Advanced biological sequential treatment of mature landfill leachate using aerobic activated sludge SBR and fungal bioreactor

Abstract

This study utilized Penicillium spp. to treat mature landfill leachate (MLL) in a continuous bioreactor and batch experimental tests under non-sterile conditions. MLL characteristics such as chemical oxygen demand (COD), soluble COD (sCOD), total carbon (TC), total organic carbon (TOC), and color removal efficiency were determined. The lignocellulosic enzymatic activity of laccase (Lac), lignin-peroxidase (LiP), and manganese-peroxidase (MnP) was also determined. The batch experimental test was carried out with raw and pretreated MLL containing the initial NH₄⁺–N concentrations of 0, 105, 352, and 914 mg/L. A maximum COD reduction of 41% and maximum enzymatic activity of 193, 37, and 25 U/L for Lac, LiP and MnP was recorded for the MLL containing 352 mg/L NH₄⁺–N. The continuous bioreactor exhibited maximum values of 52, 54, 60, 58, and 75 percentage of COD, sCOD, TC, TOC, and color removal efficiency with MLL containing 352 mg/L NH₄⁺–N that was pretreated at HRT 120 h, while the maximum detected lignocellulosic enzymatic activities were 149, 27, and 16 U/L for Lac, LiP, and MnP, respectively. A total of 64% COD reduction was achieved from the raw MLL considering 12% COD and 100% NH₄⁺–N reduction in the aerobic activated sludge sequencing batch reactor pretreatment process. The steady and higher removal efficiency of the bioreactor over the entire study period is promising for further exploration to enhance removal of refractory contaminants from the MLL.

Electronic supplementary material

The online version of this article (10.1007/s40201-020-00466-z) contains supplementary material, which is available to authorized users.

Introduction

Leachate is defined as mature landfill leachate (MLL) when landfill age is higher than 10 years. It is considered as one of the highly polluted wastewater due to the existence of various inherent toxic contaminants such as humic and fulvic substances, xenobiotic endocrine disrupting compounds, trace organic contaminants, heavy metals, higher concentration of ammonium, and lower biodegradable portion [1–3]. Although, biological processes have effectively been used to remediate young leachate with higher ratio between biochemical oxygen (BOD) and chemical oxygen demand (COD) (BOD/COD >0.5), MLL treatment still remains challenging [4, 5]. Conversely, physico-chemical process such as advanced oxidation, chemical precipitation, absorption-adsorption, coagulation-flocculation, and ion-exchange have been investigated to treat MLL [6], but high operating cost and limited versatility [7] have restricted the applicability and use of these processes. Consequently, treatment of leachate together with municipal sewage in the wastewater treatment plant (WWTP) is still considered as the most economical option. The inadequacy of conventional WWTP to remove recalcitrant components necessitate innovation of sustainable and advanced biological treatment option to remediate MLL [8, 9]. Deployment of lignocellulosic biomass such as fungi is an environmental-friendly, cost effective, and sustainable technological approach. Fungi can survive under relatively higher toxic and environmentally stressed conditions (i.e. low pH and nutrient deficiency). Additionally, the lignocellulosic fungi release diverse extracellular enzymes such as lignin-peroxidase (LiP), manganese-peroxidase (MnP), versatile peroxidase (VP), and laccase (Lac) to degrade a broad array of recalcitrant fractions [10].

Fungal bioreactors such as stirred tank, airlift, bubble column, and fluidized bed reactors are designed with the provision of adequate oxygen transfer capability, an essential nutrient for efficient fungal growth, to treat dyes and wastewaters [11, 12]. Bioreactor performance is limited by several factors which make them inconvenient in many cases such as continuous agitation of stirred tank reactor that causes high shear stress and leads to the mycelium eruption [13, 14], insufficient mixing in airlift reactor with high biomass densities [12], and non-uniform mixing in bubble column reactor due to scale-up and high pressure drop with packed column reactor [15].

Although fungal treatment of landfill leachate is a promising approach, most of the reported studies of efficient removal of contaminants from leachate was carried out in batch experiments with either young (BOD/COD >0.5) or intermediate leachate (BOD/COD: 0.1–0.5), while studies with mature leachate (BOD/COD <0.1) alone are very limited [16]. Likewise, limited number of continuous fungal bioreactor studies have been reported [8, 17]; most studies herein have remediated young or intermediate leachate and only a single study has reportedly used MLL [18]. From the exhaustive literature review, neither full-scale studies nor successful fungal treatment of mature leachate has been investigated. On the other hand, most of the fungal biodegradation studies have been conducted under controlled aseptic condition and the influence of autochthonous microorganisms during the treatment of MLL under not-sterile condition is not well documented.

