nature of the treated wastewater and on the process conditions. In Europe, the yearly production of sewage sludge is estimated at 10.13 million tons of dry matter (Collivignarelli et al., 2015a) and it is expected to reach the amount of 13 million tons (Kelessidis and Stasinakis, 2012) by 2020. Moreover, the sludge treatment and disposal costs can be equal to 50e60% of the total operating costs of a WWTP (Campos et al., 2009). Nowadays, several sewage sludge reuse options (e.g. for agricultural application, as substrate in constructed wetlands, in cement production, for ceramic making, ….) have been identified
* Corresponding author. ). E-mail address: [email protected] (A. Abba
and investigated in the scientific literature (Ahmad et al., 2016). However, the chemical-physical-microbiological properties of the sludge can restrict its chances of reuse. Thus, minimization is anyway an interesting solution. Many WWTPs are equipped with a specific biological treatment stage aimed at stabilizing the sludge and reducing its production. Among these treatments, anaerobic digestion is widely used rez-Elvira et al., 2006; Tyagi and Lo, 2013), with the valuable (Pe opportunity to recover energy from biogas and, possibly, nutrients (especially phosphate) from the supernatant by means of chemical precipitation (Huang et al., 2015). However, during the years, many alternative solutions for sludge minimization have been studied. Interesting technologies placed in the water line seem to be: chemical oxidation, especially with ozone (Zhang et al., 2009), for the promotion of lysis-cryptic growth of biomass; oxic-settlinganaerobic process (OSA) (Demir and Filibeli, 2016); membrane bioreactor (MBR), especially in thermophilic conditions (Collivignarelli et al., 2014; 2015b). Moreover, the most important technologies applied in the sludge line are physical and chemical
http://dx.doi.org/10.1016/j.jclepro.2016.12.101 0959-6526/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
2
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
re et al., pre-treatments for improving sludge hydrolysis (Carre 2010) and thus anaerobic degradability. Among sludge final treatment alternatives, the following can be mentioned: wet oxidation (Menoni and Bertanza, 2016), and sludge-to-energy technologies, such as incineration (Panepinto et al., 2016) with the possibility of phosphorus recovery from ashes (Li et al., 2017), gasification and pyrolysis (Samolada and Zabaniotou, 2014); furthermore, a promising technology seems to be sewage sludge drying (in a specific solar greenhouse) with the use of heat from OFMSW (organic fraction from municipal solid waste) composting (Rada et al., 2014). In a previous experimentation (Collivignarelli et al., 2015c), the authors evaluated the treatability of thickened sewage sludge (coming from an industrial WWTP) by means of a thermophilic membrane bioreactor process, operating with intermittent aeration cycles. In this process, the reduction of volatile solids (VS) in the sludge is obtained through cellular lysis, promoted by the thermophilic condition and enhanced by the intermittent aeration; the alternation of aerobic and anaerobic conditions involves the activation of an uncoupled metabolism (Wei et al., 2003). The hydrolization of the fed sludge, leading to an increase of soluble COD, produces fresh organic substrate available for the thermophilic biomass. Moreover, the exothermic oxidation of this substrate releases an amount of heat which ensures autothermal conditions. In short, the results obtained by Collivignarelli et al. (2015c) demonstrate the possibility to obtain a strong sludge stabilization (the VS/ TS ratio was reduced from 70% down to 45%), the reduction of VS and COD being 64% and 57%, respectively, with hydraulic retention time (HRT) of 20 d. In the present paper, a municipal sewage sludge, instead of an industrial one, was submitted to the same treatment, in order to confirm the results obtained in the previous research. Moreover, experimental tests were finalized at the optimization of process conditions (in particular efforts were focused on the optimization of the aeration phases). Finally, the characteristics and possible recovery/disposal options of the liquid residue were investigated. The experimentation was divided into two periods. During the first one (duration 125 d) the same aeration pattern adopted in Collivignarelli et al. (2015c) was followed. In the second period (duration 300 d) the optimization of process conditions and the characterization of the permeate, in view of its final destination, were investigated.
