Pilot-Scale Sequential Anaerobic―Aerobic Biological Treatment of Waste Streams from a Paper Mill

The pulp and paper industry produces a large quantity of wastewater containing recalcitrant organic compounds. In this study, a pilot ‐ scale anaerobic–aerobic sequential system was employed to treat four different waste streams produced in a kraft pulp and paper mill. The system consisted of a 2.3 m3 packed ‐ bed anaerobic digester and a completely mixed activated sludge process. Under the applied organic loading rate to the anaerobic digester (0.2 to 4.8 kg ‐ COD m−3 d−1), a COD removal efficiency of 50–65% was achieved. After the anaerobic treatment, the BOD/COD ratio of the effluent was low (0.12 ± 0.03), suggesting that additional pretreatment is necessary for the digester effluent to be further polished aerobically. Combined with the aerobic treatment, the overall COD removal efficiency was up to 70% for the substrates evaluated. Air purging before feeding sulfide ‐ containing substrate was shown to be effective for removing sulfide toxicity in the digester. Kinetic analysis showed that the pseudo ‐ first ‐ order degradation rate constants of the evaluated substrates are 0.28–0.46 d−1 in the anaerobic digester, with a methane production yield of 0.22–0.34 m3 ‐ CH4 kg ‐ COD−1 at standard temperature and pressure (0°C, 1 atm). These values are comparable to those found for other industrial substrates, indicating that an anaerobic process is a sound treatment alternative for the evaluated waste streams. The quality of biogas produced by the substrates was excellent, containing ∼80% of methane. The application of anaerobic treatment has the potential of significantly improving the energy footprint of the pulp and paper industry. © 2013 American Institute of Chemical Engineers Environ Prog, 33: 359–368, 2014

anaerobic digestion; biogas production; activated sludge; biological treatment; pulp and paper wastewater

Pulp and paper industry produces a large quantity of wastewater of high organic strength [1] , [2] . Even with the most modern operations, ∼60 m3 of wastewater is generated for every ton of paper produced [3] . Paper manufacturing processes include debarking, pulping, separation of pulp from cooking liquor, bleaching, stock preparation, and making the final paper products [4] . The pulping and bleaching processes produce the largest volume of wastewater whose composition is highly variable depending on the implemented technologies [5] , [6] , [7] . These effluents are loaded with organic matter with relatively high chemical oxygen demand (COD, 800–4400 mg ‐ O2 L−1), biochemical oxygen demand (BOD5, 300–2800 mg ‐ O2 L−1) and color (1200–6500 color unit) [8] , [9] , [10] , [11] . They also contain process by ‐ products such as extractives, carbohydrates, lignin and its derivatives, resin acids, sterols, and various chlorinated organic substances [12] , [13] , [14] , [15] .

Treatment of bleaching effluents typically incorporates primary clarification followed by a secondary biological treatment [16] , [17] . Primary clarification such as gravitational settling or dissolved air flotation is needed to remove the suspended solids, mainly made up by non ‐ biodegradable paper sludge. The biological treatment uses an activated sludge process (ASP), mostly aerated lagoons because of its relatively lower treatment cost. Although the aerobic system is capable of removing the BOD5, the COD removal typically ranges from 20% to 50% due to the existence of refractory, high ‐ molecular ‐ weight organic compounds [18] , [19] .

Anaerobic treatment, although widely used for treating agricultural and municipal wastes [20] , [21] , has not been extensively applied in the pulp and paper industry. More recently, it had been suggested that application of Upflow Anaerobic Sludge Blanket (UASB) reactors can achieve up to 70% COD removal for synthetic kraft pulp wastewater [22] . Compared to using a single ASP, a tandem UASB ‐ ASP system performs more consistently with 20–80% greater COD removal [4] , [23] . One major advantage of anaerobic treatment is that the process is capable of treating high ‐ organic ‐ strength streams that are not suitable for aerobic processes [24] . Furthermore, it has the benefit of lower treatment cost because the produced biogas contains 50–80% of methane [25] , which is a cleaner fuel compared to coal. However, earlier treatment evaluations for pulp and paper wastewater were conducted using synthetic streams or bench ‐ scale systems having small reactor volumes ranging from 5 to 50 L [22] , [26] , [27] , [28] , [29] , [30] , [31] , making the results less relevant to large ‐ scale operation. An investigation using a pilot ‐ scale system with real ‐ world industrial streams provides the performance data that are more representative for practical operation.

In this study, the treatability of the liquid wastes from a kraft paper mill was evaluated using a pilot ‐ scale, sequential bioreactor system consisting of an anaerobic and an aerobic process. The liquid wastes included streams produced from chlorine dioxide (CD) bleaching, from alkaline extraction reinforced with oxygen and peroxide (EOP) bleaching, from chemical (sulfate) pulping (so ‐ called foul condensates, FC), and from dewatering operation of plant wasted sludge (screw press liquor, SPL). For improving the digestibility and pH adjustment, a small volume of wasted sugar water (SW) from a food processing plant was blended as a co ‐ digestion substrate. We report the treatment performance, biogas production efficacy of each substrate and system characteristics of the pilot ‐ scale operation.