Penicillium spp. exhibited as the most efficient species in our previous study [19] among the six selected fungi (three Ascomycetes filamentous fungi and three Basidiomycetes white-rot fungi) to treat MLL and therefore it was utilized in the current study. The lignocellulosic enzymatic activities (Lac, LiP, and MnP) of Penicillium spp. has previously been reported in literature ([20, 21]; X. [22]). White-rot-fungi got great attention from the researchers for fungal bioremediation due to their enzymatic activity [23]. However, only a single study has reportedly utilized filamentous fungi (Aspergillus oryzae) to treat distillery spent wash effluent through bioreactor [24]. From the authors’ best knowledge, this study appears to be the first study to report the utilization of Penicillium spp. (Ascomycetes filamentous fungus) in a continuous bioreactor to remediate refractory contaminants from the MLL under non-sterile conditions. Thus, the novelty of the present study is obvious for bioremediation of toxic refractory contaminants from MLL.

Ellouze et al. [25] reported 360 mg/L of ammonia nitrogen concentration (NH₄⁺–N) as the optimal condition for fungal treatment and higher concentration caused ammonia toxicity. The NH₄⁺–N concentration of the MLL used in the current study was >900 mg/L, therefore, a pretreatment option was utilized to reduce the NH₄⁺–N concentration before the fungal treatment. This study investigated the treatment efficiency of Penicillium spp. to remove recalcitrant fractions from MLL under the non-sterile condition in a sequential treatment option by using aerobic activated sludge sequencing batch reactor (SBR) and rotating packed bed fungal bioreactor. The utilization of SBR was considered to be pre-treatment process to reduce the ammonium concentration.

The aims of this study were: 1) to evaluate the effect of NH₄⁺–N concentration on the fungal enzymatic activities (Lac, LiP, and MnP) and COD and soluble COD (sCOD) removal efficiencies from MLL through batch tests, 2) to investigate the effect of aerobic activated sludge pretreatment process (SBR) on COD and NH₄⁺–N removal efficiency from MLL, and 3) to evaluate the effect of hydraulic retention time (HRT) of fungal bioreactor on the contaminants removal efficiencies i.e. percentage removal of COD, soluble COD (sCOD), total carbon (TC), total organic carbon (TOC), and color removal efficiencies in accordance with the Lac, LiP, and MnP. enzymatic activities.

Materials and methods

Leachate characterization

Landfill leachate was periodically collected from the Brady Road Resource Management Facility (BRRMF), Winnipeg, Canada (Well Identification Number # 24, location: 49°46′0”N, 97°11′57.5”W) during the spring, summer, and fall of 2017. The landfill is in operation since 1973 and landfilling is done with both residential and commercial wastes. The physicochemical properties of leachate were evaluated immediately after arrival to the laboratory and before each test. The samples were stored at 4 °C before conducting further tests. The historic physicochemical characterization data for MLL of the same well (# 24) was collected from BRRMF to understand the nature of the contaminants over time.

The physicochemical parameters of MLL from the laboratory analysis (Table 1) and four years historic data (Supporting document Table 4) exhibited the nature and specificity of a wide range of toxic contaminants in the existing samples, and the environmental fate, effects, exposure route, and toxicity still remain unknown for many of them ([26]; G. [27]). The age of buried waste cell (> 10 years) of the landfill and the BOD₅/COD ratio (< 0.1) validated the leachate category as fully mature [28, 29].

Table 1.

Physicochemical parameters of MLL characterized in the laboratory

Parameters	Units	Raw leachate	Wastewater
pH		8.1 ± 0.7	7 ± 0.5
BOD5	mg/L	141 ± 19	195 ± 15
COD	mg/L	1458 ± 239	410 ± 34
sCOD	mg/L	1385 ± 196	NA
Ammonia nitrogen (NH4+–N)	mg/L	914 ± 91	45 ± 11
NOx (NO2−–N and NO3−–N)	mg/L	0.60 ± 0.54	< 0.1
PO4+–P	mg/L	4.7 ± 1.49	3.69 ± 0.6
Total nitrogen (TN)	mg/L	938 ± 62	NA
Total carbon (TC)	mg/L	695 ± 75	NA
Total organic carbon (TOC)	mg/L	675 ± 65	NA
BOD5/COD		0.095 ± 0.007	0.46 ± 0.05