modifications, respectively. 2.2. Experimental plan and management strategies of the pilot plant The pilot plant was fed with a thickened sludge taken from a municipal WWTP (10,000 people equivalent e PE e predenitrification configuration, no separate sludge stabilization). The average characteristics of the thickened sludge are summarized in Table 1. The experimentation was divided into two periods. During the first one (Period 1) the pilot plant was conducted adopting the same aeration strategy described in Collivignarelli et al. (2015c): the aeration/non aeration phases were alternated every 4 h. This period was subdivided into three sub-periods (A, B and C) during which the feeding flow rate (Qin) was varied in order to investigate the influence of the HRT. The extraction of excess sludge (Qpurge) was regulated with a flow rate equal to 0.1 Qin. The feeding was provided at the beginning of the first non aeration phase of the morning, while the permeate was extracted the subsequent day, just before the feeding. Between Periods 1 and 2, the pilot plant was modified as shown in Fig. 1. During Period 2, the aeration/non aeration cycles were varied so as to assess the effect on process performance. In particular, we focused on the following aspects: VS and COD removal yields, biodegradability and distribution of nitrogen forms in the effluent (permeate). Period 2 was divided into four subperiods (D, E, F and G), each one characterized by a specific aeration pattern: a progressive increase of the duration of the non aeration phase was adopted. Moreover, in this period, Qpurge was reduced with respect to Period 1. The extraction of mixed liquor occurred essentially during the cleaning of the pre-filters; losses were also due to sludge foaming. Finally, during sub-period E and subsequent periods, the feeding was provided at the beginning of each non aeration phase, the permeate being continuously extracted. For both the experimental periods, the operative temperature (55 C), the oxygen flow rate (0.5 m3 h1), the inflow oxygen pressure (2 bar), and the dissolved oxygen (DO) thresholds for the automatic supply were kept constant. The detailed operative conditions of each experimental period and sub-period are summarized in Table 2.
2. Materials and methods 2.3. Monitoring plan and analytical methods 2.1. Pilot plant description The experimentation was carried out in a thermophilic MBR pilot plant (volume ¼ 1 m3) that works with alternate aeration (with pure oxygen)/non aeration cycles. The plant configuration and the geometrical characteristics of the pilot plant are the same as those described in Collivignarelli et al. (2015c): in particular, the ultrafiltration section consists of a vessel with 7 ceramic membranes (23 channels each, cut-off 300 kDa and pore size 10 nm). During the experimentation (between Period 1 and Period 2), the pilot plant was modified. In effect, the oxygen supply system (4 porous plate placed on the floor of the tank) was replaced with a Venturi-type device for the direct injection of pure oxygen in the recirculation line of the mixed liquor. This modification was set up in order to reduce the oxygen consumption (the mass transfer efficiency was improved by adopting this system, which is also installed in the real plant described by Collivignarelli et al., 2015d). Finally, in order to limit the foam accumulation on the water surface, a mechanical device was installed: the factors affecting foam formation have been investigated by Collivignarelli et al. (2016). Fig. 1 shows the pilot plant scheme before and after adopting these
The influent sludge was sampled weekly, all along the experimentation, while the mixed liquor and the permeate (effluent) were drawn daily. On these samples, the concentrations of COD, total nitrogen (TN), total solids (TS) and VS, along with pH, were measured. In addition, the distribution of nitrogen forms (NHþ 4 -N, NO 3 -N, NO2 -N and organic nitrogen) was determined in the permeate. All analyses were performed according to the official methods (APAT and IRSA-CNR, 2003; IRSA-CNR, 1984; APHA, 2012). Respirometric tests (Oxygen Uptake Rate e OUR and Nitrate Utilization Rate e NUR), under mesophilic conditions (temperature between 20 and 25 C), were also carried out. The aim of these tests was to evaluate the capacity of a mesophilic biomass to degrade the carbonaceous residue of the permeate, in view of a subsequent treatment in the WWTP that originated the sludge, or in view of the possible reuse as external carbon source in a denitrification process. OUR tests were performed according to ISO 8192 (2007) standard. These tests were carried out under both endogenous (respiration due to cell maintenance) and exogenous (respiration due to degradation of a substrate) conditions. As for exogenous tests, they were always operated at the same food to mass (F/M) ratio: the
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
3
Fig. 1. Pilot plant scheme before and after the upgrade.