The schematic of the pilot ‐ scale treatment system is presented in Figure [NaN] . It consists of an equalization tank (2.1 m3), a continuous stirred tank reactor (CSTR) for facultative predigestion, a packed bed anaerobic digestion column and a CSTR activated sludge process (ASP) for effluent polishing. Both the predigestion and the ASP's aeration tank are of 0.95 m3 that can be maintained at three different water levels for controlling the hydraulic retention time (HRT). The packed bed column is of cylindrical shape (1.07 m in diameter and 2.60 m in height). It has a total volume of 2.3 m3 with 85% of the volume (2.0 m3) packed with a commercial ceramic bio ‐ packing media. The packing material is 1 ‐ in Raschig (cylindrical) rings (Wisconsin Stamping & Manufacturing, WI, USA) that have 203 m2 m−3 of specific surface area. A recirculation heating pump was utilized to control the influent temperature at 36 ± 2°C (mesophilic). Two fine bubble membrane diffusers (Model AFD270, Stamford Scientific, NY, USA) were used at the bottom of the aeration tank of the ASP. The final clarifier has a volume of 0.605 m3. The pilot system was built on a property outside the fence line of a pulp and paper mill for easy access to the waste streams.

The CD/EOP/FC/SPL streams were obtained twice a week from the mill in 0.76 m3 totes. The streams were mixed in the equalization tank before being fed to the facultative predigestion tank where a commercial culture powder (Meridian Bioenergy, Inc., USA) was introduced for bio ‐ augmentation (1:5000 bioculture to COD mass ratio) at 12 ‐ h HRT. The bioculture contains a mixture of four Bacillus strains and the needed substrates for initial maturation in a separate culture tank. The bio ‐ augmentation process provided initial breakdown of the large molecules or long ‐ chained organic compounds in the equalized waste streams to enhance the anaerobic digestion rate. The effectiveness is monitored through the pH drop (typically 0.5–1.0 pH unit) in the facultative predigestion stage. The packed bed digester was seeded with 0.19 m3 of anaerobic sludge (ca. 30,000 mg L−1) from a solid digester of food wastes at the bottom of the packed bed column. The substrates were then fed continuously into the digester from the predigestion tank.

The evaluation lasted for 156 days (from October 31, 2011 to April 3, 2012) and was divided into six periods according to different feeds and operating conditions. Initially, the packed bed column was operated as a downflow digester using a flow scheme similar to a trickling filter, with a space loading rate of ∼3 kg ‐ COD d−1 m−3 (based on the packing volume) and a recirculation ratio of 5.0. In the second half of the evaluation period (after the 80th day), the digester was operated as an upflow flooded bed reactor with an HRT of 1.7–2.4 days. The feeding and the operational characteristics of the digester are summarized in Table [NaN] . During the entire evaluation period, the pH in digester was controlled at a neutral range (6.92–7.60) and the digestion occurred at mesophilic temperature (31.5–34.5°C measured in the effluent). The temperature difference between influent and effluent was 3–4°C because of the seasonally low ambient temperature (5–15°C) in the pilot plant.

Pilot ‐ scale feeding activities and operating conditions during the six evaluation periods

1 The system was in the recovery mode during Days 37–44 (see Results and Discussion for details).

• 2 *Based on the volume of packing media.
• 3 Based on total volume of the packed ‐ bed digester.

Daily liquid samples were taken from six sampling points (1–6) for chemical analysis, and gas samples were taken from sampling point (7) as indicated in Figure [NaN] . Table [NaN] shows the chemical analysis performed for the samples. The pH, dissolved oxygen (DO), oxidation ‐ reduction potential (ORP) and temperature were measured on site using a calibrated portable meter with appropriate probes (Hach HQ40D with pH101, LBOD101, and IntelliCAL ORP probes) after samples were drawn from the reactors. All laboratory measurements were performed within 2 h after sampling. The total solids (TS), volatile solids (VS), total suspended solids (TSS) and volatile suspended solids (VSS) were measured using Standard Methods 2540 [32] . The 5 ‐ day BOD was determined using Standard Methods 5210 [32] . The COD, volatile fatty acids (VFA), total alkalinity (ALK), total nitrogen (TN), total phosphorus (TP), ammonia nitrogen (AN), sulfide and sulfate were determined using respective EPA approved methods (Hach Company CFR 136.3, 141, EPA 310.2, 350.1, 353.2, 365.3, DOC 316.5) with a time ‐ lapse heating reactor (Hach DRB200) and a spectrophotometer (Hach DR3800). The chloride was determined using a calibrated portable meter (Hach HQ40D) with an ion ‐ selective electrode (IntelliCAL chloride probe). Selected analyses of four replicates of a sample for all the above ‐ mentioned parameters typically had a relative standard deviation less than 5% except those for solids (<10%) and BOD (<12%). The completeness of data used for process performance evaluation was over 95%. The cumulative volume of evolved biogas was measured continuously using a digital gas flow meter (FILL ‐ RITE Model 820, TTS Corp., IN, USA). The composition of biogas was measured using a portable gas analyzer with an infrared detector (Model GEM 2000, Landtec Inc., Colton, CA, USA) at least four times a day. Standard gases (50% CO2, 50% CH4, dry air, and 1000 ppm H2S) were use for the calibration of the gas analyzer daily.

Analytical parameters and frequency for each sampling point in the pilot ‐ scale system

• 4 DO, dissolved oxygen; N, ammonium nitrogen; ORP, oxidation–reduction potential; T, temperature; TN, total nitrogen; TP, total phosphorus; TS, total solids; TSS, total suspended solids; VFA, volatile fatty acids; VS, volatile solids; VSS, volatile suspended solids.
• 5 Two samples weekly.