Parameters are presented as average values with the corresponding standard deviation of the mean. NA = Not analyzed

Pretreatment of MLL

The raw MLL was aerobically pretreated to reduce NH₄⁺–N concentration in an SBR containing activated sludge, which was obtained from the local wastewater treatment plant (South End Water Pollution Control Centre, Winnipeg, MB, Canada). The feed of the SBR contained 70–80% MLL dose (v/v) with the primary effluent, which was also collected from the aforementioned wastewater treatment plant. The SBR was made up of a 5-L glass cylinder with a 3-L working volume and contained 6000–8000 mg/L volatile suspended solids (VSS) (Fig. 1). The reactor worked for 8 h per cycle with the provision of 7.5 h aeration period per cycle at the airflow rate of 2 L/min. Half of the working volume (1.5 L) was discharged following 5 min of settlement at the end of each cycle and same amount of raw MLL was pumped into the reactor for the next cycle. The effluent having different NH₄⁺–N concentration was collected from the SBR depending on reactor’s operating condition and considered as the pretreated MLL and stored at 4 °C for couple of days until it was used for the following fungal treatment process.

Fig. 1.

Schematic diagram of the aerobic SBR

Cultivation of fungi and batch experimental tests

Penicillium spp. was previously isolated from the MLL [19] and cultured onto malt extract agar plates (MEA) following Bardi et al., [18] at 25 °C in darkness containing 2 g peptone and 20 g each of malt extract, agar, and glucose in 1 L solution and subsequently stored at 4 °C. The fungal colony of MEA plates was cut into small pieces, about 0.1 × 0.1 cm², using a sterile scalpel and cultivated into sterilized GLY broth containing 5 g glucose and 1.9 g yeast extract in 1 L solution [18]. The mycelium was harvested and washed with sterile deionized water to remove residual GLY media components, which was used for batch experimental tests. The pH of the pretreated MLL ranged from 7.6 to 8.1 and pH was adjusted to 5 using 10% diluted H₂SO₄ solution before transferring the harvested mycelium into the resulting solution. Approximately 10 g of harvested mycelium (wet weight) was added to 100 ml pretreated MLL in 500 ml flasks for the fungal treatments except the controls and all trials were triplicated. The flasks were placed on a shaker at 150 rpm to facilitate uniform mixing at 25 °C for 2 weeks. Samples were collected before adding mycelium, immediately after treatment preparation, after 24 h, and every second day for the entire experimental duration.

Continuous fungal bioreactor

A packed bed bench-scale bioreactor was constructed using a 5 L cylinder having a working volume of 4.5 L (Fig. 2). Penicillium spp. was cultivated and immobilized onto polyethylene foam (PUF) cubes following the aforementioned fungal cultivation procedure for the fungal bioreactor. A perforated polyethylene cage coupled with the shaft of the motor rotating at 5 rpm was used to contain fungal colonized PUF cubes (nearly 60 cubes). The pH controller was set to 5 to regularly adjust the reactor pH with 10% diluted H₂SO₄ solution. Continuous aeration (2 L/min) was maintained through a diffuser at the bottom connected with an air pump. The pretreated MLL was used as the feed and two pumps were used for the inlet and outlet. The reactor was operated at 6 h cycles at the room temperature (approximately 25 °C) with 1 h discharge, 1 h feeding, and 4 h of lag time between the feeding and discharge. The inlet and outlet pumps’ flowrate were adjusted to 9.4, 6.3, 4.7, and 3.8 mL/min for 48, 72, 96, and 120 h of HRT, respectively. The inlet and outlet samples were collected daily for analysis.

Fig. 2.

Schematic of the packed bed bioreactor. 1. Inlet feed tank, 2. Outlet discharge tank, 3. Aerator, 4. pH Controller, 5. pH probe, 6. acid solution container, 7. Mixer motor, 8. Motor shaft, 9. cage with immobilized fungi in PUF cubes, 10. Air diffuser

Analytical procedures

The amount of BOD₅, COD and sCOD in the samples was determined using the Standard Methods for Examination of Water and Wastewater (SMEW, 18th Edition) [30]. COD represents unfiltered samples and sCOD represents the COD of filtered samples that was measured using 1–5 μm plain filter paper. A Hach kit TNT 827 protocol was followed to determine total nitrogen (TN). A spectrophotometer (DR2800, HACH Canada Ltd., London, ON, Canada) was used to measure COD and TN. The particulate COD (pCOD) was obtained from the difference of COD and sCOD. The pH was measured using a pH meter (Oakton, Eutech Instruments, Singapore). A flow inject analyzer (QuikChem 8500, LACHAT Instruments, Loveland, CO, USA) was used to measure the concentrations of ammonia nitrogen (NH₄⁺–N), nitrate nitrogen (NO₃⁻–N), nitrite nitrogen (NO₂⁻–N), and orthophosphate (PO₄⁺–P). The TC and TOC contents were measured using a high-temperature TOC analyzer (FormacsHT TOC Analyzer, Skalar, Tinstraat, Breda, The Netherlands) coupled with an autosampler (LAS-160) and nitrogen detector from the same supplier.