Table 1 Characteristics of the fed substrate (thickened sludge). Period 1
TS [g L1] VS [g L1] VS/TS [%] COD [mg L1] TN [mg L1] TP [mg L1] a b
Period 2
Average
CVa
nb
Average
CVa
nb
31 18 60 28,700 1750 680
0.07 0.07 0.05 0.10 0.20 0.16
17 17 17 17 17 17
33 19 57 30,000 1630 630
0.17 0.17 0.09 0.16 0.19 0.35
35 35 35 35 35 27
Coefficient of variation. Number of samples.
NUR batch tests were carried out according to Kristensen et al. (1992): the “permeate þ nitrate” substrate was dosed into a labscale temperature-controlled reactor, filled with mesophilic biomass. The depletion of NO 3 -N concentration was monitored over time for 4 h. 2.4. Data processing For each experimental sub-period, the removal yields of VS, COD and TN were determined by means of Eq. (1):
h½% ¼ ðXin Xreactor;f Xreactor;i Xout Þ=Xin *100
permeate was diluted with distilled water in such a way to obtain the typical COD concentration of the wastewater treated in the WWTP where the sludge was taken.
(1)
where, for the parameter X: Xin ¼ input load to the biological reactor all along the sub-period
Table 2 Detailed operative conditions of each experimental period and sub-period. Period
Sub-period
First e last day
Duration [d]
Qin [L d1]
Qpurge [% Qin]
Number per day and duration [h] of:
Period 1
A B C D E F G
0e61 62e89 90e125 134e212 228e283 284e369 370e430
62 28 36 79 56 86 61
50 75 50 50 100 100 100
10 10 10 1.0 1.8 1.7 4.6
3 3 3 3 3 3 3
Period 2
Aeration
4 4 4 4 4 3 2
¼ ¼ ¼ ¼ ¼ ¼ ¼
12 12 12 12 12 9 6
Non aeration 3 3 3 3 3 3 3
4 4 4 4 4 5 6
¼ ¼ ¼ ¼ ¼ ¼ ¼
12 12 12 12 12 15 18
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
4
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
Xreactor,i ¼ amount present in the reactor at the beginning of the sub-period Xreactor,f ¼ amount present in the reactor at the end of the subperiod Xout ¼ output load (due to permeate and excess sludge) all along the sub-period. 3. Results and discussion 3.1. Operative conditions The actual operative conditions of each experimental period and sub-period are reported in Table 3. The average concentration of DO in the reactor, during the aeration phases, was higher than 4 mg L1 during Period 1 while, from Period 2, after the replacement of the oxygen supply device, a slight decrease was observed (range: 2.63e3.50 mg L1). A very low DO concentration was also measured (<0.40 mg L1, except for an anomalous concentration of 1.5 mg L1 which was recorded during the sub-period A of Period 1) during the non aeration phases: injection of oxygen at regular time intervals was needed to prevent the occlusion of the oxygen supply membrane diffusers/nozzle (a too intensive injection of oxygen during sub-period A led to the anomalous value recorded). Autothermal conditions were maintained; the average mixed liquor temperature was slightly lower than 55 C. The organic loading rate (OLR) was around 1 kgCOD m3 d1 during Period 1; since sub-period E, the OLR was doubled by means of Qin increase. 3.2. VS, COD and TN concentration trend The VS, COD and TN concentration pattern in the feed, mixed liquor and permeate, respectively, is illustrated in Fig. 2. The variability of the influent stream characteristics is related to the way we adopted for feeding the plant: fresh sludge was withdrawn weekly by the WWTP. The VS and COD concentrations in the mixed liquor showed a higher variability in Period 2. This was due to the not continuous mode of extraction of excess sludge, in contrast with the procedure adopted in Period 1 (as previously specified). As regards the permeate, the concentrations of VS and COD were considerably lower than those measured in the mixed liquor. This resulted from the combined effect of the biological degradation of the organic substrate and the effective solid/liquid separation provided by the membrane unit. Finally, TN concentrations in the permeate and feed were similar. This shows that influent nitrogen was solubilized and low or negligible removal occurred. 3.3. Mass balance The removal yields of VS and COD, calculated for each sub-
