- Email: [email protected]
Potential for valorization of dehydrated paper pulp sludge for biogas production: Addition of selected hydrolytic enzymes in semi-continuous anaerobic digestion assays
Energy 126 (2017) 326e334
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Potential for valorization of dehydrated paper pulp sludge for biogas production: Addition of selected hydrolytic enzymes in semicontinuous anaerobic digestion assays Sabina Kolbl a, Petra Forte-Tav cer b, Bla z Stres c, d, * a
University of Ljubljana, Faculty of Civil and Geodetic Engineering, Jamova 2, SI-1000, Ljubljana, Slovenia University of Ljubljana, Faculty of Natural Sciences and Engineering, Snezniska ulica 5, Si-1000, Ljubljana, Slovenia University of Ljubljana, Biotechnical Faculty, Department of Animal Science, Group for Microbiology and Microbial Biotechnology, Jamnikarjeva 101, 1000 Ljubljana, Slovenia d University of Insbruck, Institute of Microbiology, Technikerstrasse 25 d, 6020 Innsbruck, Austria b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 November 2016 Received in revised form 10 March 2017 Accepted 12 March 2017 Available online 16 March 2017
The effects of five commercially available hydrolytic enzyme additives on methane yields from dehydrated paper pulp sludge (DPPS) were determined in 5L pilot-scale reactors operated in semi-continuous mode for 60 days. Methane production was 40% and 43% higher in reactors receiving Novozymes and Novalin additives, respectively, compared to controls. Effects of time of DPPS inclusion on bacterial and archaeal microbial communities were many times larger than effects of enzyme type as enzyme addition did not produce rearrangements larger than random fluctuations observed in reactors receiving only DPPS. The ratio between volatile organic acids and alkalinity signified progressive decrease in process stability until day 45 irrespective of enzyme supplementation. Complementation with clarified pig slurry (1.5% vol.) for subsequent 15 days effectively stabilized process parameters and was sufficient for microbial communities to maintain DPPS hydrolytic capacity and process additional carbon flow derived from hydrolytic activity of enzyme additives. Consequently, initially unadapted full-scale biogas plant inoculum was capable of significantly increased methane yields from DPPS. Based on annual DPPS availability in EU the potential for additional energy recovery was estimated to be in the range of nearly 1 TJ. © 2017 Published by Elsevier Ltd.
Keywords: Anaerobic digestion Paper pulp sludge Methane yield Hydrolysis Enzyme addition Semi-continuous Co-substrate
1. Introduction Dehydrated paper pulp sludge (DPPS) represents a solid waste material composed of pulp residues and ash generated from the pulping and paper-making process [1,2]. It is a solid residue recovered from the wastewater and consists primarily of complex plant polysaccharide materials such as cellulose, lignin and hemicelluloses [3]. The effluents from the pulp and paper industry cause considerable damage to the receiving waters if discharged untreated and are characterized by high biochemical oxygen demand (BOD), chemical oxygen demand (COD) and contain chlorinated compounds, suspended solids (i.e. fibers), fatty acids, lignin and
* Corresponding author. University of Ljubljana, Biotechnical Faculty, Department of Animal Science, Group for Microbiology and Microbial Biotechnology, Jamnikarjeva 101, 1000 Ljubljana, Slovenia. E-mail address: [email protected] (B. Stres). http://dx.doi.org/10.1016/j.energy.2017.03.050 0360-5442/© 2017 Published by Elsevier Ltd.
sulphur compounds. Most of the solids are removed after the mechanical treatment resulting in a DPPS that contains large quantities of fibers [2,4]. In general, most of DPPS (69%) is landfilled or incinerated (21%) and as such represents measurable financial burden to the industry: 50.000 tons of paper mill sludge are being produced and disposed at the average cost of transportation and landfill disposal of 75V per tonne by paper industry in Slovenia annually [1]. In addition to its costly disposal for the paper mills its energy potential and nutrient content are lost [5]. Another option for at least partial valorization of these resources is inclusion of DPPS into existing agricultural biogas plant anaerobic digestion (AD) located nearby. Increased methane yields can be attained using enzymatic pre-treatments including the reduction of organic matter, however, without significant installation and construction costs related to enzyme utilization [5,6]. Microorganisms that are present in the anaerobic reactor secrete enzymes to help them enhance the degradation of cellulose, hemicellulose, lipids,
S. Kolbl et al. / Energy 126 (2017) 326e334
proteins, starch and other complex plant polymers. Hydrolysis is nevertheless the rate limiting step in the anaerobic digestion [7,8] hence enzymatic pretreatments are introduced to increase the extent of organic matter depolymerization and improved methane yields from anaerobic digestion. Economical feasibility of these approaches depends on performance of enzyme types under conditions close to those utilized at full scale and additive price. So far, the effects of short-term (60 days) additions of various enzyme additives on DPPS methane yields, process parameters and microbial communities of bacteria and archaea in agricultural biogas plant inoculum have not been assessed comparatively. Agricultural biogas plants encounter substrate shortage due to market fluctuations, differences in annual cycles or unanticipated weather events. The short-term addition of DPPS as a sole substrate was tested in this study as an approach to bridge the time gap between the availability of regular agricultural substrates. In addition, enzyme additives and pig slurry complementation were tested as mitigation strategies to increase methane yields from DPPS and stabilize anaerobic digestion of DPPS. In this study (i) five different enzyme additives available on market were used to increase methane yields from DPPS over 60 days; (ii) methane yields were determined in 5 L semi-continuous reactors and a number of targeted process parameters (n ¼ 16) were monitored in all variants over the same period; (iii) rearrangements in bacterial and archaeal community structure were monitored in order to delineate the contribution of stochastic changes over time relative to those potentially related to the type of enzymatic supplementation; (iv) based on annual DPPS availability in EU the potential for additional energy recovery was estimated and sustainability of short-term DPPS inclusion into 1 MW agricultural biogas plant assessed.
2. Materials and methods 2.1. Baseline methane yields from paper pulp Methane yields relevant for DPPS were obtained from Methane yield Database [9]. Average, quartiles, median, 95% confidence intervals were calculated for paper sludge (n ¼ 38 data points). Additional entries (n ¼ 16) collected from Kinnunen et al. [10] and Meyer & Edwards [11] were submitted in order to complement the data on methane yields from DPPS and are hereafter part of the freely available Methane yield Database [7]. In addition, the experimental methane yields determined in this study were submitted to Methane Yield Database [7] under the accession numbers mdb2215-mdb2261. Theoretical methane yield of dehydrated paper pulp sludge (DPPS) was calculated using Bushwell's equation [12] assuming that most of the organically degradable part of the paper sludge was composed of cellulose (C6H10O5) with an ash content of 22% and 30% for DPPS1 and DPPS2, respectively (Table 1).
Table 1 Characteristics of substrates and inoculum used in assays. BDL- below detection limit. Parameter
Swine slurry
DPPS 1
DPPS 2
Inoculum
TS (%) VS (% TS) VS (% FM) pH VOA (mg l1) TIC (mg l1) VOA TAC1 Ntotal (mg l1) 1 NHþ 4 -N (mg l )
0.33 ± 0.01 41.03 ± 1.19 0.13 ± 0.01 7.24 ± 0 1403 ± 61 2232 ± 55 0.69 ± 0.38 2500 ± 25 300 ± 12
16.99 ± 0.02 78.37 ± 0.27 13.31 ± 0.01 6.60 ± 0.01 3549 ± 494 15087 ± 2957 0.237 ± 0.01 1055 ± 35 BDL
17.64 ± 0.02 69.83 ± 0.12 12.32 ± 0.09 6.54 ± 0.01 3212 ± 359 14932 ± 3247 0.215 ± 0.02 905 ± 40 BDL
8.53 ± 0.18 69.17 ± 0.60 5.90 ± 0.16 7.87 ± 0.01 5534 ± 247 13275 ± 175 0.417 ± 0.01 5230 ± 60 1301 ± 39
327
2.2. Semi-continuous experiments 2.2.1. Inoculum and paper pulp sludge Inoculum was collected from a mesophilic agricultural biogas plant (ABGP) Sijanec (Ormo z, Slovenia; 1 MW) following the established procedure described before [6]. Two distinct batches of DPPS, (batch one - DPPS 1; batch two - DPPS 2) were collected in the period of 1 month from 2000 PE (population equivalents) paper pulp wastewater treatment plant located at the site of the hygienic paper manufacturer Paloma d.d. (Sladki vrh, Slovenia; http://web. paloma.si/en/). DPPS samples were stored at 2 C and homogenized by mixing before use in experiments. Hygienic papers of Paloma are produced from pure cellulose, waste paper and board. Flotation and flocculation are used to separate paper pulp from wastewater and further dehydrated by rotating drums to form DPPS (130e210 t daily; 40% TS). Different types of paper that are produced within Paloma d.d. rely on distinct technological approaches and hence yield DPPS batches with distinct characteristics and theoretical methane yields. Characteristics of inoculum and all substrates are listed in Tables 1 and 2. In order to complement the semi-continuous reactors with sources of trace metals and nutrients during the observed phase of process instability (d46-60; n ¼ 14 days) clarified pig slurry (PS) [13], i.e. slurry devoid of particulate material by prolonged settling was collected from a nearby (d < 20 km) pig farm Vucja vas, Slovenia (capacity ¼ 250 pigs). 2.2.2. Residual degradation of inoculum by five enzyme brands To assess whether the existing inoculum contained a fraction of organic matter amenable to further enzymatic degradation by provided additives, five brands of commercially available products were tested (Table S1) in the first experiment. Two Automatic Methane Potential Test unit (AMPTS II, Bioprocess Control, Sweden) upgraded to 5 L scale were used in semi-continuous experiments as detailed before [6]. All anaerobic reactors (n ¼ 18) received 4000 mL inoculum from BGP Sijanec. Negative controls represented anaerobic reactors (n ¼ 3) that received tap water, while all the others received enzymatic amendments: hydrolytic enzymes Novozymes (Denmark) (n ¼ 3), hydrolytic enzymes Novalin (NovaBiotec, Germany) (n ¼ 3), Micropan Biogas additive (Eurovix, USA) (n ¼ 3), additive BFL 4400AN (Biofuture, Ireland) (n ¼ 3) and Zeolit M (Ipus, Austria) (n ¼ 3). Daily preparations and additions of enzyme additives into anaerobic digesters were conducted essentially as described before [6]. Before the start up, each reactor was flushed with N2 for 3 min. Mixing (15 min on-off regime) continued throughout the operation of reactors and temperature in reactor water baths was maintained at 38 ± 2 C. 2.2.3. Testing the efficiency of five enzyme brands In the second experiment five commercially available additives of different brands were amended to the anaerobic reactors on daily basis: hydrolytic enzymes Novozymes (Denmark) (n ¼ 3), hydrolytic enzymes Novalin (NovaBiotec, Germany) (n ¼ 3), Table 2 Chemical analysis of total metal content in both batches of DPPS substrate introduced to 5l semicontinuous reactors in this study (mg kg1 TS1). Metals
DPPS1
DPPS2
Cd Cr Cu Hg Ni Pb
0.5 10.2 59.8 0.003 3 12
0.7 18.4 63.4 0.040 9.8 12.8
328
S. Kolbl et al. / Energy 126 (2017) 326e334
Micropan Biogas additive (Eurovix, USA) (n ¼ 2), additive BFL 4400AN (Biofuture, Ireland) (n ¼ 2) and Zeolit M (Ipus, Austria) (n ¼ 2). Dosage of each additive was in accordance with manufacturer's guidelines (Table S1) following the full-scale anaerobic digestion guidelines depending on the organic loading and volatile solids or total solid content of substrate. In total, 12 anaerobic digesters were supplied with additives (Table S2), whereas the rest served as controls. Organic loading rate (OLR) at the full scale reactor of ABGP Sijanec from which inoculum was obtained for testing was 10 g VS L1 day1, five times higher than the suggested starting 2.2 g VS L1 day1 for DPPS [5,14]. The three sections (DPPS1, DPPS2 and DPPS2þPS) of 60 day semi-continuous experiment (Fig. S1) were projected to take place at full scale installation [14] in order to minimize operating costs. The 5 L reactors (n ¼ 15) of AMPTS II unit (AMPTS II, Bioprocess Control, Sweden) received 4000 mL inoculum and daily organic loading of 2.2 g VS L1 day1 equal to 66.1 g of DPPS1 or 71.4 g of DPPS2 with the addition of 183.4 mL and 179.6 mL of industrial tap water, respectively. Total working volume was 4250 mL. DPPS1 was dosed in period from day 0 to day 28, DPPS2 from day 29e45 and a mixture of DPPS2 and cleared pig slurry (PS) (60 mL day1) and water (119.6 mL day1) from day 46 to day 60 (DPPS2þPS; Table S2, Fig. S1). In order to retain mass balance equal amounts of material and liquids were introduced and removed on daily basis as described before [11]. This collected material was used for subsequent analyses of process parameters. In short, the collected samples were centrifuged for 3 min at 5000 g and supernatant was stored at 20 C until further analyses of process parameters. The remaining solid fraction was homogenized, subsampled (0.5 g) at 4 C and frozen at 20 C until DNA extraction. In addition to five-fold reduction in OLR, hydraulic retention time (HRT) used at ABGP Sijanec was shortened from original 60 dayse18 days in semi-continuous experiments as suggested before [15] for optimal performance in degradation of DPPS. Mixing (15 min on-off regime) continued throughout the operation of reactors and temperature in reactor water baths was maintained at 38 ± 2 C. Process parameters (n ¼ 16), total solids, volatile solids, C1-C6 short chain fatty acids (SCFA), pH, total inorganic carbon (TIC), volatile organic acids (VOA), the total VOA/TIC ratio, total nitrogen, ammonia, electrical conductivity, particle size distribution, and soluble chemical oxygen demand (sCOD) were determined from samples collected every 5 days. 2.2.4. Rearrangements in bacterial and archaeal microbial communities Power Fecal DNA Extraction kit (MOBIO, USA) was used to extract DNA from samples stored at 20 C. DNA quality and purity (A230, A260, A280) were assessed spectrophotometrically using NanoVue according to manufacturer's instructions. Amplification of the bacterial and archaeal genes for 16S rRNA was conducted as described before [16] according to PCR protocol: 94 C for 5 min, followed by 30 cycles of denaturation at 94 C for 30 s, annealing at 54 C for 45 s, and elongation at 72 C for 90 s. The final extension was at 72 C for 7 min. The PCR reaction mixture (25 mL) contained 2.5 mL 10 PCR buffer, 2.5 mM MgCl2, 200 mM of each nucleotide, 200 nM of each primer, 1.0 U DNA polymerase, 2e20 mg BSA/L, and 1.0 mL of undiluted template. Both forward primers used in fingerprinting of bacterial and archaeal microbial communities were marked with JOE (6-carboxy-40 , 5’-dichloro-20 , 7'-dimethoxy- fluorescein) and amplicons were separated using ABI3100 sequencer. The resulting electropherograms were converted to text formatted data in ABI3100 sequencer proprietary GENEMAPPER software and analyzed in R as described before [16].
