The characterization of dissolved organic matter in alkaline fermentation of sewage sludge with different pH for volatile fatty acids production

Water Research 164 (2019) 114924

Contents lists available at ScienceDirect

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

The characterization of dissolved organic matter in alkaline fermentation of sewage sludge with different pH for volatile fatty acids production Sijia Ma, Haidong Hu, Jinfeng Wang, Kewei Liao, Haijun Ma, Hongqiang Ren* State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing, 210023, Jiangsu, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 June 2019 Received in revised form 26 July 2019 Accepted 27 July 2019 Available online 29 July 2019

Sewage sludge (SS) alkaline fermentation (especially at pH 10) can efficiently enhance volatile fatty acids (VFAs) production. VFAs are considered an excellent carbon source for the biological nutrient removal (BNR) process. Dissolved organic matter (DOM) in fermentation liquid is the direct substrate used for producing VFAs and can greatly influence the effluent quality of BNR process. However, knowledge of DOM characteristics in sludge alkaline fermentation is limited. This study focused on the functional groups, fluorescent components and molecular features of DOM as well as molecular weight of proteins in SS alkaline fermentation (at pH uncontrolled, 7, 8, 9 and 10). Results showed a significantly improved generation of tryptophan-like and tyrosine-like substances as well as molecular weight <1 kDa proteins (381.5 ± 38.4 mg/L) was observed at pH 10 (p < 0.05). Further analysis of DOM molecular characteristics indicated that pH 10 resulted in the highest molecular diversity and the generation or degradation of easily biodegradable lipid-like and proteins/amino sugars-like formulas. The improved solubilization of DOM contributed to VFAs production. Meanwhile, increasing pH to 10 also promoted the release of hardbiodegradable organic matter, e.g., humic-like and lignin-like substances. Additionally, a high diversity of resistant N-containing organic molecules was generated at pH 10. Fermentation of SS at pH 10, is favored to enhance VFAs production and, can also result in a higher content of refractory DOM. This study helps to achieve a comprehensive understanding of SS alkaline fermentation and provides fundamental information for further treatment. © 2019 Elsevier Ltd. All rights reserved.

Keywords: Sludge alkaline fermentation VFAs production Dissolved organic matter Molecular characteristic Refractory molecules

1. Introduction The treatment of sewage sludge (SS) has remained a thorny issue until now. Because of its high organic matter content, anaerobic digestion (AD) is widely used for SS treatment and can produce volatile fatty acids (VFAs) or methane (Appels et al., 2008; Chen et al., 2007). VFAs can act as superior carbon sources for biological nutrient removal (BNR) process. AD of SS for VFAs generation is an attractive strategy for producing carbon sources for BNR process (Li et al., 2011; Yu et al., 2008). Prior to VFAs production, hydrolysis of insoluble organic matter to dissolved organic matter (DOM) and acidification of the generated DOM to VFAs are necessary steps (Chen et al., 2017; Ma et al., 2019). The rate limiting step is hydrolysis. Therefore, enhancing the hydrolysis and

* Corresponding author. E-mail address: [email protected] (H. Ren). https://doi.org/10.1016/j.watres.2019.114924 0043-1354/© 2019 Elsevier Ltd. All rights reserved.

acidification of SS is of great interest. Alkaline fermentation (especially pH 10) can greatly improve the hydrolysis and acidification of SS, resulting in a substantially enhanced VFAs production (Liu et al., 2018; Wang et al., 2019). Many studies have been performed to investigate the mechanism of improved hydrolysis and acidification by SS alkaline fermentation (SAF) (especially at pH 10) (Chen et al., 2017; Li et al., 2011; Wang et al., 2019; Zhang et al., 2009; Zheng et al., 2013). Generally, these studies point out that SAF can break extracellular polymeric substances (EPS) matrix and increase DOM concentration as well as inhibit methanogens, especially increase protein (PN) concentration (Chen et al., 2017; Maspolim et al., 2015), which significantly favors VFAs production. The increase of DOM is one of the most important reasons for VFAs production (Chen et al., 2007; Wang et al., 2019). In hydrolysis stage, complex DOM is produced (Chen et al., 2017), and DOM is the direct substrate for VFAs generation (Yekta et al., 2012). Not only the DOM concentration but also the characteristics of DOM significantly influence its transformation to

2

S. Ma et al. / Water Research 164 (2019) 114924

List of abbreviations SS VFAs BNR DOM AD SAF EPS PN DOC MW FTICR-MS TB-EPS LB-EPS SB-EPS WWTP TS VS

Sewage sludge Volatile fatty acids Biological nutrient removal Dissolved organic matter Anaerobic digestion Sewage sludge alkaline fermentation Extracellular polymeric substances Proteins Dissolved organic carbon Molecular weight Fourier-transform ion cyclotron resonance mass spectrometry Tightly bound extracellular polymeric substances Loosely bound extracellular polymeric substances Soluble extracellular polymeric substances Wastewater treatment plant Total solids Volatile solids