period, are reported in Fig. 3: the thermophilic membrane reactor (TMR) exerted a very good removal efficiency, with a progressive increase, for VS, all along the experimentation. The VS removal yields (Fig. 3a) seem to be influenced by both HRT and aeration conditions, VS solubilisation and soluble matter oxidation occurring during the non aerated and aerated phases, respectively. During Period 1, the best performance (80% removal efficiency) was achieved in sub-period C, characterized by high HRT (27 d), equal duration of the aerated/non aerated phases, effective DO depletion during the non aerated phases (unlike what was recorded during sub-period A: see Table 3). The same VS removal yield (80%) was obtained in sub-period D, when the plant worked under similar conditions. The subsequent reduction of HRT in subperiod E, down to 12 d, resulted in a slight decrease of the VS removal yield, thus confirming the role of HRT, as already observed in sub-period B. Starting from sub-period F, the increase of the duration of non aerated phases led to an increase of VS removal up to values close to the maximum theoretically achievable, corresponding to the biodegradable fraction of VS (which can be estimated as reported in Eckenfelder, 1999). Operating conditions of sub-periods F and G were also optimal as concerns COD removal (Fig. 3b), which was higher than 85%. The Venturi-type oxygen supply system proved to be more efficient with respect to the one previously installed (membrane diffusers). A reduction of consumed oxygen per kg of COD removed was in effect observed: the average value for the membrane diffusers was equal to 7.6 kgO2 kgCOD1 removed, while, in case of direct oxygen injection into the recycling stream, oxygen consumption dropped down to 1.25 kgO2 kgCOD1 removed. Finally, the mass balance of TN (data not shown) has highlighted an average loss of about 30%. This may be due to gasification, presumably ammonia stripping, biological nitrification being inhibited under thermophilic conditions (Juteau, 2006; Abeynayaka and Visvanathan, 2011; Collivignarelli et al., 2015d) and denitrification being negligible for the practical absence of nitrite and nitrate in the inlet substrate. Ammonia stripping was in effect likely favoured by the high temperature (about 55 C) and pH (average 8.3, with peaks over 9) measured in the reactor.
3.4. Nitrogen forms in the permeate The liquid effluent of the TMR (permeate) was rich in TN (Fig. 2). A specific campaign aimed to determine the distribution of nitrogen forms in the permeate was carried out. In fact, in case NHþ 4 -N was the prevailing specie, N recovery as ammonia salt solution (e.g. sulphate or nitrate) by means of stripping and subsequent acidic absorption could be a valuable alternative. By the way, the relatively high permeate temperature would favour this process. In effect, high ammonium concentrations were expected, as the result of the degradation of protein substrate (Yenigün and Demirel,
Table 3 Operative conditions for each sub-period. Period
Sub-period
DO (Aerated/Non aerated phase) [mg L1]
Qin [L d1]
T [ C]
HRT [d]
OLR [kgCOD m3 d1]
F/M [kgCOD kgTS1 d1]
Period 1
A B C D E F G
4.38/1.53 4.35/0.36 4.12/0.31 3.28/0.34 2.71/0.31 3.50/0.34 2.63/0.13
36 54 38 32 85 77 71
54.5 54.2 54.6 54.1 54.2 54.2 54.9
28 19 27 31 12 13 14
0.81 1.13 0.75 1.00 2.15 1.98 2.09
0.011 0.016 0.011 0.014 0.033 0.045 0.050
Period 2
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
5
Fig. 2. Concentration trends of VS, COD and TN in the feed, mixed liquor and permeate (*: interruption for the upgrading of the oxygen supply system; **: interruption due to mechanical failure of the membrane filtration unit; ***: interruption due to the pre-feeding and recirculation pump failure; data sets used for the calculation of mass balances are included within the grey areas).