A microplate reader spectrophotometer (Synergy 4, BioTek Instruments Inc., Winooski, VT, USA) was used to detect enzymatic activities at 25 °C. Lac activity was determined as described in Bourbonnais and Paice [31] from the oxidation of ABTS (2,2′-azino-bis-(3-ethylbenozthiazoline-6-sulfonic acid)) (specific activity of 36,000 M⁻¹ cm⁻¹) in 100 mM sodium citrate buffer at pH 3 and 420 nm. MnP activity was determined by the method described by Vyas et al. [32] from the oxidation of 25 mM DMAB (3-methyl-2-benzothiazolinone hydrazone hydrochloride) and 1 mM MBTH (3-dimethylaminobenzoic acid) (specific activity of 32,900 M⁻¹ cm⁻¹) by adding 4 mM H₂O₂ in 100 mM succinate lactate and 4 mM MnSO₄ buffer at pH 4.5 and 590 nm. LiP activity was determined following Tien and Kirk [33] from the oxidation of veratryl alcohol (3,4-dimethoxybenzyl alcohol) (specific activity of 9300 M⁻¹ cm⁻¹) by adding 0.4 mM H₂O₂ in 100 mM sodium tartrate buffer at pH 3 and 310 nm. The color removal or decolorization percentage was determined from the differences in spectrum absorbance of initial and final values during the treatment using the above-mentioned microplate reader spectrophotometer in the visible range of 380–740 nm.

Statistical analysis

The statistical analysis was carried out using SAS software (SAS 9.4, SAS Institute Inc., Cary, NC, USA). Two-way analysis of variance (ANOVA) with PROC GLIMMIX procedure and repeated measure analysis was used to evaluate the interaction effect between the inoculum time and ammonia nitrogen concentration of the pretreated MLL on the COD/sCOD removal efficiency and enzymatic activity in the batch experimental tests. The minimum values of Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Generalized Chi-Square were considered to obtain the covariance structure. One-way ANOVA analysis was carried out to evaluate the effect of HRT on contaminants’ (COD, sCOD, TC, TOC and color) removal efficiencies and enzymatic activities (Lac, LiP, MnP) from the fungal bioreactor. Possible outliers and homogeneity of variance was carefully observed, and normality of residual distribution was confirmed (Shapiro–Wilk test statistics: W > 0.9 and probability, p > 0.05) for each dataset. For the least squares multiple means comparisons, Tukey’s adjustment and denominator degrees of freedom were considered.

Results and discussion

MLL characterization and selection criteria for secretion of enhanced fungal enzymes

The MLL contains many of the contaminants in excess amount and many of them exceeded the Manitoba Water Quality Standards, Objectives, and Guidelines [34] (Supporting document Table 4), which emphasize the urgent need to adapt effective treatment options for the remediation of these diverse refractory contaminants.

The equivalent amount of ammonia nitrogen (914 ± 91 mg/L) and total nitrogen (938 ± 62 mg/L) in MLL suggest that nitrogen is present mostly in ionized form (NH₄⁺-N). Moreover, restricted nitrogen levels in the environment enhance lignocellulosic enzyme production. For instance, Ellouze et al. [25] and Ürek and Pazarlioǧlu [35] reported that the optimum value for lignocellulosic enzyme production as 20–22 mM nitrogen, which is equivalent to 360–396 mg/L NH₄⁺–N. In this current study, the aerobic activated sludge pretreatment process was successfully utilized to reduce the NH₄⁺–N concentration from 914 to 352 mg/L to enhance the subsequent fungal treatment process which is discussed in the next section.