VFAs (Chen et al., 2017; Kamjunke et al., 2017). Moreover, information about the features of refractory DOM produced during SAF should be given because the refractory DOM may adversely affect the effluent quality of BNR process or following treatment (Azman et al., 2017; Cao et al., 2019a, 2019b; Maizel and Remucal, 2017). However, our understanding of DOM characteristics in SAF is still limited. Chen et al. (2017), characterized the DOM in SAF with stepwise pH increase and found that increasing pH led to an increase of both hydrophobic and hydrophilic DOC, such as building blocks, high molecular weight (MW) and low MW protein. Recently, Cao et al. (2019b) reported that building blocks were the dominant refractory DOM in effluent of sludge alkaline fermentation liquid driven denitrification. The complex and diverse SAFdriven DOM requires a multi-method analytical characterization. Detailed information on DOM characteristics and DOM transformation is essential to obtaining a better understanding on the mechanism of improved hydrolysis/acidification by SAF and helps identify potential implications when applying the fermentation liquid to the BNR process. Fourier-transform ion cyclotron resonance mass spectrometry (FTICR-MS), can provide precise molecular formulas in the complex DOM matrices. The FTICR-MS gives novel insight and greater recognition of the molecular characteristics of complex DOM (Osborne et al., 2013; Valle et al., 2018). Proteins (PN) are the principal part of DOM and intensive PN degradation is the leading cause of enhanced VFAs production in SAF (Wu et al., 2010; Zheng et al., 2013). Macromolecular PN can be hydrolyzed into micromolecular PN at alkaline conditions. Then, the anaerobes can directly take up organic matter that is smaller than 1 KDa for VFAs production. Chen et al. (2017) found that SAF with a stepwise pH increase enhanced the solubilization of both low and high MW protein. Unfortunately, the study did not determine the content of MW < 1 KDa protein. Recently, Cao et al. (2019b) reported that high MW PN were potential refractory DOM. Moreover, types of soluble proteins can also be influenced by different pH values (Xiao et al., 2017). Those characteristics of PN would largely affect its application as feed stuff. Nevertheless, little is known about the characteristics of soluble PN in SAF with different pH values. The DOM can result from both EPS and intracellular organics in SS. The transformation of organics from pellet to tightly bound EPS (TB-EPS) and loosely bound EPS (LB-EPS) as well as soluble EPS (SB-

TCOD SCOD SRT HRT GC PS COD FTIR EEM PARAFAC SPE H/C O/C C1 C2 C3 C4 C5 C6

Total chemical oxygen demand Soluble chemical oxygen demand Solids retention time Hydraulic retention time Gas chromatograph Polysaccharide Chemical oxygen demand Fourier Transform infrared spectroscopy Three-dimensional excitation-emission matrix Parallel factor analysis Solid phase extraction The molar ratio of hydrogen to carbon The molar ratio of oxygen to carbon Component 1 of the fluorescent components Component 2 of the fluorescent components Component 3 of the fluorescent components Component 4 of the fluorescent components Component 5 of the fluorescent components Component 6 of the fluorescent components

EPS) represent the release characteristics of SS (Li et al., 2014b). It is well known that EPS is the dominant component of organics in SS. Yu et al. (2008) found that the transfer of PN from TB-EPS to SB-EPS is accompanied by the degradation of PN. However, previous studies mainly used batch experiments or characterized the EPS in terms of PN and PS content (Jin et al., 2016). The chemical characteristics of EPS are known to affect its solubilization and biodegradability. Therefore, investigations on EPS features are helpful for understanding the improved hydrolysis and acidification by SAF. This study investigated the characteristics of DOM in sludge alkaline fermentation with different pH values for VFAs production. The functional groups, fluorescent components and molecular characteristics of DOM, and molecular weight distribution of PN were analyzed. The changes in EPS characteristics were also studied. The results would deepen our understanding of enhanced sludge hydrolysis and acidification by alkaline fermentation. In addition, the potential refractory DOM characteristics were also identified and thus provide essential information for further treatment. Together, this paper helps to get a better insight into alkaline fermentation of SS for VFAs production. 2. Materials and methods 2.1. Sludge characteristics and semi-continuous fermentation reactors operation The SS used in this study was collected from the secondary clarifier in a municipal wastewater treatment plant (WWTP) (Nanjing, China). Raw SS was filtered through a 1.6 mm sieve to remove debris and stored at 4
C. Seeding sludge was collected from a mesophilic anaerobic digester and detailed information of the digester is presented in our previous study (Ma et al., 2019). The features of raw SS and inoculum were: pH 6.7 ± 0.1 and 7.1 ± 0.1; total solids (TS): 19.36 ± 0.68 and 17.94 ± 0.41 g/L; volatile solids (VS): 11.91 ± 0.37 and 9.23 ± 0.22 g/L; soluble chemical oxygen demand (SCOD): 124.7 ± 14.34 and 398.3 ± 31.25 mg/L, and total chemical oxygen demand (TCOD): 17.98 ± 1.02 and 13.55 ± 1.14 g/L. Five semi-continuous fermentation reactors were used to carry out the alkaline fermentation of SS, which were made by plexiglass and each reactors had a total and working volume of 800 mL and 400 mL. Many studies have reported that pH 11 was adverse to the