Incubation time in experiment and hydrolytic enzyme additive brand were used as explanatory variables in two-way NP-MANOVA in PAST [17] and variance partitioning as described before [18,19] to assess their contribution to changes in bacterial and archaeal microbial communities. 2.2.5. Physical and chemical analyses Determination of total solids (TS), volatile solids (VS) and soluble chemical oxygen demand (sCOD) was made according to the APHA [20]. For measurements of volatile organic acids (VOA), total inorganic carbon (TIC) and ratio VOA/TIC Biogas Titrator Manager TIM 840 (Hach Lange) was used. Total nitrogen and ammonia was determined by SANplus Continuous Flow Analyzer (Skalar Analytical B.V., Netherland) according to the manufacturer's instructions [6]. Electrical conductivity and pH were measured using multimeter HQ40D (Hach Lange). Particle size analysis was performed by Laser Particle Sizer Analysette 22 Wet Dispersion Unit (Fritsch, Germany) as described before [11] using measuring range from 0.08 to 2000 mm. Particle size was calculated through laser diffraction and Fraunhofer calculation method that is incorporated in software FRITSCH Mass Control [14]. Measurements of total Cd, Cr, Cu, Hg, Ni and Pb in DPPS were provided by ERICO Velenje according to the ISO 17294-2:2003 [21]. Ether extraction was used to prepare short chain fatty acids (SCFA) for analyses as described before [6,22] and their concentrations determined by gas chromatograph Agilent 7890. Capillary column Agilent J & W GC columns DB-FFAP, 30 m 0.530 mm x 1 mm layer of stationary phase was used. The injector and detector temperatures were 200 C and 300 C, respectively. Initial oven temperature was 70 C with residence time of 1 min, then ramped at 20 C min1 to 120 C and further 10 C min1 to final oven temperature of 200 C with residence time of 3 min. Carrier gas (mobile phase) was Helium with flow 5 mL min1, nitrogen flow 25 mL min1, detector gas was hydrogen with flow 30 mL min1, synthetic air with flow 400 mL min1. 2.3. EU wide methane potential of DPPS Annual amounts of paper sludge produced in EU were calculated based on yearly paper production assuming 0.3 m3 of DPPS produced per 1000 kg of paper [23] (http://www.cepi.org/node/ 19363). In addition, Chamber of Commerce and Industry and Statistical Office of the Republic of Slovenia provided the data on annual amounts of DPPS produced in Slovenia (http://www.stat.si/ StatWeb/en/home; https://www.gzs.si/zdruzenje_za_papirno_in_ papirno_predelovalno_industrijo). First, initial parameters for determination of methane yields from DPPS and potential energy gains were estimated from (i) the data on annual production of DPPS, (ii) the DPPS organic loading in anaerobic full scale digesters [6], and (iii) the extent of increased methane yields from DPPS as a result of enzyme addition to DPPS in semi-continuous experiments. Second, total methane yields obtained from semi-continuous experiments in this study were further distinguished into three values, corresponding to: (i) the additional methane produced from the previously undigested substrate in inoculum amended solely with various enzyme brands; (ii) the methane produced as a result of increased DPPS digestion due to hydrolytic enzyme efficiency; (iii) methane produced as a result of complementation with clarified pig slurry. Third, fluctuations in production of different types of paper within Paloma d.d. necessitate distinct technological approaches. Hence two batches of DPPS were used to estimate methane yields
S. Kolbl et al. / Energy 126 (2017) 326e334
in this study and were contrasted with previously recorded data in Methane Yield Database [9]. These collated data were used to estimate the mean, lower (25%) and upper (75%) quartiles of the short-term (60 day) economic sustainability of enzyme additive purchase for the ABGP Sijanec from which inoculum was obtained using the approach described before [6,7,11] and taking into account on-site specific parameters of particular biogas plant (ABGP Sijanec, Slovenia; internal data, unpublished [6,11]). 3. Results and discussion 3.1. Residual methane yields from inoculum The semi-continuous analysis of additional methane formation obtained from inoculum amended with enzyme additives showed that most of enzyme additives significantly increased methane yields from inoculum used in this study (Fig. S2A). Methane production from enzyme only amended reactors effectively ceased after 35 days, whereas most of the process parameters remained unaffected, irrespective of enzyme brands (Figs. S2BeD). The largest surplus of residual methane was observed for Novozymes (27%), Micropan Biogas (11%) and Novalin (4%), whereas no significant activity or even shortage were observed for Biofuture (2%) and Zeolit M (16%), respectively. Biofuture contained microbial cultures and hence operated on the lines of bioaugmentation, whereas Zeolit M relied on silicates with specific nutrient allocations and specific absorption capabilities (Table S1). Growth of novel bioaugmenting microorganisms from Biofuture additive was apparently not favored under the conditions of depleting substrate present in inoculum, whereas silicates with specific absorption capabilities most probably immobilized decreasing amounts of mineral nutrients and hence did not produce significant positive effects on existing microorganisms. The short-term (i.e. 40 day) inclusion of specific enzyme additives into 5 L semi-continuously operated reactors thus resulted in distinct residual methane outputs from the same batch of ABGP Sijanec inoculum. As previous estimates of background methane formation after enzyme addition were based on batch scale experiments [24] the results of this study represent the first estimate of proportions of yet undigested substrate in inoculum. The benefits of enzyme introduction into full-scale biogas plant are apparently substantial and could support at least the short-term campaigns. 3.2. Enzymatically increased methane yields from DPPS The comparison between the methane yields obtained from various sections of 60 day semi-continuous operation (Fig. 1AeC) of DPPS anaerobic digestion illustrates the profound differences between the studied variants after the initial ABGP substrate supply was discontinued and DPPS introduced. Particle size of DPPS was below 2 mm [6] (Fig. S3A) making DPPS particles of sufficient small size for anaerobic digestion, avoiding thus the additional particle size homogenization. The three sections (DPPS1, DPPS2 and DPPS2þPS) tested in 60 day semi-continuous experiment (Fig. S1) were projected to take place at full scale installation [14]. The sole transition from standard ABGP substrate mix to DPPS was accompanied by fivefold decrease in loading rate and consequent characteristic decrease in methane production (Fig. 1B). During the addition of DPPS1, controls exhibited average daily methane production (day 8 to day 28) of 134.6 ± 12.7 mL CH4 g1 VS1 day1, whereas variants with additives Novozymes, Novalin, Micropan Biogas, Biofuture and zeolite M all exhibited positive effects on methane production: 188.7 ± 12, 195.1 ± 10.9, 158.7 ± 13,
329
156.1 ± 7.5 and 147.7 ± 13 mL CH4 g1 VS1 day1, respectively. The onset of DPPS2 addition resulted in significantly lower average methane daily production from control reactors in period from day 29 to day 45 (82.4 ± 10.7 mL CH4 g1 VS1 day1) compared to DPPS1 (Fig. S3B), however, a decrease in methane production from Novalin, Novozymes, Micropan Biogas, Biofuture and zeolite M: 133.9 ± 12.5, 128.6 ± 13.5, 103.8 ± 7.7, 102.3 ± 10.1 and 103.3 ± 11.3 mL CH4 g1 VS1 day1, respectively, was also observed. In essence, control reactors using both batches of DPPS effectively produced methane yields in the ranges reported before [5]), however, enzyme additive supplementation was substantially more effective. Addition of pig slurry to DPPS2 in semi-continuous experiment (day 46 to day 60) was conducted in attempt to stabilize anaerobic digestion and prevent further decline in methane yields. No significant change in average daily methane production from DPPS2þPS (84.8 ± 16.3 mL CH4 g1 VS1 day1) and enzyme supplemented variants with different additives Novozymes, Novalin, Biofuture, Micropan Biogas and zeolite M was observed: 130 ± 21.2 mL, 127.6 ± 19.7, 103.3 ± 14.3, 105.5 ± 11.7 and 107 ± 10.1 mL CH4 g1 VS1 day1, respectively. Nevertheless, the addition of pig slurry as co-substrate prevented further trend of declining of methane yields observed from day 29 to day 45 for DPPS2 (Fig. 1B). The use of various experimental models and scales in methane yield estimation (e.g. batch vs. semi-continuous; e.g. 300 mL vs. 4 L) was shown to exhibit substantial effects on estimates of full-scale sustainability [6,7,11] used in industrial settings for decisionmaking. In line with these observations, up to 30% overestimates in methane yields were observed for batch experiments in comparison to semi-continuous due to the effects of differences in scale and assembly [6,25]. For example, methane yields from 160 mL batch experiments treating biosludge from paper industry was up to 30% higher compared to 6 L and 400 day semi-continuous experiment [10]. Higher methane yields can be obtained in batch assays due to prolonged (e.g. 30 day) anaerobic digestion that is long enough for the utilization of almost all degradable substrates. However, the obtained methane yields in semi-continuous assays are determined by the relationship between HRT and degradability of the substrate. This distinction is important for the industrial decision makers to selectively base their sustainability estimates on semi-continuous results rather than batch test data to avoid overestimation of methane yields. In this study, methane yields from batch DPPS2 degradation were 32.5% higher than those obtained from semi-continuous digestion of DPPS2 (Fig. S3B). Complementation DPPS2þPS resulted in three times higher methane yields in 5 L batch experiments (Fig. S3B), illustrating the extent of influence on biogas plant decision-making support. In addition, a wide range of methane yields were reported for DPPS in past experiments ranging from 59 to 370 mL CH4 g1 VS1 [9e11,26]. In this respect, to obtain methane yields relevant for industrial process planning the use of experimental procedures mirroring conditions utilized at full scale as close as possible is necessary [6,9]. Previous studies suggest that deficiencies in micronutrients can lead to diminished performance of anaerobic digestion [8,27]. Therefore aliquots of clarified pig slurry (PS; 1.5% vol.) were amended as cosubstrate along with DPPS2 from day 46 to day 60 (DPPS2þPS; Table S2) to increase methane yields and to test whether some unaccounted for micronutrient limitations (e.g. other metals like Mn, Mo, Se, Zn) or depleted macronutrients (N) or process parameters (alkalinity) were responsible for low methane production irrespective of enzyme addition. The addition of PS as cosubstrate to DPPS reactors (day 46e60) restored the positive effects of enzyme addition (Fig. 1B) showing that optimal concentrations of additional nutrients in the environment were necessary
330
S. Kolbl et al. / Energy 126 (2017) 326e334
Fig. 1. Normalized cumulative methane production from 60 day semi-continuous. anaerobic digestion of DPPS1, DPPS2 and DPPS2þPS. Methane production was normalized to control reactors receiving DPPS only (A). (B) Daily methane production with standard deviations. (C) Cumulative methane production in the same experiments. DPPS1, DPPS2, DPPS þ PS were administered to all anaerobic reactors from day 1e28, day 29e45 and day 46e60, respectively.