It was assumed that no additional nutrients were required for the biological fungal treatment process except oxygen and some easily accessible carbon source since some minerals and essential nutrients are already present in MLL with high concentration such as Na (>1500 mg/L), K (>590 mg/L), Ca (>190 mg/L), Mg (>500 mg/L), Mn (>0.25 mg/L), Fe (>1.95 mg/L), Chloride (>1700 mg/L), N (>900 mg/L), and P (>5 mg/L) (Supporting document Table 4). The addition of easily accessible carbon source in the form of wastewater (20% - 25% v/v) could be beneficial for the biodegradation process. These findings are in agreement with Saetang and Babel [17] who reported that fungal growth cannot be fully supported by the available organic compounds in MLL and therefore, additional carbon source significantly enhances fungal growth.

Effect of NH₄⁺–N concentration in pretreated MLL on fungal treatments

The effect of the ammonia nitrogen concentrations in the pretreated MLL on the percentage of COD/sCOD removal efficiency in accordance with enzymatic activities were studied via batch experimental tests. The NH₄⁺–N concentration of raw MLL was 914 mg/L, and the collected pretreated MLL concentrations were 0, 105, and 352 mg/L. The experiments were carried out with MLL containing initial NH₄⁺–N concentrations of 0, 105, 352, and 914 mg/L (which have denoted as L1, L2, L3, and L4 respectively) and inoculated with Penicillium spp. The control included raw MLL without fungal inoculation.

No enzymatic activity was observed from the control (with L4 leachate) treatment and therefore, the data was not presented. The enzymatic activity was insignificant at 24 h for all treatments (Fig. 3). It is noticeable that at 24 h, L1 and L2 showed significantly higher COD/sCOD reduction as compared to the control suggesting that better acclimatization occurred with lower NH₄⁺–N concentration (Table 2). At 96 h, no significance difference was observed among L1, L2, and L3 and after the treatment duration of 168 h, L3 expressed significantly higher COD/sCOD reduction than any other treatments. The study revealed that the fungal treatment containing 352 mg/L of ammonia nitrogen concentration (L3) exhibited higher removal efficiencies of COD (41%) and sCOD (42%) at the end of 264 h of treatment duration. The enzymatic activity was also highest with L3 leachate at the end of the treatment duration such as 193, 37, and 25 U/L of Lac, LiP and MnP, respectively (Fig. 3). The MLL with lower NH₄⁺–N concentration (L2: 105 mg/L) reduced the COD and sCOD removal to 32 and 33 percentage, respectively. At this same concentration (L2), the Lac, LiP, and MnP secretion was significantly reduced in contrast with L3 and inhibited by 20, 27, and 21 percentage, respectively. Similarly, L1 (0 mg/L NH₄⁺–N) treatment caused further reduction in COD and sCOD to 26 and 27 percentage as compared to L3 (352 mg/L NH₄⁺–N) and inhibited the secretion of all three enzymes by approximately 40%. Conversely, L4 treatment (higher concentration of NH₄⁺–N: 914 mg/L) also caused a drop in COD/sCOD removal to 17% and inhibited all three enzymes secretion by approximately 70% (Fig. 3).

Fig. 3.

Effect of ammonia nitrogen concentration of pretreated MLL on fungal enzymatic activities. Note: Two-way ANOVA was carried out for the interaction effect of treatments and time on enzymatic activities. F = fungal Penicillium spp. L1 – L4 represent pretreated and raw MLL containing 0, 105, 352, and 914 mg/L of NH₄⁺–N concentration. The number above the bar represents mean value. Sharing same letter among treatments indicates no significant difference (p < 0.05)

Table 2.

Effect of NH₄⁺–N concentration in the pretreated (inoculated by Penicillium spp.) MLL on the percentage of COD/sCOD removal efficiencies

Time (h)	24	96	168	216	264	24	96	168	216	264
Treatments	% COD					% sCOD
CT(L4)	3 m	4 lm	5 lm	5 lm	6 lk	1 n	3 lmn	3 lmn	6 klm	6 jklm
F(L1)	6 l	16 ji	21 hg	25 fg	26 fe	7 jkl	17 gh	24 ef	26 de	27 de
F(L2)	7 lk	18 hi	27 fe	31 dc	32 c	8 jk	21 fg	28 dc	32 bc	33 b
F(L3)	2 m	17 ji	29 de	37 b	41 a	2 mn	18 gh	32 bc	39 a	42 a
F(L4)	4 lm	9 k	14 j	17 ji	17 i	3 lmn	10 ij	14 hi	16 h	17 gh

Two-way ANOVA was carried out for the interaction effect of treatments and time on COD/sCOD removal efficiency. Sharing same letter among treatments indicates no significant difference (p < 0.05). CT = control (without fungi), F = fungal Penicillium spp. L1 – L4 represent pretreated and raw MLL containing 0, 105, 352, and 914 mg/L of NH₄⁺–N concentration