S. Ma et al. / Water Research 164 (2019) 114924

further VFAs production (Chen et al., 2007; Maspolim et al., 2015). Therefore, pH 10 was the highest pH value in this study. The reactors were referred to as pH uncontrolled, pH 7, pH 8, pH 9 and pH 10, respectively. The pH was maintained by adding 2 M sodium hydroxide three times a day with the help of a FE20 pH meter and probe (Mettler Toledo, LLC. Toledo, OH, USA). All reactors were stirred at 100 rpm and maintained at 37
C through a shaking cultivating chamber. Seeding sludge and raw SS was 1:1 (VS:VS). Fifty mL of digestate was discharged and 50 mL of raw SS was added to the reactors everyday, resulting in a solids retention time (SRT) of 8 d (Li et al., 2014b) and an organic loading rate of 1.49 g VS/L$d. After raw SS was added and pH was adjusted each time, N2 was used to purge the reactors for 2 min. Biogas was collected using air bags. 2.2. Analytical methods TS, VS, TCOD and SCOD were determined using the Standard Methods (APHA, 2005). Methane, H2 and VFAs concentration and compositions were analyzed using GC7890A gas chromatograph (Agilent Technologies, Santa Clara, CA, USA) according to our previous publication (Ma et al., 2019). Soluble PN and PS were detected by using Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, Rockford, IL, USA) and the phenol sulfuric acid method, respectively. VFAs, PN and PS was transferred to COD (mg/L) based on the following coefficients: 1.50 g COD, 1.06 g COD, 1.07 g COD, 1.51 g COD, 1.81 g COD, and 2.04 g COD of per gram PN, PS, acetate, propionate, butyrate, and valerate, respectively (Chen et al., 2017). All the reactors were operated for 80 days (10 SRTs). SCOD was detected every three days. After operating the reactors for >6 SRT, samples (3) were collected and analyzed. Among them, triplicate samples were used for analysis of the soluble PN molecular weight distribution. All the samples were centrifuged at 4000 g for 10 min; then, were collected and filtrated through 0.45 mm fiber filter. All the samples were collected before adding raw SS. 2.3. Analysis of DOM characteristics Freeze-dried DOM and EPS were used to detect the FTIR spectrum by a Nexus870 FTIR spectrometer (Thermo Fisher Scientific, Ashville, NC, USA) in the range of 400e4000 cm1 and resolution of 4 cm1. Three-dimensional excitation-emission matrix (EEM) was used to analyze the fluorescent components of DOM and EPS by a fluorescent spectrophotometer (F-7000, Hitachi High-Technologies Corporation, Tokyo, Japan). Furthermore, 25 DOM samples were used for the parallel factor analysis (PARAFAC). Detailed information about EEM measurement and PARAFAC are presented in the Supporting Information. For the analysis of DOM molecular characteristics, the filtrated (0.45 mm) liquid sample was adjusted to pH 2. Then, solid phase extraction (SPE) cartridges (1 g, 6 mL, functionalized styrenedivinyl-benzene polymer resin, Supelco) were used to extract the DOM. The extracted DOM was detected by a 9.4 T FTICR-MS (Bruker Daltonics, Bremen, Germany). Methods for FTICR-MS measurement can be found in our previous studies (Hu et al., 2018). All formulas were determined using Bruker DataAnalysis software (version 4.1) and MATLAB routines (Kujawinski and Behn, 2006). Other chemical constraints were described in the Supporting Information. The MW distribution of PN was classified as follows: < 1, 1e10, 10e30 and > 30 KDa. Filters (Millipore Corp. USA) with MW cutoffs of 1, 10 and 30 KDa were used to conduct the MW fractions with a 400 mL Amicon® Model 8400, stirred cell unit (MilliporeSigma Corp. Burlington, MA, USA).