for the microbial communities to either maintain their own hydrolytic activities and/or to process the additional carbon flow derived from hydrolytic action of introduced enzyme additives. Additional batch experiment complementing DPPS2 with PS (Fig. S4) was in line with this observation. This puts the nutrient deficit in DPPS observed before [5] and lack of microbial hydrolytic enzyme activity (this study) into new perspective. Although DPPS1 and DPPS2 also contained microbial biomass in addition to existing inoculum, microbial activity was apparently limited due to the increasing nutrient depletion over time (days 1e45) and hence did not actively contribute to the degradation of primary sludge derived fibers consisting more or less of cellulose [15,28]. Secondly, it also shows that nutrient limitation has the potential to diminish the positive effects of introduced hydrolytic enzyme additives. Consequently, a simple complementation adopting nutrient rich co-substrates for optimal performance has the potential to support higher methane yields derived from DPPS on full scale, possibly also at higher OLR and in long-term applications in the future (t > 60 days). The compatibility of the two substrates and short-term enzyme addition to 5 L reactors operated in semi-continuous mode enables the adoption of DPPS as an additional substrate for ABGP treating clarified pig slurries containing excess nutrients. In addition, this
enables paper industry to complement its industrial biogas reactors with additional substrates from nearby agricultural practices in order to maximize methane yields from DPPS while minimizing the remaining sludge volume. From the perspective of ABGP the short term strategy of 60 day semi-continuous DPPS degradation with enzyme additives resulted in cumulatively 40% and 43% higher methane production in reactors supplemented with Novozymes and Novalin additives, respectively compared to control reactors receiving daily DPPS only. The obtained methane yields were in comparable range as reported before for semi-continuous degradation of untreated activated sludge from Swedish pulp and paper mills [5]. However, daily introduction of novel substrate in semi-continuous operation includes also daily removal of a minor portion of preexisting inoculum and previously supplied DPPS substrate. In our case this represented daily removal of 250 g aliquot of mixed DPPS, water and inoculum from total operating volume of 4250 mL, a ratio similar to ABGP full-scale operations. Further, the major part of DPPS degradation (>90%) was completed under batch conditions in 6 days (Fig. S4A). Consequently, only minute portions of fresh DPPS could be removed due to reactor operation procedures in consecutive days and increased methane production was derived from the digestion of a major part of past additions and newly added DPPS.
S. Kolbl et al. / Energy 126 (2017) 326e334
Our results also show that different enzyme additives can exhibit markedly different efficiencies, most probably due to the distinct inoculum - paper sludge combinations and hence deserve further study. Consequently, a prescreening for selection of most promising enzyme additives suitable for local inoculum and operating conditions can easily be performed on-site in real-time as described in this study. 3.3. Process parameters during the addition of DPPS Several important process parameters changed during the shift from general ABGP substrate mix [6] to DPPS as sole substrate at lower OLR and HRT (Fig. 2AeG): pH, total nitrogen, ammonia, TIC, conductivity and sCOD. The latter two kept decreasing throughout the experiments, however the pH and TIC increased sufficiently only after pig slurry addition, that helped stabilize the indicative VOA/TIC ratio. This VOA/TIC ratio was used as parameter describing process instability [11] and showed that instability was not due to accumulation of VOA but due to the observed drop in alkalinity [15,29] accompanied by decrease in electrical conductivity of anaerobic digesters (Fig. 2E). However, no significant differences in electrical conductivity or nitrogen content between reactors could be detected. In the case of anaerobic digestion of biological wastes reported by Scano et al. [30] VOA/TIC ratio reached up to 0.65, however pH did not drop below 7. Sufficient alkalinity for stable anaerobic digestion of sludge was generally reported to be between 1000 and 5000 mg CaCO3 l1 [29], however up to 10,000 mg CaCO3 l1 were reported for stable processes treating biological wastes [31]. Significant drop in soluble chemical oxygen demand (sCOD) after transition to DPPS (Fig. 2F) reflected the polymer nature of DPPS and its effects on subsequent hydrolysis and fermentation steps. Complementation of DPPS with PS or other clarified slurries or manures has thus the potential to sustain the use of higher OLR of DPPS in the full-scale reactors, increased alkalinity and increased methane production. 3.4. Fluctuations in microbial community structure of bacteria and archaea In this study, changes in bacterial and archaeal community structure were assessed using fast automated fingerprinting approaches in order to identify the extent of change in response to the time of incubation and the type of amended enzyme additives. Bacterial and archaeal communities exhibited a gradual change over the course of DPPS administration, in line with past observations that flexible community structure correlates with stable community function in methanogenic bioreactors [32] sustaining changes in carbon and electron flow [33]. The transition from agricultural substrate to DPPS as the sole substrate introduced significant reorganization in bacterial and archaeal microbial communities in control reactors in 60 days. The results of two-way NPMANOVA (npermut ¼ 9999) for bacterial and archaeal communities, respectively, showed that the administration of various enzyme additives (pBacteria<0.001; pArchaea ¼ 0.025), and incubation time with DPPS in experiment (pBacteria ¼ 0.001; pArchaea < 0.0001) were significantly associated with the changes in microbial communities, however, their interaction was surprisingly not significant (pBacteria>0.05; pArchaea > 0.05). This suggests that the effects of enzyme addition did not propagate through time under the conditions used in this study. Variation partitioning showed that the effect of incubation time with DPPS exerted on bacterial and archaeal communities was 5 and 6 times larger, respectively, than the effects of 60-day supplementation of the same reactors with different enzyme additives.
331
This shows that the extent of fluctuations in microbial communities in response to short-term enzyme additions were many times lower and well within the range of random fluctuations observed in control reactors receiving only DPPS. Hence, DPPS induced a general adaptation response of bacterial and archaeal microbial communities, whereas addition of enzymes did not induce significant enzyme-specific adaptations of microbial communities. In addition, the extent of rearrangements in microbial communities observed over time of inclusion of DPPS was 11 times larger for bacteria than for archaea in enzyme amended and control reactors. This further illustrates that the fluctuations in microbial communities were more pronounced at the level of bacterial microbial communities responsible for hydrolysis of the novel substrate than at the level of archaea producing methane from the resulting organic acids. Taken together, the short-term adoption of various enzyme additives in anaerobic digestion did not result in extensive enzyme-related rearrangements in microbial communities, and hence could be applied in industrial settings at least under comparable OLR of DPPS in the future. Our results are in line with previous findings that structurally responsive communities were also more functionally stable [34]. The exact rearrangements in microbial functional genes (after the introduction of DPPS as a novel substrate) are currently being investigated using metagenomic approaches in order to identify microbial functional genes and taxonomic groups of bacteria and archaea linked to DPPS degradation in microbial communities before and after DPPS administration. 3.5. The EU wide energy potential of DPPS The EU wide data collection showed that paper and board production in EU reached 91.1Mt in 2014 [23] (http://www.cepi. org/node/20088). Based on the availability of 4.6 Mt of DPPS the following three scenarios of methane-energy conversion were explored (Table 3): (i) energy production from methane derived solely from DPPS (control) including residual methane produced from additionally degraded inoculum; (ii) energy production from methane derived from DPPS as sole substrate after the deduction of residual methane generated by enzymes supplied to existing inoculum (accounting for additional degradation of digestate within the same reactors due to enzyme addition); (iii) energy yields from most efficient digestion of DPPS complemented with nutrients from PS (Table 3). Based on data on DPPS availability and improved methane yields approximately 1 TJ of additional energy could be obtained from DPPS annually (Table 3). Substrate shift from ABGP substrates to DPPS in this study resulted in fivefold reduction in OLR and HRT. Consequently, threefold reduction in average methane yields was observed between day 0 in comparison to days 8e60 (Fig. 1B). However, the addition of preselected enzyme additives resulted in up to 40% higher methane yields from DPPS in comparison to controls. The use of most efficient enzyme additives resulted in 195.1 ± 10.9 mL CH4 g1 VS1 day1, approximately half of the initial methane production observed at ABGP (Fig. 1B; [6,11]). Results reported in this study were indeed obtained during a short-term 60-day semi-continuous test period intended to test the potential effectiveness of short-term DPPS introduction into fullscale agricultural biogas plant. The adopted minimum intervention approach to inoculum and substrate to be used at full scale provided a framework for real-time estimation of methane yields and approximation of process parameters relevant for full-scale biogas plant management [6,11]. The data obtained in this study were compared to published data organized within Methane Yield Database [7] in order to provide realistic estimates on economic feasibility of DPPS inclusion into full-scale biogas plants including
332
S. Kolbl et al. / Energy 126 (2017) 326e334
Fig. 2. Overview of most important anaerobic process parameters during 60 day semi-continuous anaerobic digestion of DPPS1, DPPS2 and DPPS2þPS: (A) pH of reactor mixtures; (B) total nitrogen; (C) total inorganic carbon (TIC, i. e. alkalinity); (D) anaerobic process stability assessed through VOA/TIC ratio; (E) electrical conductivity as measure of ion salt concentration; (F) sCOD; (G) ammonium concentrations.