2013). The results of this investigation are summarized in Fig. 4: the percentage distribution of the different N forms with respect to TN was calculated. The predominant component was always represented by NHþ 4 -N and only a small part of TN consisted of organic nitrogen; NO 3 -N and NO2 -N presence was negligible. This observation supports the hypothesis that the measured loss of 30% of TN was due to the stripping of NHþ 4 -N. The percentage of ammonia in the permeate (i.e. the hydrolysis efficiency), as expected, is in agreement with the VS and COD removal efficiency (see Fig. 3). 3.5. Respirometric tests on permeate The permeate showed a COD concentration in the range
700e8000 mg L1, depending on the considered sub-period. Starting from sub-period D, OUR tests were carried out. The mesophilic biomass taken from the same WWTP of the treated thickened sludge was used. Tests were conducted under both endogenous (without substrate addition) and exogenous (with dosage of the permeate as external substrate) conditions. In this case, the permeate was previously diluted as described in the Materials and methods section. In this way the biodegradability of the permeate (and possible inhibitory effects) was investigated, in view of a following treatment of the permeate in a CAS WWTP or in view of a recovery as external carbon source in denitrification processes. In Fig. 5 the results of the respirometric tests are reported as ratio between the exogenous OUR and the endogenous OUR. Measured values were always greater than 1, thus indicating the
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
6
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
Fig. 3. Mass balance calculation: VS and COD removal yields in each experimental sub-period.
Fig. 4. Percentage distribution of nitrogen forms in the outgoing permeate (mean value).
absence of acute toxicity of the substrate. Moreover, the increase of the ratio all along the experimentation shows that operating conditions kept during sub-periods F and, in particular, G are of major interest also from this perspective. It is worth noting that a higher respiration rate was measured, with respect to the one obtainable with municipal wastewater dosed as a substrate at the same COD concentration (Fig. 5 - dashed line). Concerning the possible use of the TMR permeate as external carbon source in denitrification units, the results of NUR tests (Table 4) show that the observed specific denitrification rate is higher than the one obtained by dosing municipal sewage as a carbon source; moreover, the denitrification rate obtained with the TMR permeate is comparable with the one attainable by dosing readily biodegradable COD.
3.6. Economic and industrial potential The data collected during this experimentation show that a semi-industrial scale plant (mixed liquor volume of 1 m3) is able to treat 30 kg d1 (or even more as suggest by recent findings of ongoing research activities) of thickened sewage sludge (4% dry matter content, 40,000 mgCOD L1), or 6 kg d1 of dewatered sludge (20% dry matter content and 200,000 mgCOD L1). Assuming to provide a full scale industrial plant of 1000 m3 (according to the size of a real plant for the treatment of industrial aqueous waste, which was studied by the authors e Collivignarelli et al., 2015d), this could treat about 11,000 t y1 of thickened sewage sludge (4% dry matter content and 40,000 mgCOD L1), or 2200 t y1 of dewatered sludge (20% dry matter content and
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
7
Fig. 5. Results of the OUR tests performed during sub-periods from D to G (**: interruption due to mechanical failure of the membrane filtration unit; ***: interruption due to the pre-feeding and recirculation pump failure; data sets used for the calculation of mass balances are included within the grey areas): results are expressed as the ratio between exogenous and endogenous OUR.
Table 4 Specific denitrification rate of activated sludge obtained by dosing different substrate as carbon source. Carbon source
Sludge denitrification rate 1 1 [mgNO h ] 3 eN gVSS
TMR permeate Municipal sewage Readily biodegradable COD Endogenous respiration
1.9 ± 0.3 0.6e1.0a 1.0e3.0a 0.2e0.6a
a
Source: Foladori et al., 2010.