3

2.4. EPS extraction EPS were extracted according to the method described in Sun et al. (2018). Briefly, 50 mL of digestate was collected and centrifuged (4000 g, 15 min), and the bulk solution was collected as the SB-EPS. These supernatant were further used to analyze the characteristics of DOM. The solids were resuspended to 50 mL using a 0.05% NaCl solution and was vortex-mixed for 1 min. The suspensions were then centrifuged (4000 g/10 min) and the bulk solution represented the LB-EPS. The residues were resuspended to 50 mL using a 0.05% NaCl solution, and were heated for 30 min at 60
C. After centrifugation (4000 g/15 min), the organic matter in the supernatant was the TB-EPS. Before analysis, all the supernatant was filtrated through 0.45 mm fiber filter. After operation for >6 SRT, triplicate samples were collected from each reactor for EPS analysis. 2.5. Statistical analysis Analysis of the Pearson correlation and the significant differences (t-test) were calculated by IBM SPSS Statistics ver. 17.0 software (IBM, Armonk, NY). A p-value of <0.05 was considered as indication significance. Origin 2015 (OriginLab, Northampton, MA, USA) was used for data visualization. 3. Results and discussion 3.1. Performance of the fermentation reactors The SCOD generated at pH 8 (2423 ± 54.83 mg/L), 9 (4533 ± 66.24 mg/L) and 10 (6448 ± 72.37 mg/L) were significantly higher than that at pH 7 and uncontrolled reactors (1403 ± 33.87 and 1315 ± 40.94 mg/L) (p < 0.05, t-test). The total VFAs production also increased from 722.3 ± 34.21 to 3762 ± 204.68 mg COD/L as pH increased from uncontrolled and 7 to 10 (Fig. 1c). The results are consistent with Yu et al. (2008) and Chen et al. (2007), implying that increasing pH to 10 can highly improve the hydrolysis and acidification efficiency of SS. As shown in Fig. S1, increasing pH also inhibited the CH4 production. At pH 10, no CH4 was produced and H2 yield was 15.34 mL/g VS. The improved hydrolysis and acidification, and the inhibited methanogenesis (Zheng et al., 2013) at alkaline pH accounted for the increase of SCOD and TVFAs. The soluble PN concentration at pH 8, 9 and 10 (387.5e1338.8 mg COD/L) were significantly higher than that at pH of uncontrolled and 7 (265.7e284.8 mg COD/L) (p < 0.05, t-test). Zhang et al. (2009) also found that by increasing the fermentation pH from 7 to 10, soluble PN increased from 41 to 163 mg COD/g VS. It means that alkaline treatment can efficiently promote the solubilization of PN in SS (Xiao et al., 2017). Soluble PS increased slightly with the increase of pH and its content was lower than the content of PN in all the reactors. It is agreed with previous studies that PN is the dominant fraction of DOM (Chen et al., 2007; Zhang et al., 2009). Unknown SCOD also increased from 152.3 to 185.6 mg/L to 304.1e975.4 mg/L as the pH increased from uncontrolled and 7 to 8, 9 and 10. Unknown SCOD may be composed of humic acid and fulvic acid substance, as well as DNA (Ma et al., 2019). The characteristics of DOM can largely affect its transformation to VFAs and degradation in following steps, which was an incentive for further analysis of DOM features in this study. 3.2. Characteristics of DOM in alkaline fermentation reactors 3.2.1. FTIR spectra of DOM FTIR spectra in the region of 1800e900 cm1 is believed to offer primary information about functional groups of DOM or EPS (Wang et al., 2018; Zhu et al., 2012). The FTIR results (Fig. 2a) contain five

4

S. Ma et al. / Water Research 164 (2019) 114924

regions correlated with different functional groups: amide I proteins (1700-1600 cm1), amide II proteins (1600-1500 cm1), carboxylic group-containing and hydrocarbon-like compounds (15001300 cm1), amide III proteins (1300-1200 cm1), and polysaccharides and nucleic acids (1200-900 cm1) (Wang et al., 2018; Badireddy et al., 2010). As is shown in Fig. 2a, diverse bands appeared, revealing the complex functional groups of DOM. The functional groups of the five DOM samples were similar. A closer observation found that a band at 1569 cm1 was only detected at pH 10, implying that the content of -CO-NH- at pH 10 was relatively higher than at other pH values (Wang et al., 2018). A weak band at 1157 cm1 was only detected at pH uncontrolled and 7, which means the DOM contained a relatively higher amount of polysaccharide than that at pH 8, 9 and 10. These results agree with previous studies (Wang et al., 2019; Zheng et al., 2013), implying that SAF can efficiently enhance the hydrolysis of proteins.

Fig. 1. Hydrolysis and acidification performance of sewage sludge with different pH: (a) SCOD concentration; (b) SCOD composition; (c) TVFAs composition (as COD). HAc, HPr, HBu, i-HBu, HVa and i-HVa represent acetate, propionate, butyrate, iso- butyrate, valerate and iso-valerate, respectively.