also on-site biogas plant specific sustainability parameters (ABGP Sijanec (unpublished internal data)). The short term inclusion of DPPS and its complementation with locally available pig slurry in combination with enzyme additives was accepted as viable short-
term alternative using the estimation approach described before [11], covering the expenses of enzyme purchasing, paper pulp and PS transport. Future experiments are directed toward increased OLR and HRT approaching those utilized at ABGP Sijanec during
S. Kolbl et al. / Energy 126 (2017) 326e334
333
Table 3 The potential for valorization of methane produced from DPPS in Slovenia and EU during the 60 day on-off introduction of various enzyme brands. Energy production (GJ year1)
a
Slovenia (DPPS) Slovenia (no residual methane)b Slovenia (addition of pig slurry)c EU (DPPS)a EU (no residual methane)b EU (addition of pig slurry)c
No enzymes
Novozymes
Novalin
Biofuture
M. Biogas
Zeolit M
6.9 5.0 7.0 630.4 460.2 645.5
10.1 7.4 10.3 926.4 676.3 948.7
10.3 9.9 10.5 944.9 907.1 967.6
8.1 8.3 8.3 744.1 759.0 762.0
8.2 9.6 8.4 758.7 880.1 776.9
8.0 7.1 8.2 734.8 654.0 752.4
a
Energy produced from methane generated during short-term addition of DPPS to agricultural biogas plant biomass. Energy produced from methane generated during short-term addition of DPPS to agricultural biogas plant biomass, after the deduction of residual methane produced from existing undigested residual substrate. c Energy produced from methane generated during short-term addition of DPPS to agricultural biogas plant biomass with clarified pig slurry as cosubstrate. b
regular operation with the final goal of DPPS inclusion into ABGP as long-term cosubstrate. 4. Conclusions Prescreening for most effective hydrolytic enzyme additives identified Novozymes and Novalin additives to increase methane production from DPPS for 40% and 43%, respectively. Daily supplementation of hydrolytic enzyme additives did not result in enzyme-specific rearrangements in bacterial and archaeal microbial communities. However, process instability observed in 45 days was bypassed by cost-effective complementation strategy with introduction of clarified pig slurry (1.5%). Data exploration on DPPS availability identified a substantial potential for annual energy production up to 1 TJ from enzymatically and co-substrate complemented DPPS in EU. Database linking Methane Yield Database: http://methane.fe.uni-lj.si/ Contributions SK and BS contributed equally. Performed the experiments: SK. Analyzed the data: SK, BS. Contributed reagents/chemicals/tools/ ideas: SK, PFT, BS. Wrote the paper: BS, SK. All authors agreed on the final version of the manuscript. Competing interests The authors declare no competing interests. Acknowledgments The authors acknowledge the support of Keter Invest energetika d.o.o. and Paloma d.d.. »Operation part financed by the European Union, European Social Fund. Operation implemented in the framework of the Operational Programme for Human Resources Development for the Period 2007e2013, Priority axis 1: Promoting entrepreneurship and adaptability, Main type of activity 1.1.: Experts and researchers for competitive enterprises.« SK acknowledges Programme "Water Science and Technology, and Geotechnic": P2-0180 (B). COST ACTION ES1302: European Network on Ecological Functions of Trace Metals in Anaerobic Biotechnologies is acknowledged for fruitful discussions and exchange of ideas during the final preparation of this manuscript.
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.energy.2017.03.050. References [1] Zule J, Cernec F, Likon M. Chemical properties and biodegradability of waste paper mill sludges to be used for landfill covering. Waste Manag Res 2007;25: 538e46. http://dx.doi.org/10.1177/0734242X07079188. [2] Mehdi Dashtban YH, Dashtban M, Chen S, Qin W, Song R. Lignin in paper mill sludge is degraded by white-rot fungi in submerged fermentation. J Microb Biochem Technol 2015:7. http://dx.doi.org/10.4172/1948-5948.1000201. [3] Jackson MJ, Line MA. Organic composition of a pulp and paper mill sludge determined by FTIR, 13C CP MAS NMR, and chemical extraction techniques. J Agric Food Chem 1997;45:2354e8. http://dx.doi.org/10.1021/jf960946l. [4] Bajpai P. Composition of waste. manag. pulp pap. mill waste. Cham: Springer International Publishing; 2015. p. 19e29. http://dx.doi.org/10.1007/978-3319-11788-1_3. [5] Karlsson A, Truong X, Gustavsson J, Svensson BH, Nilsson F, Ejlertsson J. Anaerobic treatment of activated sludge from Swedish pulp and paper mills e biogas production potential and limitations. Environ Technol 2011;32: 1559e71. http://dx.doi.org/10.1080/09593330.2010.543932. [6] Kolbl S, Paloczi A, Panjan J, Stres B. Addressing case specific biogas plant tasks: industry oriented methane yields derived from 5L Automatic Methane Potential Test Systems in batch or semi-continuous tests using realistic inocula, substrate particle sizes and organic loading. Bioresour Technol 2014. http:// dx.doi.org/10.1016/j.biortech.2013.12.010. [7] Batstone DJ, Jensen PD. Treatise on water science. Elsevier; 2011. http:// dx.doi.org/10.1016/B978-0-444-53199-5.00097-X. [8] Romero-Güiza MS, Vila J, Mata-Alvarez J, Chimenos JM, Astals S. The role of additives on anaerobic digestion: a review. Renew Sustain Energy Rev 2016;58:1486e99. http://dx.doi.org/10.1016/j.rser.2015.12.094. [9] Murovec B, Kolbl S, Stres B. Methane Yield Database: online infrastructure and bioresource for methane yield data and related metadata. Bioresour Technol 2015. http://dx.doi.org/10.1016/j.biortech.2015.04.021. €-Outinen A, Rintala J. Mesophilic anaerobic digestion of pulp [10] Kinnunen V, Yla and paper industry biosludgeelong-term reactor performance and effects of thermal pretreatment. Water Res 2015;87:105e11. http://dx.doi.org/10.1016/ j.watres.2015.08.053. [11] Meyer T, Edwards EA. Anaerobic digestion of pulp and paper mill wastewater and sludge. Water Res 2014;65:321e49. http://dx.doi.org/10.1016/ j.watres.2014.07.022. [12] Deublein D, Steinhauser A. Biochemistry. biogas from waste renew. resour.. Weinheim, Germany: Wiley-VCH Verlag GmbH & Co. KGaA; 2010. p. 97e100. http://dx.doi.org/10.1002/9783527632794.ch9. [13] Piccinini S, Cortellini L, Bonazzi G. Separation of pig slurry by sedimentation. In: Van Der Meer HG, Unwin RJ, Van Dijk TA, Ennik GC, editors. Anim. Manure grassl. Fodd. Crop. Fertil. or Waste? Proc. an int. Symp. Eur. Grassl. Fed. Wageningen, Netherlands, 31 Auguste3 Sept. 1987. Dordrecht: Springer Netherlands; 1987. p. 385e8. http://dx.doi.org/10.1007/978-94-009-36591_45. [14] Kolbl S, Panjan J, Stres B. Mixture of primary and secondary municipal wastewater sludge as a short-term substrate in 2 MW agricultural biogas plant: site-specific sustainability of enzymatic and ultrasound pretreatments n.d. doi:10.1002/jctb.4883. [15] Bayr S, Rintala J. Thermophilic anaerobic digestion of pulp and paper mill primary sludge and co-digestion of primary and secondary sludge. Water Res 2012;46:4713e20. http://dx.doi.org/10.1016/j.watres.2012.06.033. [16] Franke-Whittle IH, Manici LM, Insam H, Stres B. Rhizosphere bacteria and fungi associated with plant growth in soils of three replanted apple orchards. Plant Soil 2015;395:317e33. http://dx.doi.org/10.1007/s11104-015-2562-x. [17] Hammer Ø, Harper Dat a T, Ryan PD. PAST: paleontological statistics software
334
[18]
[19]
[20] [21] [22]
[23] [24]
[25]
[26]
S. Kolbl et al. / Energy 126 (2017) 326e334 package for education and data analysis. Palaeontol Electron 2001;4(1):1e9. http://dx.doi.org/10.1016/j.bcp.2008.05.025. Stres B, Philippot L, Faganeli J, Tiedje JM. Frequent freeze-thaw cycles yield diminished yet resistant and responsive microbial communities in two temperate soils: a laboratory experiment. FEMS Microbiol Ecol 2010;74: 323e35. http://dx.doi.org/10.1111/j.1574-6941.2010.00951.x. Stres B, Sul WJ, Murovec B, Tiedje JM. Recently deglaciated high-altitude soils of the Himalaya: diverse environments, heterogenous bacterial communities and long-range dust inputs from the upper troposphere. PLoS One 2013;8: e76440. http://dx.doi.org/10.1371/journal.pone.0076440. APHA. Standard methods for the examination of water and wastewater. Washington, USA: American Water Works Assn; 2005. ISO 17294-2. Water quality e application of inductively coupled plasma mass spectrometry (ICP-MS) e Part 2: determination of 62 elements. 2003. Novak D, Franke-Whittle IH, Pirc ET, Jerman V, Insam H, Logar RM, et al. Biotic and abiotic processes contribute to successful anaerobic degradation of cyanide by UASB reactor biomass treating brewery waste water. Water Res 2013;47:3644e53. http://dx.doi.org/10.1016/j.watres.2013.04.027. CEPI. Preliminary statistics. 2013. 2012e3, http://www.cepi.org/node/20088. ndez-Cegrí V, Borja R. Anaerobic digestion of Raposo F, De la Rubia MA, Ferna solid organic substrates in batch mode: an overview relating to methane yields and experimental procedures. Renew Sustain Energy Rev 2012;16: 861e77. http://dx.doi.org/10.1016/j.rser.2011.09.008. Ruffino B, Fiore S, Roati C, Campo G, Novarino D, Zanetti M. Scale effect of anaerobic digestion tests in fed-batch and semi-continuous mode for the technical and economic feasibility of a full scale digester. Bioresour Technol 2015;182:302e13. http://dx.doi.org/10.1016/j.biortech.2015.02.021. Steffen F, Requejo A, Ewald C, Janzon R, Saake B. Anaerobic digestion of fines from recovered paper processing e influence of fiber source, lignin and ash content on biogas potential. Bioresour Technol 2016;200:506e13. http:// dx.doi.org/10.1016/j.biortech.2015.10.014.
[27] Voelklein MA, O’ Shea R, Jacob A, Murphy JD. Role of trace elements in single and two-stage digestion of food waste at high organic loading rates. Energy 2017;121:185e92. http://dx.doi.org/10.1016/j.energy.2017.01.009. ve J, Dewil R. Principles and potential of the [28] Appels L, Baeyens J, Degre anaerobic digestion of waste-activated sludge. Prog Energy Combust Sci 2008;34:755e81. http://dx.doi.org/10.1016/j.pecs.2008.06.002. [29] Lin Y, Wang D, Wu S, Wang C. Alkali pretreatment enhances biogas production in the anaerobic digestion of pulp and paper sludge. J Hazard Mater 2009;170:366e73. http://dx.doi.org/10.1016/j.jhazmat.2009.04.086. [30] Scano EA, Asquer C, Pistis A, Ortu L, Demontis V, Cocco D. Biogas from anaerobic digestion of fruit and vegetable wastes: experimental results on pilot-scale and preliminary performance evaluation of a full-scale power plant. Energy Convers Manag 2014;77:22e30. http://dx.doi.org/10.1016/ j.enconman.2013.09.004. [31] Banks CJ, Chesshire M, Heaven S, Arnold R. Anaerobic digestion of sourcesegregated domestic food waste: performance assessment by mass and energy balance. Bioresour Technol 2011;102:612e20. http://dx.doi.org/10.1016/ j.biortech.2010.08.005. [32] Fernandez AS, Hashsham SA, Dollhopf SL, Raskin L, Glagoleva O, Dazzo FB, et al. Flexible community structure correlates with stable community function in methanogenic bioreactor communities perturbed by glucose. Appl Environ Microbiol 2000;66:4058e67. http://dx.doi.org/10.1128/AEM.66.9.40584067.2000. ndez A, Huang S, Seston S, Xing J, Hickey R, Criddle C, et al. How stable is [33] Ferna stable? Function versus community composition. Appl Environ Microbiol 1999;65:3697e704. [34] Hashsham SA, Fernandez AS, Dollhopf SL, Dazzo FB, Hickey RF, Tiedje JM, et al. Parallel processing of substrate correlates with greater functional stability in methanogenic bioreactor communities perturbed by glucose. Appl Environ Microbiol 2000;66:4050e7. http://dx.doi.org/10.1128/AEM.66.9.40504057.2000.