Ellouze et al. [25] reported similar observations in terms of fungal growth and enzymatic activities. They reported 360 mg/L NH₄⁺–N concentration (20 mM nitrogen) as the optimum condition for fungal growth and exhibited reduction in LiP and MnP activities by 50 and 60 percentage for P. chrysosporium, and reduction in Lac activity by 30% for T. trogii with 673 mg/L NH₄⁺–N (2 g/L of NH₄Cl) concentration as compared to the optimum condition. Ürek and Pazarlioǧlu [35] also reported highest MnP activity at 396 mg/L NH₄⁺–N (22 mM nitrogen) concentration. These observations are also in agreement with many white-rot-fungi behavior suggesting that the secretion of lignocellulosic enzyme enhances at limited nitrogen levels [36]. However, similar observations of enzymatic inhibition at very lower ammonia nitrogen concentration (0–396 mg/L NH₄⁺–N) were not reported by Ellouze et al. [25] and Ürek and Pazarlioǧlu [35]. The possible reason could be due to the difference of fungal culture media. The aforementioned studies were conducted with the addition of NH₄Cl in solid-state culture media [25, 35] whereas the our current study was conducted with MLL by reducing ammonia nitrogen concentration through aerobic activated sludge pretreatment process.

Effect of aerobic activated sludge pretreatment on MLL

The initial volume in the aerobic activated sludge SBR was 20% (v/v) MLL, which was successively increased by 5–10% over the next 2–3 months to acclimatize with 100% MLL. With the 100% MLL in feed, the SBR achieved 11–13 g/L mix liquor total suspended solids (MLTSS), 6–8 g/L mix liquor volatile suspended solid (MLVSS), and SVI < 100, which suggests good settleability [37]. At stable operating condition, 99–100% NH₄⁺–N removal was achieved irrespective of the MLL dose in the feed (Fig. 4). High NH₄⁺–N removal efficiency from leachate has also been reported in other studies. For instance, Xu et al. [38] reported 96.7% NH₄⁺–N removal efficiency from leachate (BOD₅/COD ratio of 0.15 with an initial NH₄⁺–N of 1451 mg/L) with relatively higher effluent NH₄⁺–N (48 mg/L). In another study, Wang et al. [39] also achieved 99% NH₄⁺–N removal rate by introducing 30 min intermittent aeration followed by 24 h SBR cycle.

Fig. 4.

COD and NH₄⁺-N concentration of aerobic activated sludge pretreated MLL. Note: Paired t-test was done between influent and effluent for COD and NH₄⁺-N reduction. Sharing same letter between influent and effluent indicates no significant difference (p < 0.05). Error bar represents standard error of the mean

In order to obtain the pretreated MLL for the following fungal treatment process, effluent with three levels of NH₄⁺–N concentration (0, 105, and 352 mg/L) was collected. The influent and effluent characteristics of the SBR for these three categories of pretreated MLL (L1, L2 and L3) illustrate that the sole aerobic activated sludge process is not efficient enough to remove COD (effluent COD: 1363, 1397, 1561 mg/L for L1, L2, and L3 respectively), although the pared t-test between influent and effluent shows significantly different (Fig. 4). Marañón et al. [40] also demonstrated that sole activated sludge process could not adequately remove COD from old leachate. The possible reason could be due to the low biodegradability fraction and diverse toxicity of the recalcitrant components in MLL [4, 41]. For all three categories of SBR effluents, approximately 10–12 percentage of COD removal rate was exhibited with 80% MLL feed (Fig. 4). This observation agreed the previous findings [6] that used the same MLL and reported a 14% COD reduction via aerobic activated sludge SBR fed with 40–65% MLL.