3.2.2. DOM fluorescent components EEM-PARAFAC modeling of the 25 samples generated a six components model (Fig. S2 and Table 1). The identified components consisted of: four protein-like components (tryptophan-like: C1, C2 and C3; tyrosine-like: C4), one fulvic-like component (C5) and one humic-like component (C6) (Ishii and Boyer, 2012; Li et al., 2014a). The fluorescence intensity of the components increased with the increase of pH (Fig. 2b), especially the protein-like components (p < 0.05). This finding agrees with the FTIR results, illustrating that increasing fermentation pH to 10 can efficiently enhance the hydrolysis of proteins. Previous studies also observed that proteinlike, fulvic-like and humic-like substances in liquid increased with the increase of pH in alkaline pretreatment of SS (Li et al., 2016; Xiao et al., 2017). The fluorescence intensity of protein-like substances was higher than that of fulvic-like and humic-like substances in all the reactors. Protein-like substances (C1, C2, C3 and C4) are the major and easily biodegradable substrate (especially compared to humic-like substances) for VFAs production. Li et al. (2014a) reported that tryptophan-like materials degraded gradually while tyrosine-like materials fluctuated over time during a 43 days's AD of dewatered SS. The accumulation of these DOM at alkaline pH might be due to the combined effect of low SRT (8 d), the highly enhanced release of protein, the possibly inhibited acidogenesis (Chen et al., 2017) and their biodegradable mechanisms (Li et al., 2014a). While fulvic-like and humic-like substances are hard biodegradable materials (Cao et al., 2019a; Zhang et al., 2016). Li et al. (2019) found that humic acids can promote acidification of SS. Polycyclic aromatic hydrocarbon can enhance acetic acid production in SS fermentation (Luo et al., 2016). Increasing fermentation pH has been shown to increase the hydrolysis of protein-like, fulvic-like and humic-like substances, which further favors VFAs production, especially at pH 10. However, the highest VFAs and SCOD production, and fulvic-like and humic-like substances were simultaneous at a fermentation pH 10. Cao et al. (2019b) found that fulvic-like and humic-like substances in fermentation liquid were difficult to be degraded in sequencing batch reactor driven denitrification process; moreover, an extended HRT also had a minor effect on the removal efficiency. The presence of this type of DOM may inhibit the N removal efficiency and deteriorate the effluent (Fernandes et al., 2014; Zhang et al., 2018). Therefore, further studies are required to determine more efficient removal of fulvic-like and humic-like substances in SAF liquid (pH 10), especially when considering the reuse of treated wastewater in WWTPs (Michael-Kordatou et al., 2015; Ni et al., 2011). 3.2.3. Molecular characteristics Van Krevelen diagrams (Fig. S3), according to the molar ratio of

S. Ma et al. / Water Research 164 (2019) 114924

5

Fig. 2. (a) FTIR spectra (1800-900 cm1) of the DOM and (b) fluorescence intensity of the six EEM components of the DOM in the reactors with different pH. C1, C2, C3 and C4 represent the protein-like components (Ex/Em ¼ 275/332, 270/340, 275/352 and 270/300 nm, respectively); C5 represents the fulvic-like components (Ex/Em ¼ 240, 310/384); C6 represents the humic-like components (255,360/456 nm).

Table 1 Description of the identified EEM-PARAFAC components. Component

Ex/Em (nm)

Description

C1 C2 C3 C4 C5 C6

275/332 270/340 275/352 270/300 240, 310/384 255, 360/456

Tryptophan-like Tryptophan-like Tryptophan-like Tyrosine-like Fulvic-like Humic-like

H/C and O/C, were used to analyze the entire formulas detected in the DOM samples (Kamjunke et al., 2017). DOM formulas were mainly located in the region of H/C of 0.7e2.2 and the O/C of 0.1e0.8, which agree with previous studies (Ma et al., 2019; Yekta et al., 2012). The amount of assigned formulas of the reactors with pH uncontrolled, 7, 8, 9 and 10 were 4493, 4682, 5338, 5545 and 5668, respectively. The increased diversity of assigned formulas illustrated that the hydrolysis of organic matter was enhanced by increasing the fermentative pH. To further distinguish the DOM characteristics, the differential molecular formulas between pH uncontrolled and pH 7, 8, 9 and 10 were analyzed (Fig. 3). With the increase of pH, more and more differential molecular formulas appeared. Lipid-like and H/C > 1.8 of proteins/amino sugars-like molecules decreased at pH 8, 9 and 10, compared with pH uncontrolled. These compounds form the major and easily degradable substrates to fermentative bacteria for VFAs production. Combined with the increased VFAs production (Fig. 1b) and the decreased EPS content (Fig. 5), a reasonable inference is that the hydrolysis and degradation of these compounds was highly enhanced at pH of 8, 9 and 10. H/C < 1.7 of proteins/amino sugars-like DOM are potential aromatic compounds (Hu et al., 2018) and increased with the increase of pH, which is consistent with the observations by EEM. Increasing DOM molecules were found in the carbohydrate-like region with the increase of pH and its diversity was lower than proteins/amino sugars. Increasing lignin-like and tannins-like substances were also found with the increase of pH. Lignin-like substances and tanninslike substances are potential aromatic and hard-biodegradable material (Ma et al., 2019; Hertkorn et al., 2006). The degradation