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Potential for valorization of dehydrated paper pulp sludge for biogas production: Addition of selected hydrolytic enzymes in semicontinuous anaerobic digestion assays Sabina Kolbl a, Petra Forte-Tav cer b, Bla z Stres c, d, * a
University of Ljubljana, Faculty of Civil and Geodetic Engineering, Jamova 2, SI-1000, Ljubljana, Slovenia University of Ljubljana, Faculty of Natural Sciences and Engineering, Snezniska ulica 5, Si-1000, Ljubljana, Slovenia University of Ljubljana, Biotechnical Faculty, Department of Animal Science, Group for Microbiology and Microbial Biotechnology, Jamnikarjeva 101, 1000 Ljubljana, Slovenia d University of Insbruck, Institute of Microbiology, Technikerstrasse 25 d, 6020 Innsbruck, Austria b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 November 2016 Received in revised form 10 March 2017 Accepted 12 March 2017 Available online 16 March 2017
The effects of five commercially available hydrolytic enzyme additives on methane yields from dehydrated paper pulp sludge (DPPS) were determined in 5L pilot-scale reactors operated in semi-continuous mode for 60 days. Methane production was 40% and 43% higher in reactors receiving Novozymes and Novalin additives, respectively, compared to controls. Effects of time of DPPS inclusion on bacterial and archaeal microbial communities were many times larger than effects of enzyme type as enzyme addition did not produce rearrangements larger than random fluctuations observed in reactors receiving only DPPS. The ratio between volatile organic acids and alkalinity signified progressive decrease in process stability until day 45 irrespective of enzyme supplementation. Complementation with clarified pig slurry (1.5% vol.) for subsequent 15 days effectively stabilized process parameters and was sufficient for microbial communities to maintain DPPS hydrolytic capacity and process additional carbon flow derived from hydrolytic activity of enzyme additives. Consequently, initially unadapted full-scale biogas plant inoculum was capable of significantly increased methane yields from DPPS. Based on annual DPPS availability in EU the potential for additional energy recovery was estimated to be in the range of nearly 1 TJ. © 2017 Published by Elsevier Ltd.
Keywords: Anaerobic digestion Paper pulp sludge Methane yield Hydrolysis Enzyme addition Semi-continuous Co-substrate
1. Introduction Dehydrated paper pulp sludge (DPPS) represents a solid waste material composed of pulp residues and ash generated from the pulping and paper-making process [1,2]. It is a solid residue recovered from the wastewater and consists primarily of complex plant polysaccharide materials such as cellulose, lignin and hemicelluloses [3]. The effluents from the pulp and paper industry cause considerable damage to the receiving waters if discharged untreated and are characterized by high biochemical oxygen demand (BOD), chemical oxygen demand (COD) and contain chlorinated compounds, suspended solids (i.e. fibers), fatty acids, lignin and
* Corresponding author. University of Ljubljana, Biotechnical Faculty, Department of Animal Science, Group for Microbiology and Microbial Biotechnology, Jamnikarjeva 101, 1000 Ljubljana, Slovenia. E-mail address: [email protected] (B. Stres). http://dx.doi.org/10.1016/j.energy.2017.03.050 0360-5442/© 2017 Published by Elsevier Ltd.
sulphur compounds. Most of the solids are removed after the mechanical treatment resulting in a DPPS that contains large quantities of fibers [2,4]. In general, most of DPPS (69%) is landfilled or incinerated (21%) and as such represents measurable financial burden to the industry: 50.000 tons of paper mill sludge are being produced and disposed at the average cost of transportation and landfill disposal of 75V per tonne by paper industry in Slovenia annually [1]. In addition to its costly disposal for the paper mills its energy potential and nutrient content are lost [5]. Another option for at least partial valorization of these resources is inclusion of DPPS into existing agricultural biogas plant anaerobic digestion (AD) located nearby. Increased methane yields can be attained using enzymatic pre-treatments including the reduction of organic matter, however, without significant installation and construction costs related to enzyme utilization [5,6]. Microorganisms that are present in the anaerobic reactor secrete enzymes to help them enhance the degradation of cellulose, hemicellulose, lipids,
S. Kolbl et al. / Energy 126 (2017) 326e334
proteins, starch and other complex plant polymers. Hydrolysis is nevertheless the rate limiting step in the anaerobic digestion [7,8] hence enzymatic pretreatments are introduced to increase the extent of organic matter depolymerization and improved methane yields from anaerobic digestion. Economical feasibility of these approaches depends on performance of enzyme types under conditions close to those utilized at full scale and additive price. So far, the effects of short-term (60 days) additions of various enzyme additives on DPPS methane yields, process parameters and microbial communities of bacteria and archaea in agricultural biogas plant inoculum have not been assessed comparatively. Agricultural biogas plants encounter substrate shortage due to market fluctuations, differences in annual cycles or unanticipated weather events. The short-term addition of DPPS as a sole substrate was tested in this study as an approach to bridge the time gap between the availability of regular agricultural substrates. In addition, enzyme additives and pig slurry complementation were tested as mitigation strategies to increase methane yields from DPPS and stabilize anaerobic digestion of DPPS. In this study (i) five different enzyme additives available on market were used to increase methane yields from DPPS over 60 days; (ii) methane yields were determined in 5 L semi-continuous reactors and a number of targeted process parameters (n ¼ 16) were monitored in all variants over the same period; (iii) rearrangements in bacterial and archaeal community structure were monitored in order to delineate the contribution of stochastic changes over time relative to those potentially related to the type of enzymatic supplementation; (iv) based on annual DPPS availability in EU the potential for additional energy recovery was estimated and sustainability of short-term DPPS inclusion into 1 MW agricultural biogas plant assessed.
2. Materials and methods 2.1. Baseline methane yields from paper pulp Methane yields relevant for DPPS were obtained from Methane yield Database [9]. Average, quartiles, median, 95% confidence intervals were calculated for paper sludge (n ¼ 38 data points). Additional entries (n ¼ 16) collected from Kinnunen et al. [10] and Meyer & Edwards [11] were submitted in order to complement the data on methane yields from DPPS and are hereafter part of the freely available Methane yield Database [7]. In addition, the experimental methane yields determined in this study were submitted to Methane Yield Database [7] under the accession numbers mdb2215-mdb2261. Theoretical methane yield of dehydrated paper pulp sludge (DPPS) was calculated using Bushwell's equation [12] assuming that most of the organically degradable part of the paper sludge was composed of cellulose (C6H10O5) with an ash content of 22% and 30% for DPPS1 and DPPS2, respectively (Table 1).
Table 1 Characteristics of substrates and inoculum used in assays. BDL- below detection limit. Parameter
Swine slurry
DPPS 1
DPPS 2
Inoculum
TS (%) VS (% TS) VS (% FM) pH VOA (mg l1) TIC (mg l1) VOA TAC1 Ntotal (mg l1) 1 NHþ 4 -N (mg l )
0.33 ± 0.01 41.03 ± 1.19 0.13 ± 0.01 7.24 ± 0 1403 ± 61 2232 ± 55 0.69 ± 0.38 2500 ± 25 300 ± 12
16.99 ± 0.02 78.37 ± 0.27 13.31 ± 0.01 6.60 ± 0.01 3549 ± 494 15087 ± 2957 0.237 ± 0.01 1055 ± 35 BDL
17.64 ± 0.02 69.83 ± 0.12 12.32 ± 0.09 6.54 ± 0.01 3212 ± 359 14932 ± 3247 0.215 ± 0.02 905 ± 40 BDL
8.53 ± 0.18 69.17 ± 0.60 5.90 ± 0.16 7.87 ± 0.01 5534 ± 247 13275 ± 175 0.417 ± 0.01 5230 ± 60 1301 ± 39
327
2.2. Semi-continuous experiments 2.2.1. Inoculum and paper pulp sludge Inoculum was collected from a mesophilic agricultural biogas plant (ABGP) Sijanec (Ormo z, Slovenia; 1 MW) following the established procedure described before [6]. Two distinct batches of DPPS, (batch one - DPPS 1; batch two - DPPS 2) were collected in the period of 1 month from 2000 PE (population equivalents) paper pulp wastewater treatment plant located at the site of the hygienic paper manufacturer Paloma d.d. (Sladki vrh, Slovenia; http://web. paloma.si/en/). DPPS samples were stored at 2 C and homogenized by mixing before use in experiments. Hygienic papers of Paloma are produced from pure cellulose, waste paper and board. Flotation and flocculation are used to separate paper pulp from wastewater and further dehydrated by rotating drums to form DPPS (130e210 t daily; 40% TS). Different types of paper that are produced within Paloma d.d. rely on distinct technological approaches and hence yield DPPS batches with distinct characteristics and theoretical methane yields. Characteristics of inoculum and all substrates are listed in Tables 1 and 2. In order to complement the semi-continuous reactors with sources of trace metals and nutrients during the observed phase of process instability (d46-60; n ¼ 14 days) clarified pig slurry (PS) [13], i.e. slurry devoid of particulate material by prolonged settling was collected from a nearby (d < 20 km) pig farm Vucja vas, Slovenia (capacity ¼ 250 pigs). 2.2.2. Residual degradation of inoculum by five enzyme brands To assess whether the existing inoculum contained a fraction of organic matter amenable to further enzymatic degradation by provided additives, five brands of commercially available products were tested (Table S1) in the first experiment. Two Automatic Methane Potential Test unit (AMPTS II, Bioprocess Control, Sweden) upgraded to 5 L scale were used in semi-continuous experiments as detailed before [6]. All anaerobic reactors (n ¼ 18) received 4000 mL inoculum from BGP Sijanec. Negative controls represented anaerobic reactors (n ¼ 3) that received tap water, while all the others received enzymatic amendments: hydrolytic enzymes Novozymes (Denmark) (n ¼ 3), hydrolytic enzymes Novalin (NovaBiotec, Germany) (n ¼ 3), Micropan Biogas additive (Eurovix, USA) (n ¼ 3), additive BFL 4400AN (Biofuture, Ireland) (n ¼ 3) and Zeolit M (Ipus, Austria) (n ¼ 3). Daily preparations and additions of enzyme additives into anaerobic digesters were conducted essentially as described before [6]. Before the start up, each reactor was flushed with N2 for 3 min. Mixing (15 min on-off regime) continued throughout the operation of reactors and temperature in reactor water baths was maintained at 38 ± 2 C. 2.2.3. Testing the efficiency of five enzyme brands In the second experiment five commercially available additives of different brands were amended to the anaerobic reactors on daily basis: hydrolytic enzymes Novozymes (Denmark) (n ¼ 3), hydrolytic enzymes Novalin (NovaBiotec, Germany) (n ¼ 3), Table 2 Chemical analysis of total metal content in both batches of DPPS substrate introduced to 5l semicontinuous reactors in this study (mg kg1 TS1). Metals
DPPS1
DPPS2
Cd Cr Cu Hg Ni Pb
0.5 10.2 59.8 0.003 3 12
0.7 18.4 63.4 0.040 9.8 12.8
328
S. Kolbl et al. / Energy 126 (2017) 326e334
Micropan Biogas additive (Eurovix, USA) (n ¼ 2), additive BFL 4400AN (Biofuture, Ireland) (n ¼ 2) and Zeolit M (Ipus, Austria) (n ¼ 2). Dosage of each additive was in accordance with manufacturer's guidelines (Table S1) following the full-scale anaerobic digestion guidelines depending on the organic loading and volatile solids or total solid content of substrate. In total, 12 anaerobic digesters were supplied with additives (Table S2), whereas the rest served as controls. Organic loading rate (OLR) at the full scale reactor of ABGP Sijanec from which inoculum was obtained for testing was 10 g VS L1 day1, five times higher than the suggested starting 2.2 g VS L1 day1 for DPPS [5,14]. The three sections (DPPS1, DPPS2 and DPPS2þPS) of 60 day semi-continuous experiment (Fig. S1) were projected to take place at full scale installation [14] in order to minimize operating costs. The 5 L reactors (n ¼ 15) of AMPTS II unit (AMPTS II, Bioprocess Control, Sweden) received 4000 mL inoculum and daily organic loading of 2.2 g VS L1 day1 equal to 66.1 g of DPPS1 or 71.4 g of DPPS2 with the addition of 183.4 mL and 179.6 mL of industrial tap water, respectively. Total working volume was 4250 mL. DPPS1 was dosed in period from day 0 to day 28, DPPS2 from day 29e45 and a mixture of DPPS2 and cleared pig slurry (PS) (60 mL day1) and water (119.6 mL day1) from day 46 to day 60 (DPPS2þPS; Table S2, Fig. S1). In order to retain mass balance equal amounts of material and liquids were introduced and removed on daily basis as described before [11]. This collected material was used for subsequent analyses of process parameters. In short, the collected samples were centrifuged for 3 min at 5000 g and supernatant was stored at 20 C until further analyses of process parameters. The remaining solid fraction was homogenized, subsampled (0.5 g) at 4 C and frozen at 20 C until DNA extraction. In addition to five-fold reduction in OLR, hydraulic retention time (HRT) used at ABGP Sijanec was shortened from original 60 dayse18 days in semi-continuous experiments as suggested before [15] for optimal performance in degradation of DPPS. Mixing (15 min on-off regime) continued throughout the operation of reactors and temperature in reactor water baths was maintained at 38 ± 2 C. Process parameters (n ¼ 16), total solids, volatile solids, C1-C6 short chain fatty acids (SCFA), pH, total inorganic carbon (TIC), volatile organic acids (VOA), the total VOA/TIC ratio, total nitrogen, ammonia, electrical conductivity, particle size distribution, and soluble chemical oxygen demand (sCOD) were determined from samples collected every 5 days. 2.2.4. Rearrangements in bacterial and archaeal microbial communities Power Fecal DNA Extraction kit (MOBIO, USA) was used to extract DNA from samples stored at 20 C. DNA quality and purity (A230, A260, A280) were assessed spectrophotometrically using NanoVue according to manufacturer's instructions. Amplification of the bacterial and archaeal genes for 16S rRNA was conducted as described before [16] according to PCR protocol: 94 C for 5 min, followed by 30 cycles of denaturation at 94 C for 30 s, annealing at 54 C for 45 s, and elongation at 72 C for 90 s. The final extension was at 72 C for 7 min. The PCR reaction mixture (25 mL) contained 2.5 mL 10 PCR buffer, 2.5 mM MgCl2, 200 mM of each nucleotide, 200 nM of each primer, 1.0 U DNA polymerase, 2e20 mg BSA/L, and 1.0 mL of undiluted template. Both forward primers used in fingerprinting of bacterial and archaeal microbial communities were marked with JOE (6-carboxy-40 , 5’-dichloro-20 , 7'-dimethoxy- fluorescein) and amplicons were separated using ABI3100 sequencer. The resulting electropherograms were converted to text formatted data in ABI3100 sequencer proprietary GENEMAPPER software and analyzed in R as described before [16].