200,000 mgCOD L1). The running costs of this plant, based on the experience acquired throughout the 11 years operation of a full scale facility (Collivignarelli et al., 2015d), can be estimated as follows: (i) power consumption, around 10 kWh m3 of MBR permeate (essentially for mixing and pumping); (ii) pure oxygen supply, around 1.2 kgO2 kgCOD1 removed; (iii) costs of membrane maintenance and depreciation between 0.5 and 1.5 V m3 of MBR permeate (this value is strongly influenced by the type of biomass, the presence of polymers, precipitates …). Based on these figures and on the sludge disposal cost (strictly depending on local conditions) the evaluation of economic sustainability can be performed, case by case. 4. Conclusions The MBR thermophilic process with intermittent aeration proved to be a promising option for the reduction of sewage sludge in municipal WWTPs. The main results of the experimentation can be summarized as follows: HRT and aeration conditions are crucial process parameters: with HRT <15 d and 2 h aeration - 6 h non aeration cycles the removal efficiency of VS and COD was appreciably greater than
80%; these performances are even better than those obtained in a previous research, where an industrial, instead of a municipal, sludge was treated (Collivignarelli et al., 2015c). The direct injection of pure oxygen into the sludge recirculation line allows to limit oxygen consumption to around 1.25 kgO2 kgCOD1 removed. The liquid effluent of the process (permeate) shows a better biodegradability under mesophilic conditions than municipal wastewater, thus allowing its final treatment in a conventional municipal WWTP; moreover, the permeate can be valorised (after ammonia removal) as carbon source in a denitrification unit. The TMR process converts the organic nitrogen present in the fed sludge into ammonia: in one of the experiments the conversion was complete. The ammonia can be extracted both from the reactor itself (about 30% being stripped out from the mixed liquor at the operative temperature and pH conditions) and by means of the permeate stripping (70%) and subsequently recovered as ammonium sulphate. Acknowledgements The authors wish to thank Idroclean S.p.A., ASMortara S.p.A., Aquagest S.r.l. and ASMia S.r.l. for their technical and financial support to the research. References Abeynayaka, A., Visvanathan, C., 2011. Mesophilic and thermophilic aerobic batch biodegradation, utilization of carbon and nitrogen sources in high-strength wastewater. Bioresour. Technol. 102 (3), 2358e2366. Ahmad, T., Ahmad, S., Alam, M., 2016. Sustainable management of water treatment sludge through 3‘R’ concept. J. Clean. Prod. 124, 1e13. American Public Health Association (APHA), 2012. Standard Methods for the Examination of Water and Wastewater, twenty-second ed. Washington DC, USA. APAT, IRSA-CNR, 2003. Analytical Methods for the Waters (In Italian). Materials and
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101
8
M.C. Collivignarelli et al. / Journal of Cleaner Production xxx (2016) 1e8
Guidelines 29/2003. Campos, J.l., Otero, L., Franco, A., Mosquera-Corral, A., Roca, E., 2009. Ozonation strategies to reduce sludge production of seafood industry WWTP. Bioresour. Technol. 100, 1069e1073. re, H., Dumas, C., Battimelli, A., Batstone, D.J., Delgene s, J.P., Steyer, J.P., Carre Ferrer, I., 2010. Pretreatment methods to improve sludge anaerobic degradability: a review. J. Hazard. Mater. 183, 1e15. , A., Bertanza, G., 2014. Treatment of high strength pharCollivignarelli, M.C., Abba maceutical wastewater in a thermophilic aerobic membrane reactor (TAMR). Water Res. 63, 190e198. , A., Padovani, S., Frascarolo, M., Sciunnach, D., Turconi, M., Collivignarelli, M.C., Abba Orlando, M., 2015a. Recovery of sewage sludge on agricultural land in Lombardy: current issues and regulatory scenarios. Environ. Eng. Manag. J. 14 (7), 1477e1486. Collivignarelli, M.C., Abb a, A., Bertanza, G., 2015b. Why use a Thermophilic Aerobic Membrane Reactor for the treatment of industrial wastewater/liquid waste? Environ. Technol. 36 (16), 2115e2124. Collivignarelli, M.C., Castagnola, F., Sordi, M., Bertanza, G., 2015c. Treatment of sewage sludge in a thermophilic membrane reactor (TMR) with alternate aeration cycles. J. Environ. Manag. 162, 132e138. Collivignarelli, M.C., Bertanza, G., Sordi, M., Pedrazzani, R., 2015d. High-strength wastewater treatment in a pure oxygen thermophilic process: 11-year operation and monitoring of different plant configurations. Water Sci. Technol. 71 (4), 588e596. Collivignarelli, M.C., Castagnola, F., Sordi, M., Bertanza, G., 2016. Sewage sludge treatment in a thermophilic membrane reactor (TMR): factors affecting foam formations. Environ. Sci. Pollut. Res. http://dx.doi.org/10.1007/s11356-0167983-4. Available online 4 November 2016, (in press). Demir, O., Filibeli, A., 2016. The investigation of the sludge reduction efficiency and mechanisms in oxic-settling-anaerobic (OSA) process. Water Sci. Technol. 73 (10), 2331e2323. Eckenfelder Jr., W.W., 1999. Industrial Water Pollution Control, third ed. McGraw Hill International. ISBN: 978e0070393646. Foladori, P., Andreottola, G., Ziglio, G., 2010. Sludge Reduction Technologies in Wastewater Treatment Plants. IWA Publishing, London, UK. ISBN: 9781843392781. Huang, H., Liu, J., Ding, L., 2015. Recovery of phosphate and ammonia nitrogen from the anaerobic digestion supernatant of activated sludge by chemical precipitation. J. Clean. Prod. 102, 437e446.