of carboxylic and alicyclic organic compounds can lead to an increase of lignin-like substances (Hertkorn et al., 2006). Moreover, lignocellulosic matter is one of the components of SS (Hao et al., 2017). The numerous lignin-like molecules in high pH reactors (especially pH 10) were most likely attributed to the high degradation of proteins and/or the low degradation of the hydrolytic lignocellulosic matter. Therefore, it is suggested that both higher molecular diversity and greater generation or degradation of easily degradable substrates accounted for the improved VFAs production in SAF. Meanwhile, hard biodegradable molecules were also greatly generated (especially pH 10). N-containing compounds are related to the transformation of proteins in SS (Xiao et al., 2017). The number of N-containing molecules was 2558 (uncontrolled), 2484 (7), 2703 (8), 2789 (9) and 2894 (10), which indicated that hydrolysis or transformation of those compounds were enhanced by increasing pH. Lu et al. (2018) also found that alkaline pretreatment led to a large amount of Ncompounds. The large numbers of N-containing molecules in the reactors may be because proteins are the major component of SS and the key substrate for bacteria or alkalinity. Among the N-containing molecules, 2303 (pH 10), 2181 (pH 9), 2072 (pH 8), 1924 (pH 7) and 1825 (pH uncontrolled) molecules were located in ligninlike and tannins-like regions. Xiao et al. (2017) reported that the dominant N species was pyridine-N in sludge after alkaline pretreatment, and cyclization of amine-N intermediates may generate heterocyclic-N compounds. The high diversity of hardbiodegradable N-containing molecules at pH 10 may increase the diversity of soluble N-containing organic matter in effluent of BNR (Eom et al., 2017), which would be adverse to the effluent quality of BNR process. 3.2.4. Molecular weight distribution of soluble proteins The MW distribution of soluble PN in the reactors with different pH is shown in Fig. 4. MW < 30 kDa PN was more dominant than MW > 30 kDa PN in all reactors. Xiao et al. (2017) also found that after a 30-min pretreatment of SS at pH of 6.3e10, MW < 20 kDa PN was the dominant PN. MW < 1 kDa PN increased from 53.26 ± 7.98 to 381.5 ± 38.4 mg/L as the pH increased (p < 0.05), implying that the microbes in reactors with higher pH possessed more PN that can be directly degraded. This phenomenon also accounted for the

6

S. Ma et al. / Water Research 164 (2019) 114924

Fig. 3. Comparison of the van Krevelen diagram of DOM at four different pH (7, 8, 9 and 10) reactors against the pH uncontrolled reactor. The differential molecular formulas between the pH uncontrolled reactor and pH 7 (a), pH 8 (b), pH 9 (c) pH 10 (d). The numbered circles in each diagram represent the following: (1) lipids; (2) proteins/amino sugars; (3) carbohydrates; (4) unsaturated hydrocarbons; (5) lignin; (6) tannins; and (7) condensed aromatics.

enhanced VFAs production at SAF. Chen et al. (2017) also found that low MW PN was accumulated with increased pH levels and the

profiles of those DOM might be a reason for this phenomenon. Besides the mentioned reasons in section 3.2.2, the mainly accumulated proteins/amino sugars-like DOM at high pH levels is tend to have low H/C (<1.7) and high unsaturated degree (Table 2), which also likely limits their quick degradation under restrained retention time (Hu et al., 2018; Yekta et al., 2012). The percentage of MW < 1 kDa PN at pH of uncontrolled, 7, 8, 9 and 10 was 30.06%, 33.35%, 44.55%, 52.60% and 42.74%, respectively. This phenomenon illustrates that increasing pH promotes the transformation of PN from high MW to low MW. The decreased percentage of

Table 2 Counts of molecule formulas of the DOM in the SS alkaline fermentation reactors with different pH.

Fig. 4. Molecular weight distribution of soluble protein generated in the reactors with different pH.

Reactor code

uncontrolled

7

8

9

10

Total CHO CHON CHOS CHONS Average DBE

4493 1071 1825 728 776 6.6

4682 1383 1924 619 577 7

5338 1580 2072 740 668 7.1

5545 1755 2181 625 528 7.3

5668 1782 2303 599 533 7.5

The reactor code of uncontrolled, 7, 8, 9 and 10 represent the reactor operating at pH of uncontrolled, 7, 8, 9 and 10, respectively.