Incubation time in experiment and hydrolytic enzyme additive brand were used as explanatory variables in two-way NP-MANOVA in PAST [17] and variance partitioning as described before [18,19] to assess their contribution to changes in bacterial and archaeal microbial communities. 2.2.5. Physical and chemical analyses Determination of total solids (TS), volatile solids (VS) and soluble chemical oxygen demand (sCOD) was made according to the APHA [20]. For measurements of volatile organic acids (VOA), total inorganic carbon (TIC) and ratio VOA/TIC Biogas Titrator Manager TIM 840 (Hach Lange) was used. Total nitrogen and ammonia was determined by SANplus Continuous Flow Analyzer (Skalar Analytical B.V., Netherland) according to the manufacturer's instructions [6]. Electrical conductivity and pH were measured using multimeter HQ40D (Hach Lange). Particle size analysis was performed by Laser Particle Sizer Analysette 22 Wet Dispersion Unit (Fritsch, Germany) as described before [11] using measuring range from 0.08 to 2000 mm. Particle size was calculated through laser diffraction and Fraunhofer calculation method that is incorporated in software FRITSCH Mass Control [14]. Measurements of total Cd, Cr, Cu, Hg, Ni and Pb in DPPS were provided by ERICO Velenje according to the ISO 17294-2:2003 [21]. Ether extraction was used to prepare short chain fatty acids (SCFA) for analyses as described before [6,22] and their concentrations determined by gas chromatograph Agilent 7890. Capillary column Agilent J & W GC columns DB-FFAP, 30 m 0.530 mm x 1 mm layer of stationary phase was used. The injector and detector temperatures were 200 C and 300 C, respectively. Initial oven temperature was 70 C with residence time of 1 min, then ramped at 20 C min1 to 120 C and further 10 C min1 to final oven temperature of 200 C with residence time of 3 min. Carrier gas (mobile phase) was Helium with flow 5 mL min1, nitrogen flow 25 mL min1, detector gas was hydrogen with flow 30 mL min1, synthetic air with flow 400 mL min1. 2.3. EU wide methane potential of DPPS Annual amounts of paper sludge produced in EU were calculated based on yearly paper production assuming 0.3 m3 of DPPS produced per 1000 kg of paper [23] (http://www.cepi.org/node/ 19363). In addition, Chamber of Commerce and Industry and Statistical Office of the Republic of Slovenia provided the data on annual amounts of DPPS produced in Slovenia (http://www.stat.si/ StatWeb/en/home; https://www.gzs.si/zdruzenje_za_papirno_in_ papirno_predelovalno_industrijo). First, initial parameters for determination of methane yields from DPPS and potential energy gains were estimated from (i) the data on annual production of DPPS, (ii) the DPPS organic loading in anaerobic full scale digesters [6], and (iii) the extent of increased methane yields from DPPS as a result of enzyme addition to DPPS in semi-continuous experiments. Second, total methane yields obtained from semi-continuous experiments in this study were further distinguished into three values, corresponding to: (i) the additional methane produced from the previously undigested substrate in inoculum amended solely with various enzyme brands; (ii) the methane produced as a result of increased DPPS digestion due to hydrolytic enzyme efficiency; (iii) methane produced as a result of complementation with clarified pig slurry. Third, fluctuations in production of different types of paper within Paloma d.d. necessitate distinct technological approaches. Hence two batches of DPPS were used to estimate methane yields
S. Kolbl et al. / Energy 126 (2017) 326e334
in this study and were contrasted with previously recorded data in Methane Yield Database [9]. These collated data were used to estimate the mean, lower (25%) and upper (75%) quartiles of the short-term (60 day) economic sustainability of enzyme additive purchase for the ABGP Sijanec from which inoculum was obtained using the approach described before [6,7,11] and taking into account on-site specific parameters of particular biogas plant (ABGP Sijanec, Slovenia; internal data, unpublished [6,11]). 3. Results and discussion 3.1. Residual methane yields from inoculum The semi-continuous analysis of additional methane formation obtained from inoculum amended with enzyme additives showed that most of enzyme additives significantly increased methane yields from inoculum used in this study (Fig. S2A). Methane production from enzyme only amended reactors effectively ceased after 35 days, whereas most of the process parameters remained unaffected, irrespective of enzyme brands (Figs. S2BeD). The largest surplus of residual methane was observed for Novozymes (27%), Micropan Biogas (11%) and Novalin (4%), whereas no significant activity or even shortage were observed for Biofuture (2%) and Zeolit M (16%), respectively. Biofuture contained microbial cultures and hence operated on the lines of bioaugmentation, whereas Zeolit M relied on silicates with specific nutrient allocations and specific absorption capabilities (Table S1). Growth of novel bioaugmenting microorganisms from Biofuture additive was apparently not favored under the conditions of depleting substrate present in inoculum, whereas silicates with specific absorption capabilities most probably immobilized decreasing amounts of mineral nutrients and hence did not produce significant positive effects on existing microorganisms. The short-term (i.e. 40 day) inclusion of specific enzyme additives into 5 L semi-continuously operated reactors thus resulted in distinct residual methane outputs from the same batch of ABGP Sijanec inoculum. As previous estimates of background methane formation after enzyme addition were based on batch scale experiments [24] the results of this study represent the first estimate of proportions of yet undigested substrate in inoculum. The benefits of enzyme introduction into full-scale biogas plant are apparently substantial and could support at least the short-term campaigns. 3.2. Enzymatically increased methane yields from DPPS The comparison between the methane yields obtained from various sections of 60 day semi-continuous operation (Fig. 1AeC) of DPPS anaerobic digestion illustrates the profound differences between the studied variants after the initial ABGP substrate supply was discontinued and DPPS introduced. Particle size of DPPS was below 2 mm [6] (Fig. S3A) making DPPS particles of sufficient small size for anaerobic digestion, avoiding thus the additional particle size homogenization. The three sections (DPPS1, DPPS2 and DPPS2þPS) tested in 60 day semi-continuous experiment (Fig. S1) were projected to take place at full scale installation [14]. The sole transition from standard ABGP substrate mix to DPPS was accompanied by fivefold decrease in loading rate and consequent characteristic decrease in methane production (Fig. 1B). During the addition of DPPS1, controls exhibited average daily methane production (day 8 to day 28) of 134.6 ± 12.7 mL CH4 g1 VS1 day1, whereas variants with additives Novozymes, Novalin, Micropan Biogas, Biofuture and zeolite M all exhibited positive effects on methane production: 188.7 ± 12, 195.1 ± 10.9, 158.7 ± 13,
329
156.1 ± 7.5 and 147.7 ± 13 mL CH4 g1 VS1 day1, respectively. The onset of DPPS2 addition resulted in significantly lower average methane daily production from control reactors in period from day 29 to day 45 (82.4 ± 10.7 mL CH4 g1 VS1 day1) compared to DPPS1 (Fig. S3B), however, a decrease in methane production from Novalin, Novozymes, Micropan Biogas, Biofuture and zeolite M: 133.9 ± 12.5, 128.6 ± 13.5, 103.8 ± 7.7, 102.3 ± 10.1 and 103.3 ± 11.3 mL CH4 g1 VS1 day1, respectively, was also observed. In essence, control reactors using both batches of DPPS effectively produced methane yields in the ranges reported before [5]), however, enzyme additive supplementation was substantially more effective. Addition of pig slurry to DPPS2 in semi-continuous experiment (day 46 to day 60) was conducted in attempt to stabilize anaerobic digestion and prevent further decline in methane yields. No significant change in average daily methane production from DPPS2þPS (84.8 ± 16.3 mL CH4 g1 VS1 day1) and enzyme supplemented variants with different additives Novozymes, Novalin, Biofuture, Micropan Biogas and zeolite M was observed: 130 ± 21.2 mL, 127.6 ± 19.7, 103.3 ± 14.3, 105.5 ± 11.7 and 107 ± 10.1 mL CH4 g1 VS1 day1, respectively. Nevertheless, the addition of pig slurry as co-substrate prevented further trend of declining of methane yields observed from day 29 to day 45 for DPPS2 (Fig. 1B). The use of various experimental models and scales in methane yield estimation (e.g. batch vs. semi-continuous; e.g. 300 mL vs. 4 L) was shown to exhibit substantial effects on estimates of full-scale sustainability [6,7,11] used in industrial settings for decisionmaking. In line with these observations, up to 30% overestimates in methane yields were observed for batch experiments in comparison to semi-continuous due to the effects of differences in scale and assembly [6,25]. For example, methane yields from 160 mL batch experiments treating biosludge from paper industry was up to 30% higher compared to 6 L and 400 day semi-continuous experiment [10]. Higher methane yields can be obtained in batch assays due to prolonged (e.g. 30 day) anaerobic digestion that is long enough for the utilization of almost all degradable substrates. However, the obtained methane yields in semi-continuous assays are determined by the relationship between HRT and degradability of the substrate. This distinction is important for the industrial decision makers to selectively base their sustainability estimates on semi-continuous results rather than batch test data to avoid overestimation of methane yields. In this study, methane yields from batch DPPS2 degradation were 32.5% higher than those obtained from semi-continuous digestion of DPPS2 (Fig. S3B). Complementation DPPS2þPS resulted in three times higher methane yields in 5 L batch experiments (Fig. S3B), illustrating the extent of influence on biogas plant decision-making support. In addition, a wide range of methane yields were reported for DPPS in past experiments ranging from 59 to 370 mL CH4 g1 VS1 [9e11,26]. In this respect, to obtain methane yields relevant for industrial process planning the use of experimental procedures mirroring conditions utilized at full scale as close as possible is necessary [6,9]. Previous studies suggest that deficiencies in micronutrients can lead to diminished performance of anaerobic digestion [8,27]. Therefore aliquots of clarified pig slurry (PS; 1.5% vol.) were amended as cosubstrate along with DPPS2 from day 46 to day 60 (DPPS2þPS; Table S2) to increase methane yields and to test whether some unaccounted for micronutrient limitations (e.g. other metals like Mn, Mo, Se, Zn) or depleted macronutrients (N) or process parameters (alkalinity) were responsible for low methane production irrespective of enzyme addition. The addition of PS as cosubstrate to DPPS reactors (day 46e60) restored the positive effects of enzyme addition (Fig. 1B) showing that optimal concentrations of additional nutrients in the environment were necessary
330
S. Kolbl et al. / Energy 126 (2017) 326e334
Fig. 1. Normalized cumulative methane production from 60 day semi-continuous. anaerobic digestion of DPPS1, DPPS2 and DPPS2þPS. Methane production was normalized to control reactors receiving DPPS only (A). (B) Daily methane production with standard deviations. (C) Cumulative methane production in the same experiments. DPPS1, DPPS2, DPPS þ PS were administered to all anaerobic reactors from day 1e28, day 29e45 and day 46e60, respectively.