International Organization for Standardization (ISO), 2007. ISO 8192:2007: Water Quality e Test of Inhibition of Oxygen Consumption by Activated Sludge for Carbonaceous and Ammonium Oxidation. ISO, Geneva. IRSA-CNR, 1984. Analytical Methods for Sludge. Technological Parameters (In Italian). Polygraph Institute of National Mint, Rome. IRSA-CNR, No. 64, ed. Juteau, P., 2006. Review of the use of aerobic thermophilic bioprocesses for the treatment of swine waste. Livest. Sci. 102 (3), 187e196. Kelessidis, A., Stasinakis, A.S., 2012. Comparative study of the methods used for treatment and final disposal of sewage sludge in European countries. Waste Manag. 32, 1186e1195. Kristensen, G., Jørgensen, P.E., Henze, M., 1992. Characterization of functional micro-organism groups and substrate in activated sludge and wastewater by AUR, NUR and OUR. Water Sci. Technol. 25 (6), 43e57. Li, R., Teng, W., Li, Y., Wang, W., Cui, R., Yang, T., 2017. Potential recovery of phosphorus during the fluidized bed incineration of sewage sludge. J. Clean. Prod. 140, 964e970. http://dx.doi.org/10.1016/j.jclepro.2016.06.177. Menoni, L., Bertanza, G., 2016. Wet Oxidation of sewage sludge: a mathematical model for estimating the performance based on the VSS/TSS ratio. Chem. Eng. J. 306, 685e692. Panepinto, D., Fiore, S., Genon, G., Acri, M., 2016. Thermal valorization of sewer sludge: perspectives for large wastewater treatment plants. J. Clean. Prod. 137, 1323e1329. rez-Elvira, S.I., Nieto Diez, P., Fdz-Polanco, F., 2006. Sludge minimisation techPe nologies. Rev. Environ. Sci. Bio/Technol. 5 (4), 375e398. Rada, E.C., Ragazzi, M., Villotti, S., Torretta, V., 2014. Sewage sludge drying by energy recovery from OFMSW composting: preliminary feasibility evaluation. Waste Manag. 34 (5), 895e866. Samolada, M.C., Zabaniotou, A.A., 2014. Comparative assessment of municipal sewage sludge incineration, gasification and pyrolysis for a sustainable sludgeto-energy management in Greece. Waste Manag. 34 (2), 411e420. Tyagi, V.K., Lo, S.-L., 2013. Sludge: a waste or renewable source for energy and resources recovery? Renew. Sustain Energy Rev. 25, 708e728. Wei, Y., Van Houten, R.T., Borger, A.R., Eikelboom, D.H., Fan, Y., 2003. Minimization of excess sludge production for biological wastewater treatment. Water Res. 37, 4453e4467. Yenigün, O., Demirel, B., 2013. Ammonia inhibition in anaerobic digestion: a review. Process Biochem. 48, 901e911. Zhang, G., Yang, J., Liu, H., Zhang, J., 2009. Sludge ozonation: disintegration, supernatant changes and mechanisms. Bioresour. Technol. 100, 1505e1509.
Please cite this article in press as: Collivignarelli, M.C., et al., Minimization of municipal sewage sludge by means of a thermophilic membrane bioreactor with intermittent aeration, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.12.101