S. Ma et al. / Water Research 164 (2019) 114924

MW < 1 kDa PN at pH 10 may result from the efficient degradation of those PN. Previous studies have reported that PN-degrading microbes were significantly accumulated in SS fermentation at pH 10 (Huang et al., 2016; Zheng et al., 2013). Alkaline fermentation of SS not only enhanced the hydrolysis of PN but also enhanced the transformation of PN from high MW to low MW, which further favored the VFAs production. MW > 30 kDa PN also showed a significant increase at pH 10 (176.4 ± 18.0 mg/L). High MW PN can account for 13.37%e17.39% of the refractory DOM in effluent of BNR process using sludge alkaline fermentation liquid as its carbon source (Cao et al., 2019b). Enhanced transformation of high MW PN to VFAs in the reactor at pH 10 would be meaningful, because of those PN are also good substrate for fermentative bacteria but may be adverse to BNR process effluent quality. 3.3. EPS characteristics in alkaline fermentation reactors 3.3.1. EPS distribution and compositions pH clearly affected the distribution of PN in different fractions of EPS, while PS was almost uniformly distributed along sludge EPS layers (Fig. 5), which agrees with previous studies (Yu et al., 2008). Total EPS content increased with the increase of pH, which means that cell lysis and transformation of organic matter from pellets to EPS was enhanced by increasing pH, especially to SB-EPS. The dissociation of acidic groups in EPS and the repulsions between the negatively charged EPS are responsible for the enhanced transformation of organic matter (Gonzalez et al., 2018). LB-EPS content was lower than TB-EPS in all reactors, especially at pH 10. This could be due to the availability of LB-EPS to microbes. 3.3.2. FTIR spectra of TB and LB-EPS In general, the peaks detected at other pH values, became weak or absent at pH 10 (Fig. S4). For LB-EPS, the bands appearing at 1700-1500 cm1 at pH 10 were weaker when exposed to other pH. The band at 1569 cm1 was not detected at pH 10 but was detected at the other pH values; moreover, this band was only detected in DOM at pH 10 (Fig. 2a). This phenomenon illustrates that at pH 10, proteins in LB-EPS were largely released. And LB-EPS are available to microbes (Yu et al., 2008). Thus, the large differences among the FTIR spectra of LB-EPS might be due to the combined alkali and biotic effect. Hydrolysis of TB-EPS and degradation of resultant organics played a key role in protein solubilization (Xiao et al., 2017). In the FTIR spectra of TB-EPS, the bands at 1520 cm1, 1590 cm1 and 1630 cm1 became weak at the increase of pH. The results indicate that alkaline fermentation can efficiently attack proteins in the TB-

7

EPS layer and convert those proteins into bulk liquid. A band at 1410 cm1 at pH 10 was weaker there than at any of the other pH values. Two weak bands at 1350 cm1 and 1330 cm1 were only detected in TB-EPS at pH uncontrolled, 7 and 8. The results imply that increasing fermentative pH to 9 or 10 can promote the hydrolysis of carboxylic group-containing and hydrocarbon-like compounds in TB-EPS. However, two weak peaks at 1002 cm1 and 1157 cm1 were only detected in TB-EPS at pH 9 and 10. Compared with PS, alkaline fermentation can better promote the release of PN in TB-EPS. The results suggest that increasing the fermentation pH to 10 can efficiently promote the release and hydrolysis of organic compounds (especially proteins) in LB and TB-EPS. 3.3.3. EEM spectra of TB and LB-EPS EEM spectra of LB and TB-EPS in the reactors with different pH are shown in Fig. S5 and Fig. S6. Peaks belonging to aromatic protein-like, humic-like and fulvic-like substances were observed in all the samples. In LB and TB-EPS, the fluorescence intensity of those peaks was the lowest at pH 10. An obvious decrease in the protein-like substances' fluorescence intensity appeared at pH 9 and 10. This phenomenon is consistent with the observations by FTIR, implying the efficient hydrolysis of EPS (especially protein) occurring at pH 10. The decrease of humic-like and fulvic-like substances’ fluorescence intensity with increasing pH also explains the increase of fluorescence intensity of those substances in DOM (SB-EPS). In general, the fluorescence intensity of these peaks in TB-EPS was higher than in LB-EPS. Xiao et al. (2017) observed similar results in SS alkaline pretreatment. LB-EPS is the outer layer and its availability for microbes may be the reason behind this phenomenon. 3.4. Implications of DOM characteristics in sludge alkaline fermentation This study found out that increasing the fermentative pH to 10 results in the highest easily biodegradable molecules and molecular diversity, which contributes to the maximum VFAs production. Meanwhile, pH 10 also enhances the release of hard-biodegradable materials (e.g., lignin-like substances). Moreover, a large amount of N-containing hard-biodegradable molecules were generated at pH 10. A recent study has reported that sequencing batch reactor driven BNR process has difficulty in degrading humic-like and fulvic-like substances (Cao et al., 2019b). The problem lies in their less permeable, not easily biodegradable nature, which has proven to be adverse to the effluent quality and advanced treatment, i.e., increase the diversity of effluent organic matter and do harm to the reuse of treated wastewater in WWTPs (Michael-Kordatou et al.,

Fig. 5. EPS composition and distribution of the sludge in the reactors with different pH: (a) proteins; (b) polysaccharides.