The characteristics of the evaluated waste streams are summarized in Table [NaN] . The COD concentrations range from 2500 to 4800 mg L−1. These liquid wastes have generally low solid contents (TS < 1 wt %). FC, EOP and SPL are slightly to moderately alkaline (pH 8.4–9.3). To bring the pH of these substrate down to a level suitable for anaerobic treatment, a waste sugar water (SW, COD = 408,000 mg L−1, pH 3.99, TN 62 mg L−1, TP 28 mg L−1) was blended into the substrates. A volume blending ratio of approximately 200:1 (substrate to SW) in the equalization tank was sufficient to bring the mixed liquid's pH to 7.0–7.5. CD is slightly acidic (pH 5.2). It was blended with either FC or EOP before treatment. FC also has a moderately high sulfide concentration (52 mg L−1).

Initial characteristics of the evaluated waste streams from the paper mill

• 6 n is the number of waste stream samples analyzed.
• 7 dCOD is the dissolved COD concentration, which represents the organic load excluding the contribution from suspended solids.

The operating parameters and performance of the packed bed digester for each of the six periods (Table [NaN] ) are summarized in Table [NaN] . For comparison, the typical values for liquid digesters are also listed. The low DO concentration (near detection limit of the DO probe) and ORP value (<−200 mV) indicate that the system was strictly anaerobic. VFA concentration was lower than 350 mg L−1 as acetic acid (HAc) except during the second operating period when the VFA level was elevated to >1000 mg L−1 as HAc. VFA accumulation has been suggested to significantly decrease pH and cause in vitro toxicity [33] . Total alkalinity was mostly over 1500 mg L−1 as CaCO3 except during the first operating period, mainly due to the low buffering capacity of FC as the substrate. Nevertheless, the pH (6.9–7.6) for the entire evaluation period was within the optimal range of anaerobic treatment (pH 6.5–7.5). TN and TP concentration in the effluent was relatively low (40–70 mg L−1 as N and 8–15 mg L−1 as P) because no nutrients were provided to observe the treatability of the substrates directly. COD removal in the digester ranged from 40 to 75% (except during the second period, Figure [NaN] ) at about 2 days of HRT. The relatively short HRT compared to typical liquid digesters (1–15 days) was required because the primary goal of the pilot plant experiment was to collect data for designing a full ‐ scale commercial treatment unit that treats a large quantity of waste streams from a paper mill (15–20 million gallons per day).

Summary of chemical characteristics in the anaerobic digester and the digestion performance

Figure [NaN] shows the time series plots of substrate COD concentration entering the digester, the biogas production and gas quality (Figure [NaN] a) and COD removal (Figure [NaN] b). The vertical dash lines divide the operational periods as indicated in Table [NaN] . Even with the equalization of the substrate, the organic strength of incoming waste streams still had a moderate variability, with COD concentrations from 3000 to 5000 mg L−1. About 10 days after the initial seeding, the COD removal reached a relatively stable value. From Days 10–32, the COD removal efficiency ranged mostly between 65 and 75%. The daily biogas production reached up to 2 m3 d−1 at standard temperature and pressure (STP, 0°C, 1 atmosphere; all gas volumes mentioned hereafter have been normalized to STP) with CH4 content in the range of 75–80%. During this period, the FC condensate was purged with air in the balanced tank for at least 2 h before the facultative predigestion stage to avoid the potential impact from the high H2S content. The 2 ‐ h purging time was determined by a laboratory air stripping experiment in a 2 ‐ L batch reactor without pH adjustment (Figure [NaN] ). After the 2 ‐ h stripping, the sulfide concentration decreased from 75 to 1 mg L−1 for the tested FC sample. This was caused by both H2S volatilization and aqueous sulfide oxidation by dissolved oxygen, which contributed to the increasing sulfate concentration with respect to time.

From Days 32 to 37 (before switching the substrate to EOP), the air purging was stopped to observe if sulfide stripping played an important role in the treatment of FC. Immediately after stopping the air purging, the H2S content in the biogas increased to more than 5000 ppm (the upper detection limit of the portable gas meter). During this period, the COD removal efficiency dropped to 20–30% (Figure [NaN] ). The biogas production rate and CH4 content also sharply decreased. Visually, there was white scum appeared in the digester effluent and the effluent pH dropped to 6.3, suggesting the accumulation of VFAs. The compromised digestion performance was attributed to the sulfide toxicity caused by two possible reasons. One is that the sulfate reducing bacteria became predominated during this period. This led to the continuous conversion of existing sulfate in FC to sulfide and thus decreased the activity of methanogens [38] . The other was due to presence of excessive H2S in the digester at the lowered pH. Hydrogen sulfide (H2S) is a weak diprotic acid with the first acid dissociation constant at 10−7 M. At pH < 7, the speciation is dominated by H2S, the most toxic form of the sulfide species [38] . The volatilization of H2S also contributed to the increased H2S concentration measured in the biogas.