for the microbial communities to either maintain their own hydrolytic activities and/or to process the additional carbon flow derived from hydrolytic action of introduced enzyme additives. Additional batch experiment complementing DPPS2 with PS (Fig. S4) was in line with this observation. This puts the nutrient deficit in DPPS observed before [5] and lack of microbial hydrolytic enzyme activity (this study) into new perspective. Although DPPS1 and DPPS2 also contained microbial biomass in addition to existing inoculum, microbial activity was apparently limited due to the increasing nutrient depletion over time (days 1e45) and hence did not actively contribute to the degradation of primary sludge derived fibers consisting more or less of cellulose [15,28]. Secondly, it also shows that nutrient limitation has the potential to diminish the positive effects of introduced hydrolytic enzyme additives. Consequently, a simple complementation adopting nutrient rich co-substrates for optimal performance has the potential to support higher methane yields derived from DPPS on full scale, possibly also at higher OLR and in long-term applications in the future (t > 60 days). The compatibility of the two substrates and short-term enzyme addition to 5 L reactors operated in semi-continuous mode enables the adoption of DPPS as an additional substrate for ABGP treating clarified pig slurries containing excess nutrients. In addition, this
enables paper industry to complement its industrial biogas reactors with additional substrates from nearby agricultural practices in order to maximize methane yields from DPPS while minimizing the remaining sludge volume. From the perspective of ABGP the short term strategy of 60 day semi-continuous DPPS degradation with enzyme additives resulted in cumulatively 40% and 43% higher methane production in reactors supplemented with Novozymes and Novalin additives, respectively compared to control reactors receiving daily DPPS only. The obtained methane yields were in comparable range as reported before for semi-continuous degradation of untreated activated sludge from Swedish pulp and paper mills [5]. However, daily introduction of novel substrate in semi-continuous operation includes also daily removal of a minor portion of preexisting inoculum and previously supplied DPPS substrate. In our case this represented daily removal of 250 g aliquot of mixed DPPS, water and inoculum from total operating volume of 4250 mL, a ratio similar to ABGP full-scale operations. Further, the major part of DPPS degradation (>90%) was completed under batch conditions in 6 days (Fig. S4A). Consequently, only minute portions of fresh DPPS could be removed due to reactor operation procedures in consecutive days and increased methane production was derived from the digestion of a major part of past additions and newly added DPPS.
S. Kolbl et al. / Energy 126 (2017) 326e334
Our results also show that different enzyme additives can exhibit markedly different efficiencies, most probably due to the distinct inoculum - paper sludge combinations and hence deserve further study. Consequently, a prescreening for selection of most promising enzyme additives suitable for local inoculum and operating conditions can easily be performed on-site in real-time as described in this study. 3.3. Process parameters during the addition of DPPS Several important process parameters changed during the shift from general ABGP substrate mix [6] to DPPS as sole substrate at lower OLR and HRT (Fig. 2AeG): pH, total nitrogen, ammonia, TIC, conductivity and sCOD. The latter two kept decreasing throughout the experiments, however the pH and TIC increased sufficiently only after pig slurry addition, that helped stabilize the indicative VOA/TIC ratio. This VOA/TIC ratio was used as parameter describing process instability [11] and showed that instability was not due to accumulation of VOA but due to the observed drop in alkalinity [15,29] accompanied by decrease in electrical conductivity of anaerobic digesters (Fig. 2E). However, no significant differences in electrical conductivity or nitrogen content between reactors could be detected. In the case of anaerobic digestion of biological wastes reported by Scano et al. [30] VOA/TIC ratio reached up to 0.65, however pH did not drop below 7. Sufficient alkalinity for stable anaerobic digestion of sludge was generally reported to be between 1000 and 5000 mg CaCO3 l1 [29], however up to 10,000 mg CaCO3 l1 were reported for stable processes treating biological wastes [31]. Significant drop in soluble chemical oxygen demand (sCOD) after transition to DPPS (Fig. 2F) reflected the polymer nature of DPPS and its effects on subsequent hydrolysis and fermentation steps. Complementation of DPPS with PS or other clarified slurries or manures has thus the potential to sustain the use of higher OLR of DPPS in the full-scale reactors, increased alkalinity and increased methane production. 3.4. Fluctuations in microbial community structure of bacteria and archaea In this study, changes in bacterial and archaeal community structure were assessed using fast automated fingerprinting approaches in order to identify the extent of change in response to the time of incubation and the type of amended enzyme additives. Bacterial and archaeal communities exhibited a gradual change over the course of DPPS administration, in line with past observations that flexible community structure correlates with stable community function in methanogenic bioreactors [32] sustaining changes in carbon and electron flow [33]. The transition from agricultural substrate to DPPS as the sole substrate introduced significant reorganization in bacterial and archaeal microbial communities in control reactors in 60 days. The results of two-way NPMANOVA (npermut ¼ 9999) for bacterial and archaeal communities, respectively, showed that the administration of various enzyme additives (pBacteria<0.001; pArchaea ¼ 0.025), and incubation time with DPPS in experiment (pBacteria ¼ 0.001; pArchaea < 0.0001) were significantly associated with the changes in microbial communities, however, their interaction was surprisingly not significant (pBacteria>0.05; pArchaea > 0.05). This suggests that the effects of enzyme addition did not propagate through time under the conditions used in this study. Variation partitioning showed that the effect of incubation time with DPPS exerted on bacterial and archaeal communities was 5 and 6 times larger, respectively, than the effects of 60-day supplementation of the same reactors with different enzyme additives.
331
This shows that the extent of fluctuations in microbial communities in response to short-term enzyme additions were many times lower and well within the range of random fluctuations observed in control reactors receiving only DPPS. Hence, DPPS induced a general adaptation response of bacterial and archaeal microbial communities, whereas addition of enzymes did not induce significant enzyme-specific adaptations of microbial communities. In addition, the extent of rearrangements in microbial communities observed over time of inclusion of DPPS was 11 times larger for bacteria than for archaea in enzyme amended and control reactors. This further illustrates that the fluctuations in microbial communities were more pronounced at the level of bacterial microbial communities responsible for hydrolysis of the novel substrate than at the level of archaea producing methane from the resulting organic acids. Taken together, the short-term adoption of various enzyme additives in anaerobic digestion did not result in extensive enzyme-related rearrangements in microbial communities, and hence could be applied in industrial settings at least under comparable OLR of DPPS in the future. Our results are in line with previous findings that structurally responsive communities were also more functionally stable [34]. The exact rearrangements in microbial functional genes (after the introduction of DPPS as a novel substrate) are currently being investigated using metagenomic approaches in order to identify microbial functional genes and taxonomic groups of bacteria and archaea linked to DPPS degradation in microbial communities before and after DPPS administration. 3.5. The EU wide energy potential of DPPS The EU wide data collection showed that paper and board production in EU reached 91.1Mt in 2014 [23] (http://www.cepi. org/node/20088). Based on the availability of 4.6 Mt of DPPS the following three scenarios of methane-energy conversion were explored (Table 3): (i) energy production from methane derived solely from DPPS (control) including residual methane produced from additionally degraded inoculum; (ii) energy production from methane derived from DPPS as sole substrate after the deduction of residual methane generated by enzymes supplied to existing inoculum (accounting for additional degradation of digestate within the same reactors due to enzyme addition); (iii) energy yields from most efficient digestion of DPPS complemented with nutrients from PS (Table 3). Based on data on DPPS availability and improved methane yields approximately 1 TJ of additional energy could be obtained from DPPS annually (Table 3). Substrate shift from ABGP substrates to DPPS in this study resulted in fivefold reduction in OLR and HRT. Consequently, threefold reduction in average methane yields was observed between day 0 in comparison to days 8e60 (Fig. 1B). However, the addition of preselected enzyme additives resulted in up to 40% higher methane yields from DPPS in comparison to controls. The use of most efficient enzyme additives resulted in 195.1 ± 10.9 mL CH4 g1 VS1 day1, approximately half of the initial methane production observed at ABGP (Fig. 1B; [6,11]). Results reported in this study were indeed obtained during a short-term 60-day semi-continuous test period intended to test the potential effectiveness of short-term DPPS introduction into fullscale agricultural biogas plant. The adopted minimum intervention approach to inoculum and substrate to be used at full scale provided a framework for real-time estimation of methane yields and approximation of process parameters relevant for full-scale biogas plant management [6,11]. The data obtained in this study were compared to published data organized within Methane Yield Database [7] in order to provide realistic estimates on economic feasibility of DPPS inclusion into full-scale biogas plants including
332
S. Kolbl et al. / Energy 126 (2017) 326e334
Fig. 2. Overview of most important anaerobic process parameters during 60 day semi-continuous anaerobic digestion of DPPS1, DPPS2 and DPPS2þPS: (A) pH of reactor mixtures; (B) total nitrogen; (C) total inorganic carbon (TIC, i. e. alkalinity); (D) anaerobic process stability assessed through VOA/TIC ratio; (E) electrical conductivity as measure of ion salt concentration; (F) sCOD; (G) ammonium concentrations.