Contaminants’ removal and enzymatic activities in fungal bioreactor

The flask level sequential treatment revealed that the maximum COD (41%) and sCOD (42%) reduction occurred when MLL was first treated with aerobic activated sludge (at 352 mg/L NH₄⁺–N concentration) following the fungal (Penicillium spp.) treatment. The final scale-up studies were carried out following this sequence and the performance of fungal bioreactor was evaluated based on the percent removal efficiencies of COD, sCOD, TC, TOC, and color from the pretreated MLL. In the fungal bioreactor, COD and sCOD removal efficiencies were further enhanced to 48 and 50 percentage (max. 52 and 54 percentage respectively) on day 87 with HRT 120 h as compared to the batch tests of 41 and 42 percentage, respectively (Fig. 5). Moreover, an additional 12% COD removal was achieved by the aerobic activated sludge SBR during the pretreatment process. This observation is analogous with the finding of Ghosh and Swati [8], where a 71% COD reduction in flask test was further increased to 79% in a sequential bioreactor. The sequential treatment with two or more micro-organisms is beneficial as the first treatment initiates the catabolic reactions while the subsequent treatment activates the remaining metabolic pathways for complete mineralization [42]. On the other hand, the feed for the sequential treatment process was prepared with 80% raw MLL and 20% municipal sewer wastewater, where wastewater contributed approximately 80–90 mg/L of COD. The higher BOD/COD ratio of wastewater (≈ 0.5) as compared to MLL (< 0.1) indicated that wastewater added little amount of easily accessible carbon source for the microorganisms which might further enhance the microbial activity.

Fig. 5.

Percent removal efficiencies of COD, sCOD, Color, TC, and TOC from the fungal bioreactor and their corresponding Lac, LiP and MnP enzymatic activities

The fungal bioreactor demonstrated stable removal efficiencies and enzymatic activities since the startup of the bioreactor and operation was continued for 150 days with varying HRT (Fig. 5). The startup HRT (72 h) was subsequently increased to 96 h and 120 h, and finally reduced to 48 h to observe the effect of HRT on contaminants’ removal efficiencies and enzymatic activities. The system took several days to adjust to the change in HRT and regain stable removal efficiencies and enzymatic activities and these transition phases were indicated as 72 (T), 96 (T), 120 (T), and 48 (T) for the corresponding HRT of 72, 96, 120, and 48 h, respectively. The transition phase was longer, approximately 8 days, during the change in HRT from 120 h to 48 h, while it was about 5 days in all other cases. With the increase in HRT, the COD and sCOD reduction was significantly increased due to the longer contact time between MLL and fungus. Specifically, the average COD reduction of 31, 35, 43, and 48 percentage and sCOD reduction of 33, 35, 44, and 50 percentage were recorded at HRT of 48, 72, 96, and 120 h, respectively. The Lac and LiP enzymatic activities were also significantly higher in most cases, while MnP activities were statistically insignificant and low values were recorded over the entire operational period (Fig. 5 and Table 3). In terms of TC, TOC, and color removal efficiencies and Lac and MnP enzymatic activities, there was no significant difference between the HRT 96 and 120 h suggesting that 96 h could be considered as optimal contact time between MLL and fungus in the bioreactor. The maximum TC, TOC and color removals were 60, 58, and 75 percentage on day 99 while the maximum Lac, and LiP activities were 149 and 27 U/L on day 87 and MnP activity of 16 U/L was recorded on day 99 at HRT 120 h (Fig. 5).

Table 3.

Effect of HRT on contaminants’ removal efficiency and enzymatic activity in fungal bioreactor

Operating days	HRT (h)	Percent removal					Enzymatic activities (U/L)
		COD	sCOD	TC	TOC	Color	Lac	LiP	MnP
1–4	72 (T)	15 E	15 E	13 C	11 D	7 E	13 C	4 C	0 A
5–37	72	35 C	35 C	38 B	36 CB	47 C	66 BC	10 C	5 A
38–43	96 (T)	40 BC	38 C	NA	NA	NA	82 AB	13 BC	6 A
44–72	96	43 B	44 B	50 A	50 A	68 AB	93 A	14 B	5 A
73–79	120 (T)	43 B	48 AB	48 A	45 AB	66 AB	117 A	17 AB	5 A
80–105	120	48 A	50 A	54 A	52 A	73 A	143 A	22 A	8 A
106–114	48 (T)	37 C	36 C	26 BC	24 CD	56 CB	80 ABC	12 BC	6 A
115–148	48	31 D	33 D	22 C	21 D	32 D	34 C	6 C	2 A

The COD, sCOD, TC, and TOC removal efficiencies followed a similar trend during all phases of HRT except at 48 h, when the COD/sCOD reduction was approximately 10% higher than the TC/TOC due to the higher variation in TC/TOC (standard deviation for COD/sCOD <2, while for TC/TOC ≈ 10). The color removal efficiency was much higher in contrast to the COD, sCOD, TC, and TOC at HRT of 72, 96, and 120 h. For instance, color removal was almost 1.5-fold higher than COD during those periods. Similarly, Lu et al. [43] reported higher percentage of color removal (85.5%) than COD reduction (54.6%) in a continuous bioreactor.