During the first two periods (Table [NaN] ), FC was more readily treated in the packed bed digester compared to EOP because FC contains low molecular weight organic compounds (lower ‐ carbon fatty acids, methanol, and dimethyl sulfide) [26] , [39] that are more easily digestible. Earlier lab ‐ scale treatability studies indicated that FC has a relatively higher bio ‐ degradability (>80%) compared to the waste streams produced from softwood and hardwood processing (∼30%) [40] , [41] . Another explanation is the partial volatilization of the organic contents caused by the gas ‐ liquid exchange when the packed bed digester was operated like a trickling filter. As shown in Figure [NaN] , the COD of FC can be removed considerably (by 26%) after 2 ‐ h air stripping. The volatilization occurred in the digester can contribute to the higher COD removal efficiency.

From Day 45–81, EOP was fed as the primary substrate and the digester was reseeded in the same fashion as the initial seeding (0.19 m3 of anaerobic sludge, TSS = ∼30,000 mg L−1). After the reseeding, the biogas production and quality recovered somewhat (0.45 ± 0.19 m3 d−1 and 75.2% ± 9.3% CH4, Figure [NaN] b). However, the COD removal is relatively low (25.9% ± 9.9%). Visual observation of the internal packing material revealed that there was considerably lower biofilm formation, probably due to a combination of the much higher COD and dissolved solids (estimated by the difference of TS and TSS measurements in Table [NaN] ) in EOP [42] and the continuous recirculation flow that sloughed off the biomass. When FC was the primary substrate (Days 10–32), there was a visibly greater amount of biofilm formed on the media.

In an attempt to improve the treatment performance, the flow scheme of the digester was modified from the downflow anaerobic trickling filter to upflow flooded bed digester on Day 82 (Table [NaN] ). The change of feed flow scheme greatly improved the biodegradation (Figure [NaN] a). Both the COD removal and biogas production increased continuously over the next 7 days. During Days 82–135, the mean COD removal was 56.2% ± 5.4% (peaked at >70%). The mean biogas production was 1.1 ± 0.35 m3 d−1 with a CH4 content at 81.5% ± 1.9%. During this period, the BOD5 concentrations were also measured twice a week. The BOD5 removal by the digester consistently ranged from 80 to 90%, with a mean of 86.5% ± 3.4%, suggesting that the majority of the easily biodegradable organics was removed.

On Day 136, CD was blended as a co ‐ substrate at 50:50 volume ratio (Table [NaN] , the fourth period). Due to the lower pH of the CD (Table [NaN] ), the SW was not supplemented for pH adjustment (Table [NaN] , the fifth period). Finally, SPL was fed with a volume ratio of SPL:EOP:CD at 20:40:40 (Table [NaN] , the sixth period). During these three periods, the incoming COD had a decreasing trend because of the substantially lower dissolved COD (dCOD) of the SPL and CD (Table [NaN] ) and the exclusion of the waste sugar water as the co ‐ substrate. This led to a decrease of gas production rate. Nevertheless, with the decreased COD removal efficiency, the biogas yield was increased (Table [NaN] ) and the biogas quality in terms of CH4 content remained excellent (Figure [NaN] ).

Figure [NaN] shows the scattered plot between the COD removal and organic loading rate (OLR). The COD removal efficiency was not significantly affected by varying the OLR, resulting in a linear increase of COD mass removal with respect to the applied OLR (Figure [NaN] ). This suggests that the system should have a greater treatment capacity than the OLR range applied during the evaluation period. There are a few data points that deviated from the main cluster in the scatter plot. The data points represent the period when VFA accumulation and potential H2S toxicity occurred during Days 32–37.

To estimate the methane production yield of the evaluated substrates, the cumulative CH4 production at STP and the cumulative COD mass digested were plotted against the time axis (Figure [NaN] ). After excluding the periods associated with substrate changes and system recovery, there are six linear stages that the system was performing consistently. The methane yield was calculated as the ratio of the slopes of the two curves in Figure [NaN] . The values, shown in Table [NaN] , range from 0.22 to 0.34 m3 ‐ CH4 kg ‐ COD−1 for the substrates evaluated. The methane production yields are comparable to those reported for other industrial waste streams (0.15 ‐ 0.45 m3 ‐ CH4 kg ‐ COD−1, Table [NaN] ).

Comparison of reported anaerobic degradation rate constants, methane yields, substrate concentrations, and methane contents

8 The degradation kinetics cannot be assessed using pseudo ‐ first ‐ order kinetics because of the downflow PBR configuration.

Kinetic analysis was performed to estimate the rate of substrate degradation. Given the concentrations measured in this pilot ‐ scale study, a pseudo ‐ first ‐ order kinetic law was applied. Similar approaches were also utilized in earlier studies [42] , [47] , [51] , which provides the basis for data comparison. The kinetics of an upflow packed bed reactor can be treated similarly to a plug flow reactor. Therefore, the effluent concentration can be expressed as:

C=C0e(−kθ)

where C0 represents initial the substrate COD concentration (mg L−1) and θ is the HRT (d−1). The mean COD concentrations in the influent and the effluent were applied to calculate k during Periods #3 to #6 (the data obtained during Period #1 and #2 are not applicable to Eq. since the digester was operated similar to an anaerobic trickling filter). The values are in the range of 0.28–0.46 d−1 (Table [NaN] ).