also on-site biogas plant specific sustainability parameters (ABGP Sijanec (unpublished internal data)). The short term inclusion of DPPS and its complementation with locally available pig slurry in combination with enzyme additives was accepted as viable short-
term alternative using the estimation approach described before [11], covering the expenses of enzyme purchasing, paper pulp and PS transport. Future experiments are directed toward increased OLR and HRT approaching those utilized at ABGP Sijanec during
S. Kolbl et al. / Energy 126 (2017) 326e334
333
Table 3 The potential for valorization of methane produced from DPPS in Slovenia and EU during the 60 day on-off introduction of various enzyme brands. Energy production (GJ year1)
a
Slovenia (DPPS) Slovenia (no residual methane)b Slovenia (addition of pig slurry)c EU (DPPS)a EU (no residual methane)b EU (addition of pig slurry)c
No enzymes
Novozymes
Novalin
Biofuture
M. Biogas
Zeolit M
6.9 5.0 7.0 630.4 460.2 645.5
10.1 7.4 10.3 926.4 676.3 948.7
10.3 9.9 10.5 944.9 907.1 967.6
8.1 8.3 8.3 744.1 759.0 762.0
8.2 9.6 8.4 758.7 880.1 776.9
8.0 7.1 8.2 734.8 654.0 752.4
a
Energy produced from methane generated during short-term addition of DPPS to agricultural biogas plant biomass. Energy produced from methane generated during short-term addition of DPPS to agricultural biogas plant biomass, after the deduction of residual methane produced from existing undigested residual substrate. c Energy produced from methane generated during short-term addition of DPPS to agricultural biogas plant biomass with clarified pig slurry as cosubstrate. b
regular operation with the final goal of DPPS inclusion into ABGP as long-term cosubstrate. 4. Conclusions Prescreening for most effective hydrolytic enzyme additives identified Novozymes and Novalin additives to increase methane production from DPPS for 40% and 43%, respectively. Daily supplementation of hydrolytic enzyme additives did not result in enzyme-specific rearrangements in bacterial and archaeal microbial communities. However, process instability observed in 45 days was bypassed by cost-effective complementation strategy with introduction of clarified pig slurry (1.5%). Data exploration on DPPS availability identified a substantial potential for annual energy production up to 1 TJ from enzymatically and co-substrate complemented DPPS in EU. Database linking Methane Yield Database: http://methane.fe.uni-lj.si/ Contributions SK and BS contributed equally. Performed the experiments: SK. Analyzed the data: SK, BS. Contributed reagents/chemicals/tools/ ideas: SK, PFT, BS. Wrote the paper: BS, SK. All authors agreed on the final version of the manuscript. Competing interests The authors declare no competing interests. Acknowledgments The authors acknowledge the support of Keter Invest energetika d.o.o. and Paloma d.d.. »Operation part financed by the European Union, European Social Fund. Operation implemented in the framework of the Operational Programme for Human Resources Development for the Period 2007e2013, Priority axis 1: Promoting entrepreneurship and adaptability, Main type of activity 1.1.: Experts and researchers for competitive enterprises.« SK acknowledges Programme "Water Science and Technology, and Geotechnic": P2-0180 (B). COST ACTION ES1302: European Network on Ecological Functions of Trace Metals in Anaerobic Biotechnologies is acknowledged for fruitful discussions and exchange of ideas during the final preparation of this manuscript.
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.energy.2017.03.050. References [1] Zule J, Cernec F, Likon M. Chemical properties and biodegradability of waste paper mill sludges to be used for landfill covering. Waste Manag Res 2007;25: 538e46. http://dx.doi.org/10.1177/0734242X07079188. [2] Mehdi Dashtban YH, Dashtban M, Chen S, Qin W, Song R. Lignin in paper mill sludge is degraded by white-rot fungi in submerged fermentation. J Microb Biochem Technol 2015:7. http://dx.doi.org/10.4172/1948-5948.1000201. [3] Jackson MJ, Line MA. Organic composition of a pulp and paper mill sludge determined by FTIR, 13C CP MAS NMR, and chemical extraction techniques. J Agric Food Chem 1997;45:2354e8. http://dx.doi.org/10.1021/jf960946l. [4] Bajpai P. Composition of waste. manag. pulp pap. mill waste. Cham: Springer International Publishing; 2015. p. 19e29. http://dx.doi.org/10.1007/978-3319-11788-1_3. [5] Karlsson A, Truong X, Gustavsson J, Svensson BH, Nilsson F, Ejlertsson J. Anaerobic treatment of activated sludge from Swedish pulp and paper mills e biogas production potential and limitations. Environ Technol 2011;32: 1559e71. http://dx.doi.org/10.1080/09593330.2010.543932. [6] Kolbl S, Paloczi A, Panjan J, Stres B. Addressing case specific biogas plant tasks: industry oriented methane yields derived from 5L Automatic Methane Potential Test Systems in batch or semi-continuous tests using realistic inocula, substrate particle sizes and organic loading. Bioresour Technol 2014. http:// dx.doi.org/10.1016/j.biortech.2013.12.010. [7] Batstone DJ, Jensen PD. Treatise on water science. Elsevier; 2011. http:// dx.doi.org/10.1016/B978-0-444-53199-5.00097-X. [8] Romero-Güiza MS, Vila J, Mata-Alvarez J, Chimenos JM, Astals S. The role of additives on anaerobic digestion: a review. Renew Sustain Energy Rev 2016;58:1486e99. http://dx.doi.org/10.1016/j.rser.2015.12.094. [9] Murovec B, Kolbl S, Stres B. Methane Yield Database: online infrastructure and bioresource for methane yield data and related metadata. Bioresour Technol 2015. http://dx.doi.org/10.1016/j.biortech.2015.04.021. €-Outinen A, Rintala J. Mesophilic anaerobic digestion of pulp [10] Kinnunen V, Yla and paper industry biosludgeelong-term reactor performance and effects of thermal pretreatment. Water Res 2015;87:105e11. http://dx.doi.org/10.1016/ j.watres.2015.08.053. [11] Meyer T, Edwards EA. Anaerobic digestion of pulp and paper mill wastewater and sludge. Water Res 2014;65:321e49. http://dx.doi.org/10.1016/ j.watres.2014.07.022. [12] Deublein D, Steinhauser A. Biochemistry. biogas from waste renew. resour.. Weinheim, Germany: Wiley-VCH Verlag GmbH & Co. KGaA; 2010. p. 97e100. http://dx.doi.org/10.1002/9783527632794.ch9. [13] Piccinini S, Cortellini L, Bonazzi G. Separation of pig slurry by sedimentation. In: Van Der Meer HG, Unwin RJ, Van Dijk TA, Ennik GC, editors. Anim. Manure grassl. Fodd. Crop. Fertil. or Waste? Proc. an int. Symp. Eur. Grassl. Fed. Wageningen, Netherlands, 31 Auguste3 Sept. 1987. Dordrecht: Springer Netherlands; 1987. p. 385e8. http://dx.doi.org/10.1007/978-94-009-36591_45. [14] Kolbl S, Panjan J, Stres B. Mixture of primary and secondary municipal wastewater sludge as a short-term substrate in 2 MW agricultural biogas plant: site-specific sustainability of enzymatic and ultrasound pretreatments n.d. doi:10.1002/jctb.4883. [15] Bayr S, Rintala J. Thermophilic anaerobic digestion of pulp and paper mill primary sludge and co-digestion of primary and secondary sludge. Water Res 2012;46:4713e20. http://dx.doi.org/10.1016/j.watres.2012.06.033. [16] Franke-Whittle IH, Manici LM, Insam H, Stres B. Rhizosphere bacteria and fungi associated with plant growth in soils of three replanted apple orchards. Plant Soil 2015;395:317e33. http://dx.doi.org/10.1007/s11104-015-2562-x. [17] Hammer Ø, Harper Dat a T, Ryan PD. PAST: paleontological statistics software
334
[18]
[19]
[20] [21] [22]
[23] [24]
[25]
[26]
S. Kolbl et al. / Energy 126 (2017) 326e334 package for education and data analysis. Palaeontol Electron 2001;4(1):1e9. http://dx.doi.org/10.1016/j.bcp.2008.05.025. Stres B, Philippot L, Faganeli J, Tiedje JM. Frequent freeze-thaw cycles yield diminished yet resistant and responsive microbial communities in two temperate soils: a laboratory experiment. FEMS Microbiol Ecol 2010;74: 323e35. http://dx.doi.org/10.1111/j.1574-6941.2010.00951.x. Stres B, Sul WJ, Murovec B, Tiedje JM. Recently deglaciated high-altitude soils of the Himalaya: diverse environments, heterogenous bacterial communities and long-range dust inputs from the upper troposphere. PLoS One 2013;8: e76440. http://dx.doi.org/10.1371/journal.pone.0076440. APHA. Standard methods for the examination of water and wastewater. Washington, USA: American Water Works Assn; 2005. ISO 17294-2. Water quality e application of inductively coupled plasma mass spectrometry (ICP-MS) e Part 2: determination of 62 elements. 2003. Novak D, Franke-Whittle IH, Pirc ET, Jerman V, Insam H, Logar RM, et al. Biotic and abiotic processes contribute to successful anaerobic degradation of cyanide by UASB reactor biomass treating brewery waste water. Water Res 2013;47:3644e53. http://dx.doi.org/10.1016/j.watres.2013.04.027. CEPI. Preliminary statistics. 2013. 2012e3, http://www.cepi.org/node/20088. ndez-Cegrí V, Borja R. Anaerobic digestion of Raposo F, De la Rubia MA, Ferna solid organic substrates in batch mode: an overview relating to methane yields and experimental procedures. Renew Sustain Energy Rev 2012;16: 861e77. http://dx.doi.org/10.1016/j.rser.2011.09.008. Ruffino B, Fiore S, Roati C, Campo G, Novarino D, Zanetti M. Scale effect of anaerobic digestion tests in fed-batch and semi-continuous mode for the technical and economic feasibility of a full scale digester. Bioresour Technol 2015;182:302e13. http://dx.doi.org/10.1016/j.biortech.2015.02.021. Steffen F, Requejo A, Ewald C, Janzon R, Saake B. Anaerobic digestion of fines from recovered paper processing e influence of fiber source, lignin and ash content on biogas potential. Bioresour Technol 2016;200:506e13. http:// dx.doi.org/10.1016/j.biortech.2015.10.014.
[27] Voelklein MA, O’ Shea R, Jacob A, Murphy JD. Role of trace elements in single and two-stage digestion of food waste at high organic loading rates. Energy 2017;121:185e92. http://dx.doi.org/10.1016/j.energy.2017.01.009. ve J, Dewil R. Principles and potential of the [28] Appels L, Baeyens J, Degre anaerobic digestion of waste-activated sludge. Prog Energy Combust Sci 2008;34:755e81. http://dx.doi.org/10.1016/j.pecs.2008.06.002. [29] Lin Y, Wang D, Wu S, Wang C. Alkali pretreatment enhances biogas production in the anaerobic digestion of pulp and paper sludge. J Hazard Mater 2009;170:366e73. http://dx.doi.org/10.1016/j.jhazmat.2009.04.086. [30] Scano EA, Asquer C, Pistis A, Ortu L, Demontis V, Cocco D. Biogas from anaerobic digestion of fruit and vegetable wastes: experimental results on pilot-scale and preliminary performance evaluation of a full-scale power plant. Energy Convers Manag 2014;77:22e30. http://dx.doi.org/10.1016/ j.enconman.2013.09.004. [31] Banks CJ, Chesshire M, Heaven S, Arnold R. Anaerobic digestion of sourcesegregated domestic food waste: performance assessment by mass and energy balance. Bioresour Technol 2011;102:612e20. http://dx.doi.org/10.1016/ j.biortech.2010.08.005. [32] Fernandez AS, Hashsham SA, Dollhopf SL, Raskin L, Glagoleva O, Dazzo FB, et al. Flexible community structure correlates with stable community function in methanogenic bioreactor communities perturbed by glucose. Appl Environ Microbiol 2000;66:4058e67. http://dx.doi.org/10.1128/AEM.66.9.40584067.2000. ndez A, Huang S, Seston S, Xing J, Hickey R, Criddle C, et al. How stable is [33] Ferna stable? Function versus community composition. Appl Environ Microbiol 1999;65:3697e704. [34] Hashsham SA, Fernandez AS, Dollhopf SL, Dazzo FB, Hickey RF, Tiedje JM, et al. Parallel processing of substrate correlates with greater functional stability in methanogenic bioreactor communities perturbed by glucose. Appl Environ Microbiol 2000;66:4050e7. http://dx.doi.org/10.1128/AEM.66.9.40504057.2000.