A previous continuous bioreactor study with Bjerkandera adusta to treat MLL from the same landfill (and same well # 24) [18] resulted in average sCOD reduction of 48% during the first stage with 0.5 g/L of glucose addition at the startup and HRT of 72 h whereas 16% sCOD reduction was recorded from MLL alone. The sCOD reduction decreased to 14% when startup glucose was depleted, and no additional glucose was added during the second stage; which increased the sCOD reduction to 24% while 0.5 g/L glucose was again added at the third stage. During the entire study period, no lignocellulosic enzymatic activity was reported and COD/sCOD reduction trend was not stable. In the current study, higher and steady sCOD removal efficiency and lignocellulosic enzymatic activities suggest better acclimatization for Penicillium spp., which could lead to higher performance. In a separate study, Ren [6] reported 43 and 31 percentage COD reduction from the same MLL by using granular activated sludge SBR fed with 45 and 90 percentage MLL dosage, respectively, which suggests that sequential treatment of aerobic granular sludge and lignocellulosic fungi is a feasible approach to enhance COD removal efficiency from MLL. The result of the current study is promising considering the nature of the MLL with low biodegradability and presence of toxic refractory contaminants. Moreover, a maximum COD reduction of 64% was achieved from the raw MLL considering the 12% COD reduction from pretreatment process. Further exploratory study to understand the chemical characteristics of the degraded components could be beneficial to optimize the performance of the current process.

Conclusion

This study investigated the lignocellulosic enzymatic activities and contaminants’ removal efficiencies (COD, sCOD, TC, TOC, and color) from MLL through batch experimental tests and sequential treatment utilizing an aerobic activated sludge SBR (pretreatment) and a continuous backed bed fungal (Penicillium spp.) bioreactor. The maximum COD reduction (41%) was observed from the pretreated MLL with 352 mg/L NH₄⁺–N concentration through batch tests and COD reduction efficiency was decreased regardless of the change in NH₄⁺–N concentration. The optimal treatment condition (352 mg/L NH₄⁺–N) exhibited 193, 37, and 25 U/L of Lac, LiP and MnP enzymatic activity, respectively. The maximum COD reduction efficiency was further increased to 52% in the continuous bioreactor fed with pretreated MLL at HRT 120 h. The total COD reduction of 64% was obtained from raw MLL including the 12% COD reduction from pretreatment process. The fungal bioreactor exhibited 60, 58, and 75 percentage of TC, TOC, and color removals efficiency in accordance with the 149, 27, and 16 U/L of Lac, LiP, and MnP enzymatic activities. The findings of this treatment process seem to be promising and the integration of advanced and sustainable technological approaches could be evaluated to enhance the removal of refractory contaminants from the MLL.

Abbreviations

ANOVA

Analysis of variance

BOD

Biochemical oxygen demand

BRRMF

Brady road resource management facility

Ca

Calcium

COD

Chemical oxygen demand

CT

Control

F

Fungus

Fe

Iron

h

Hour

HRT

Hydraulic retention time

K

Potassium

L

Liter

Lac

Laccase

LiP

Ligni-peroxidase

mg

Milligram

Mg

Magnesium

MLL

Mature landfill leachate

MLTSS

Mix liquor total suspended solids

MLVSS

Mix liquor volatile suspended solid

mM

Millimolar

Mn

Manganese

MnP

Manganese-peroxidase

N

Nitrogen

Na

Sodium

nm

Nanometer

P

Phosphorus

pCOD

Particulate chemical oxygen demand

PUF

Polyethylene foam

rpm

Revolution per minute

SBR

Sequencing batch reactor

sCOD

Soluble chemical oxygen demand

SVI

Sludge volume index

TC

Total carbon

TN

Total nitrogen

TOC

Total organic carbon

U/L

Unit per liter

v/v

Volume / volume

VSS

Volatile suspended solids

WWTP

Wastewater treatment plant

Authors’ contributions

Mr. Mofizul Islam performed the overall entire experimental study including experimental design, reactors operation, samples collection and analysis, data interpretation, and manuscript writing. Mr. Qian Xu contributed to design and operating of the aerobic activated sludge bioreactor. Dr. Qiuyan Yuan supervised all aspects of the experimental study and contributed to improve the writing of the manuscript.

Funding information

The financial support was appropriated to the Graduate Enhancement of Tri-Council Stipends (GETS) and Create-H2O programs, University of Manitoba for this research from the Natural Sciences and Engineering Research Council of Canada (NSERC).

Availability of data and material

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Compliance with ethical standards

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.