Compared to the methane yields and first ‐ order degradation rate constants reported for other liquid substrates using different reactors (k = 0.15–1.2 d−1), the values found in this study are comparable to those in the earlier assessments. The somewhat slower kinetics is probably due to the difference in substrate digestibility, the reactor design, and the greater difficulty in controlling the feeding and operating conditions in a pilot ‐ scale experiment (all other studies were performed using bench ‐ scale systems). The substrate utilization rate constants in this study (0.28–0.46 d−1) are substantially lower than those of slaughterhouse wastewater (1.2 d−1[46] ) and fruit ‐ processing wastewater (0.89 d−1[47] ) digested in fluidized bed and cell bioreactors; comparable to that of palm oil mill effluent (0.36 d−1[48] ) treated in anaerobic pond; and higher than that of the soluble fraction in olive mill solid residue (0.145 d−1[42] ) digested in a CSTR. The methane production yields are consistent with other substrates (Table [NaN] ) and the biogas quality is excellent (∼80%, Table [NaN] ). We concluded that the packed bed digester performed satisfactorily for anaerobic treatment of the evaluated pulp and paper waste streams.

Starting from the third operating period, the ASP system was initiated for polishing the effluent from the anaerobic digester. The aeration tank was seeded with 0.76 m3 recycle sludge obtained from a local municipal activated sludge plant. On Day 124, BOD data (two samples per week) were collected to assess the performance of the ASP for polishing the digester effluent. The HRT of the aeration tank was maintained at 8 h, the DO was kept at 2–3 mg L−1, and the sludge recycle ratio was controlled at about 0.5 using a piston pump. During the period when BOD data were collected, the mean concentration mixed liquor suspended solid in aeration tank was 3234 ± 351 mg L−1. TKN (Total Kjeldahl Nitrogen) and TP concentrations were 14.0 ± 2.1 mg L−1 as N and 5.5 ± 0.7 mg L−1 as P, respectively, suggesting no nutrient deficiency.

The performance of the ASP system is shown in Figure [NaN] . During the third period (before Day 142), the influent to the aeration tank had a BOD concentration of 218 ± 61 mg L−1. After polishing by the aerobic process, the BOD concentration of the effluent was 73 ± 51 mg L−1. The BOD removal was ∼65%. The BOD removal decreased to ∼40% during the period when CD was used as a co ‐ substrate (after Day 142). After anaerobic treatment, the BOD/COD ratio of the stream was quite low (0.12 ± 0.03), indicating low aerobic biodegradability (< 0.2, [37] , [52] ). After the aerobic treatment by the ASP, the BOD/COD ratio further decreased to 0.07 ± 0.02. The low BOD/COD ratio of the effluent from the anaerobic digester was probably due to the presence of recalcitrant organic compounds in the pulp and paper waste streams. To improve the aerobic biological treatability of the digester effluent, additional pretreatment processes capable of breaking down recalcitrant organic compounds such thermal hydrolysis [53] and oxidative pretreatment [54] , [55] , [56] would be necessary. Ultimately, the discharge requirement and the overall treatment cost govern the selection of treatment alternatives. Combining both anaerobic and aerobic treatment, the overall removal efficiency was 65–70% for most of the evaluation period.

In this study, we evaluated the treatability of the four waste streams from the pulp and paper industry in a pilot plant consisting of a packed bed anaerobic digester and an activated sludge process. It was found that all waste streams are readily treatable. The anaerobic treatment removed 50–65% of substrate COD. Coupled with the aerobic treatment using a completely mixed activated sludge process, the overall COD removal efficiency was 55–70%. Kinetic analysis showed that the pseudo ‐ first ‐ order degradation rate constants of the evaluated substrates in the digester are in the range of 0.28–0.46 d−1, with a methane production yield of 0.22–0.34 m3 ‐ CH4 kg ‐ COD−1. The application of anaerobic treatment has the potential of significantly improving the energy footprints of the pulp and paper industry.

During the pilot plant study, the packed bed was operated as an anaerobic trickling filter and as an upflow flooded anaerobic digester. For the waste stream produced by the bleaching process (EOP), the upflow configuration yielded a better treatment performance in terms of COD removal. For substrate containing moderately high sulfide concentration (FC), air stripping before feeding the substrate into the liquid digester effectively removed sulfide toxicity in the digester. The biogas quality produced by the substrates was excellent, containing ∼80% methane. After the anaerobic treatment, the BOD/COD ratio of the effluent was low (0.12 ± 0.03). This suggests that additional pretreatment is required for the effluent to be further treated aerobically.

ALK total alkalinity

AN ammonia nitrogen

ASP activated sludge process

BOD 5 5 ‐ day biochemical oxygen demand

CD waste stream from Chlorine Dioxide bleaching process

COD chemical oxygen demand

CSTR continuous stirred tank reactor

EOP waste stream from alkaline Extraction and Oxygen & Peroxide bleaching process

EPA Environmental Protection Agency

FC Foul Condensate, a waste stream from chemical pulping process

HRT hydraulic retention time

OLR organic loading rate

ORP oxidation reduction potential

PBR packed bed reactor

SPL screw press liquor, a waste stream from paper sludge dewatering process

STP standard temperature and pressure (0°C, 1 atmosphere)

SW sugar water, a waste stream from a food processing plant

TKN Total Kjeldahl Nitrogen

TN total nitrogen

TP total phosphorous

TS total solids

TSS total suspended solids

UASB upflow anaerobic sludge blanket reactor

VFA volatile fatty acids

VS volatile solids

VSS volatile suspended solids

This study was supported by MeadWestvaco Evadale TX facility (Project No: MWV0001). The authors would like to thank Dipendra Wagle, Sophia Yang, Yolanda Wang, Brandon Corace, and Erik Corace for their assistance in the field operation and laboratory analysis. The assistance of Gary Colson, Michael Clapper, and Robert Sasser in supporting this work and in obtaining waste streams is greatly appreciated. The administrative assistance of Stewart Cairns, Reid Sweet, and Thomas Sitton are also acknowledged.

1 Satyawali, Y., Pant, D., Singh, A. & Srivastava, R.K. ( 2009 ). Treatment of rayon grade pulp drain effluent by upflow anaerobic fixed packed bed reactor (UAFPBR), Journal of Environmental Biology, 30, 667 – 672.

• 2 Moiseev, A., Schroeder, H., Kotsaridou-Nagel, M., Geissen, S.U. & Vogelpohl, A. ( 2004 ). Photocatalytical polishing of paper ‐ mill effluents. Water Science and Technology, 49, 325 – 330.
• 3 Thompson, G., Swain, J., Kay, M. & Forster, C.F. ( 2001 ). The treatment of pulp and paper mill effluent: A review. Bioresource Technology, 77, 275 – 286.
• 4 Tezel, U., Guven, E., Erguder, T.H. & Demirer, G.N. ( 2001 ). Sequential (anaerobic/aerobic) biological treatment of Dalaman SEKA pulp and paper industry effluent. Waste Management, 21, 717 – 724.
• 5 Ugurlu, M. & M.H. Karaoglu. ( 2009 ). Removal of AOX, total nitrogen and chlorinated lignin from bleached Kraft mill effluents by UV oxidation in the presence of hydrogen peroxide utilizing TiO(2) as photocatalyst. Environmental Science and Pollution Research International, 16, 265 – 273.
• 6 Mounteer, A.H., Souza, L.C., & Silva, C.M. ( 2007 ). Potential for enhancement of aerobic biological removal of recalcitrant organic matter in bleached pulp mill effluents. Environmental Technology, 28, 157 – 164.
• 7 Zhu, C., Wang, L., Kong, L., Yang, X., Zheng, S., Chen, F., MaiZhi, F. & Zong, H. ( 2000 ). Photocatalytic degradation of AZO dyes by supported TiO2 + UV in aqueous solution. Chemosphere, 41, 303 – 309.
• 8 Rintala, J.A. & Lepisto, S.S. ( 1998 ). Thermophilic anaerobic treatment of sulphur rich forest industry wastewater. Biodegradation, 9, 225 ‐ 232.
• 9 Buzzini, A.P., Patrizzi, L.J., Motheo, A.J. & Pires, E.C. ( 2007 ). Preliminary evaluation of the electrochemical and chemical coagulation processes in the post ‐ treatment of effluent from an upflow anaerobic sludge blanket (UASB) reactor. Journal of Environmental Management, 85, 847 – 857.
• 10 Sheldon, M.S., Zeelie, P.J., & Edwards, W. ( 2012 ). Treatment of paper mill effluent using an anaerobic/aerobic hybrid side ‐ stream membrane bioreactor. Water Science and Technology, 65, 1265 – 1272.
• 11 Singhal, A. & Thakur, I.S. ( 2009 ). Decolourization and detoxification of pulp and paper mill effluent by Emericella nidulans var. nidulans. Journal of Hazardous Materials, 171, 619 – 625.
• 12 Xie, T.M., Abrahamsson, K., Fogelqvist, E. & Josefsson, B. ( 1986 ). Distribution of chlorophenolics in a marine environment. Environmental Science and Technology, 20, 457 – 463.
• 13 Kostamo, A., Holmbom, B., & Kukkonen, J.V. ( 2004 ). Fate of wood extractives in wastewater treatment plants at kraft pulp mills and mechanical pulp mills. Water Research, 38, 972 – 982.
• 14 Aaltonen, T.M., Jokinen, E.I., Lappivaara, J. Markkula, S.E., Salo, H.M., Leppanen, H. & Lammi, R. ( 2000 ). Effects of primary ‐ and secondary ‐ treated bleached kraft mill effluents on the immune system and physiological parameters of roach. Aquatic Toxicology, 51, 55 – 67.
• 15 Leiviska, T., Ramo, J., Nurmesniemi, H., Poykio, R. & Kuokkanen, T. ( 2009 ). Size fractionation of wood extractives, lignin and trace elements in pulp and paper mill wastewater before and after biological treatment. Water Research, 43, 3199 – 3206.
• 16 Pereira, R., Antunes, S.C., Goncalves, A.M., Marques, S.M., Goncalves, F., Ferreira, F., Freitas, A.C., Rocha-Santos, T.A., Diniz, M.S., Castro, L., Peres, I. & Duarte, A.C. ( 2009 ). The effectiveness of a biological treatment with Rhizopus oryzae and of a photo ‐ Fenton oxidation in the mitigation of toxicity of a bleached kraft pulp mill effluent. Water Research, 43, 2471 – 2480.
• 17 Baer, K.N., Bankston, C.R., Mosadeghi, S. & Schlenk, D. ( 2009 ). The effects of pulp and paper mill effluent on physiological and hematological endpoints in fingerling largemouth bass ( Micropterus salmoides ). Drug and Chemical Toxicology, 32, 59 – 67.
• 18 Villamar, C.A., Jarpa, M., Decap, J. & Vidal, G. ( 2009 ). Aerobic moving bed bioreactor performance: A comparative study of removal efficiencies of kraft mill effluents from Pinus radiata and Eucalyptus globulus as raw material. Water Science and Technology, 59, 507 – 514.
• 19 Chandra, R., Raj, A., Purohit, H.J. & Kapley, A. ( 2007 ). Characterisation and optimisation of three potential aerobic bacterial strains for kraft lignin degradation from pulp paper waste. Chemosphere, 67, 839 – 846.
• 20 Chen, G.Y., Zheng, Z., Chang, Z.Z., Ye, X.M. & Luo, Y. ( 2011 ). Effect of NaOH ‐ treatment on dry ‐ thermophilic anaerobic digestion of Spartina alterniflora. Environmental Science, 32, 2158 – 2163.
• 21 Villegas, J.D., de Laclos, H.F., Dovat, J., Membrez, Y. & Holliger, C. ( 2011 ). Nitrogen removal from digested manure in a simple one ‐ stage process. Water Science and Technology, 63, 1991 – 1996.
• 22 Buzzini, A.P., Gianotti, E.P., & Pires, E.C. ( 2005 ). UASB performance for bleached and unbleached kraft pulp synthetic wastewater treatment. Chemosphere, 59, 55 – 61.
• 23 Lerner M., Stahl, N., & Galil, N.I. ( 2007 ). Aerobic vs. anaerobic ‐ aerobic biotreatment: paper mill wastewater. Environmental Engineering Science, 24, 277 – 285.
• 24 Morgan ‐ Sagastume, F. & Allen, D.G. ( 2003 ). Effects of temperature transient conditions on aerobic biological treatment of wastewater. Water Research, 37, 3590 – 3601.
• 25 Mackie, I. & Bryant, P. ( 1995 ). Anaerobic digestion of cattle waste at mesophilic and thermophilic temperature. Applied Microbiology and Biotechnology, 43, 346 – 350.
• 26 Zhou, W., Imai, T., Ukita, M. Li, F. & Yuasa, A. ( 2007 ). Effect of limited aeration on the anaerobic treatment of evaporator condensate from a sulfite pulp mill. Chemosphere, 66, 924 – 929.
• 27 Buzzini, A.P., Miwa, D.W., Motheo, A.J. & Pires, E.C. ( 2006 ). Use of electrochemical oxidation process as post ‐ treatment for the effluents of a UASB reactor treating cellulose pulp mill wastewater. Water Science and Technology, 54, 207 – 213.
• 28 Mahadevaswamy, M., Murthy, B.M. & Girijamma, A.R. ( 2004 ). Performance evaluation of up ‐ flow anaerobic sludge blanket (UASB) reactor for treatment of paper mill wastewater. Journal of Environmental Science (China), 16, 194 – 198.
• 29 Lee, Y. ‐ W., Choi, J.-Y., Kim, J.-O. & Chung, J. ( 2010 ). Evaluation of UASB/CO 2 stripping system for simultaneous removal of organics and calcium in linerboard wastewater. Environmental Progress and Sustainable Energy, 30, 187 – 195.
• 30 Buzzini, A.P. & Pires, E.C. ( 2007 ). Evaluation of a upflow anaerobic sludge blanket reactor with partial recirculation of effluent used to treat wastewaters from pulp and paper plants. Bioresource Technology, 98, 1838 – 1848.
• 31 Latif, M.A., Ghufran, R., Wahid, Z.A. & Ahmad, A. ( 2011 ). Integrated application of upflow anaerobic sludge blanket reactor for the treatment of wastewaters. Water Research 45, 4683 – 4699.
• 32 Lenore, S.C. & Andrew, D.E. ( 1998 ). APHA, AWWA and WEF standard methods for the examination of water and wastewater, Washington, DC, USA : APHA ‐ WWA ‐ WEF.
• 33 Gray, N.F. ( 2010 ). Water technology: An introduction for environmental scientists and engineers ( 3rd edition ), Oxford : Elseiver.
• 34 Alley, E.R. ( 2007 ). Water Quality Control Handbook ( 2nd Edition ), Colombus, OH : WEF+McGraw ‐ Hill.
• 35 Tchobanoglous, G., Burton, F., & Stensel, H. ( 2002 ). Wastewater engineering: Treatment and reuse ( 4th Edition ), Columbus, OH : McGraw ‐ Hill.
• 36 Khanal, S.K. ( 2008 ). Anaerobic biotechnology for bioenergy production principles and applications ( 1st Edition ), Hoboken, NJ : Wiley ‐ Blackwell.
• 37 Gilbert, E. ( 1987 ). Biodegradability of ozonation products as a function of COD and DOC elimination by example of substituted aromatic substances. Water Research, 21, 1273 – 1278.
• 38 Chen, Y., Cheng, J.J., & Creamer, K.S. ( 2008 ). Inhibition of anaerobic digestion process: A review. Bioresource Technology, 99, 4044 – 4064.
• 39 Dufresne, R., Liard, A., & Blum, M.S. ( 2001 ). Anaerobic treatment of condensates: Trial at a kraft pulp and paper mill. Water Environment Research, 73, 103 – 109.
• 40 Norrman, J., Narbuvold, R., & Nystrom, L. ( 1984 ). Anaerobic treatability of waste water from pulp and paper industries. Biotechnology Advances, 2, 329 – 345.
• 41 Yang, M.I., Edwards, E.A., & Allen, D.G. ( 2010 ). Anaerobic treatability and biogas production potential of selected in ‐ mill streams. Water Science and Technology, 62, 2427 – 2434.
• 42 Rincon, B., Sanchez, E., Raposo, F., Borja, R., Travieso, L., Martin, M.A. & Martin, A. ( 2008 ). Effect of the organic loading rate on the performance of anaerobic acidogenic fermentation of two ‐ phase olive mill solid residue. Waste Management, 28, 870 – 877.
• 43 Xie, K., Lin, H.J., Mahendran, B., Bagley, D.M. Leung, K.T. Liss, S.N. & Liao, B.Q. ( 2010 ). Performance and fouling characteristics of a submerged anaerobic membrane bioreactor for kraft evaporator condensate treatment. Environmental Technology, 31, 511 – 521.
• 44 Driessen, W. & Wasenius, C.O. ( 1994 ). Combined anaerobic/aerobic treatment of peroxide bleached TMP mill effluent. Water Science and Technology, 29, 381 – 389.
• 45 Lee, Y., Choi, J., Kim, J. & Chung, J.W. ( 2010 ). Evaluation of UASB/CO 2 stripping system for simultaneous removal of organics and calcium in linerboard wastewater. Environmental Progress and Sustainable Energy, 30, 187 – 195.
• 46 Borja, R., Banks, C.J., & Wang, Z. ( 1995 ). Kinetic evaluation of an anaerobic fluidized ‐ bed reactor treating slaughterhouse waste ‐ water. Bioresource Technology, 52, 163 – 167.
• 47 Borja, R. & Banks, C.J. ( 1994 ). Kinetic study of anaerobic digestion of fruit ‐ processing wastewater in immobilized ‐ cell bioreactors. Biotechnology and Applied Biochemistry, 20, 79 – 92.
• 48 Wong, K.K. & Springer, A.M. ( 1981 ). A first ‐ order kinetic model for designing anaerobi ponds in the treatment of palm oil mill effluent. Agricultural Wastes, 3, 35 – 42.
• 49 Zinatizadeh, A.A., Mohamed, A.R., and Najafpour, G. ( 2006 ). Kinetic evaluation of palm oil mill effluent digestion in a high rate up ‐ flow anaerobic sludge fixed film bioreactor. Process Biochemistry, 41, 1038 – 1046.
• 50 Faisal, M. & Unno, H. ( 2001 ). Kinetic analysis of palm oil mill wastewater treatment by a modified anaerobic baffle reactor. Biochemical Engineering Journal, 9, 25 – 31.
• 51 Borja, R., Martin, A., Banks, C.J., Alonso, V. & Chica, A. ( 1995 ). A kinetic study of anaerobic digestion of olive mill wastewater at mesophilic and thermophilic temperatures. Environmental Pollution, 88, 13 – 18.
• 52 Alvares, A.B., Diaper, C., & Parsons, S.A. ( 2001 ). Partial oxidation by ozone to remove recalcitrance from wastewaters—A review. Environmental Technology, 22, 409 – 427.
• 53 Wang, Z.J. & Wang, W. ( 2005 ). Enhancement of sewage sludge anaerobic digestibility by thermal hydrolysis pretreatment. Huan Jing Ke Xue, 26, 68 – 71.
• 54 Cokgor, E.U., Alaton, I.A., Karahan, O., Dogruel, S. & Orhon, D. ( 2004 ). Biological treatability of raw and ozonated penicillin formulation effluent. Journal of Hazardous Materials, 116, 159 – 166.
• 55 Brenner, A., Belkin, S. & Abeliovich, A. ( 1994 ). Development of a pretreatment program to improve biological treatability of high strength and toxic industrial wastewater. Water Science Technology, 29, 29 – 37.
• 56 Kim, Y.K. & Huh, I.R. ( 1997 ). Enhancing biological treatability of landfill leachate by chemical oxidation. Environmental Engineering Science, 14, 73 – 79.

Graph: (a) Schematic diagram of the pilot ‐ scale treatment system. The system is a functional design with transfer pumps for flow transport. Points #1 to #6 are the liquid sampling locations and Point #7 is the gas sampling location. (b) Photo of anaerobic section. (c) Photo of the aerobic section. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Graph: Biogas production and COD removal of the anaerobic digester: (a) daily biogas production and major gas components (CH 4 and CO 2 ), and (b) feeding COD concentration and COD removal efficiency. Details of the six operational periods are shown in Table . “M” denotes the maintenance and recovery period.

Graph: Results of laboratory air stripping experiments for a foul condensate sample. After 2 h of stripping, the COD concentration decreased by 26%, and sulfide concentration decreased from 75 mg L −1 to 1 mg L −1.

Graph: Scatter plots between organic loading rate and COD mass removal. Data points in the circled area are those recorded when H 2 S toxicity occurred during the period "M" in Figure .

Graph: Cumulative CH 4 production (at standard temperature and pressure) and cumulative COD mass digested during the evaluation period. There are six linear periods (Table ) during which the data were used for calculating the methane yield.

Graph: Performance of the aerobic treatment as a polishing step for the digester effluent. After the treatment, BOD/COD ration was consistently low.