8

S. Ma et al. / Water Research 164 (2019) 114924

2015; Ni et al., 2011). Taken together, fermentation of SS at pH 10 can significantly enhance VFAs production, but further studies are needed to reduce the hard-biodegradable organic matter. Coagulation has been widely used to separate alkaline fermentation liquid (Li et al., 2011; Liu et al., 2018). Recently, Song et al. (2019) reported that the removal efficiency of humic acid by coagulation using AlCl3 can reach to 80%. Moreover, using alkaline fermentation liquid as carbon source, humic acid can improve short-cut nitrification-denitrification (Ji and Chen. 2010). If combining with membrane bioreactors, the hard-biodegradable DOM can also be removed. The results of this study provide fundamental information about DOM characteristics and add to the body of knowledge needed to achieve a comprehensive understanding of SS alkaline fermentation. 4. Conclusions This paper focused on the functional groups, fluorescent components and molecular characteristics of DOM, as well as the molecular weight of PN in SAF for VFAs production. Results show that increasing pH to 10 can lead to a significantly enhanced hydrolysis of tryptophan-like and tyrosine-like substances and increases the content of low MW PN (p < 0.05). EPS analysis also confirmed the enhanced hydrolysis of those DOM. DOM molecular features revealed that pH 10 results in the highest molecular diversity and generation or degradation of easily degradable substrates. The enhanced hydrolysis of those DOM contributes to the VFAs production. However, increasing fermentative pH also promotes the generation of hard-biodegradable organic matter, e.g., humic-like and lignin-like substances. At pH 10, a large amount of hard biodegradable N-containing molecules were also observed. Anaerobic fermentation of SS at pH 10, is favored to enhance VFAs production and, can also result in a higher content of refractory DOM. To further promote SS alkaline fermentation as an appropriate solution for both SS treatment and VFAs production, more studies are required to assess the influence of SS alkaline fermentation liquid on effluent quality of BNR process and find treatment methods to eliminate refractory DOM. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This work was supported by the National Water Pollution Control and Treatment Science and Technology Major Project (2017ZX07202-001), Natural Science Foundation of Jiangsu Province (BK20180346), National Natural Science Foundation of China (51808281) and Fundamental Research Funds for the Central Universities (021114380091). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.watres.2019.114924. References APHA, 2005. Standard Methods for the Examination of Water and Wastewater, twenty-first ed. American Public Health Association, Washington, DC, USA. ve, J., Dewil, R., 2008. Principles and potential of the Appels, L., Baeyens, J., Degre anaerobic digestion of waste-activated sludge. Prog. Energ. Combust 34 (6), 755e781.

Azman, S., Khadem, A.F., Plugge, C.M., Stams, A.J., Bec, S., Zeeman, G., 2017. Effect of humic acid on anaerobic digestion of cellulose and xylan in completely stirred tank reactors: inhibitory effect, mitigation of the inhibition and the dynamics of the microbial communities. Appl. Microbiol. Biotechnol. 101 (2), 889e901. Badireddy, A.R., Chellam, S., Gassman, P.L., Engelhard, M.H., Lea, A.S., Rosso, K.M., 2010. Role of extracellular polymeric substances in bioflocculation of activated sludge microorganisms under glucose-controlled conditions. Water Res. 44, 4505e4516. Cao, S., Qian, T., Zhou, Y., 2019a. New insights on the sludge fermentation liquid driven denitrification: evaluation of the system performance and effluent organic matter (EfOM). Bioresour. Technol. 289, 121621. https://doi.org/10.1016/ j.biortech.2019.121621. Cao, S., Sun, F., Lu, D., Zhou, Y., 2019b. Characterization of the refractory dissolved organic matters (rDOM) in sludge alkaline fermentation liquid driven denitrification: effect of HRT on their fate and transformation. Water Res. 159, 135e144. Chen, Y., Jiang, S., Yuan, H., Zhou, Q., Gu, G., 2007. Hydrolysis and acidification of waste activated sludge at different pHs. Water Res. 41 (3), 683e689. Chen, Y., Jiang, X., Xiao, K., Shen, N., Zeng, R.J., Zhou, Y., 2017. Enhanced volatile fatty acids (VFAs) production in a thermophilic fermenter with stepwise pH increase - investigation on dissolved organic matter transformation and microbial community shift. Water Res. 112, 261e268. Eom, H., Borgatti, D., Paerl, H.W., Park, C., 2017. Formation of low molecular-weight dissolved organic nitrogen in predenitrification biological nutrient removal systems and its impact on eutrophication in coastal waters. Environ. Sci. Technol. 51 (7), 3776e3783. Fernandes, T.V., van Lier, J.B., Zeeman, G., 2014. Humic acid-like and fulvic acid-like inhibition on the hydrolysis of cellulose and tributyrin. BioEnergy Res. 8 (2), 821e831. Gonzalez, A., Hendriks, A., van Lier, J.B., de Kreuk, M., 2018. Pre-treatments to enhance the biodegradability of waste activated sludge: elucidating the rate limiting step. Biotechnol. Adv. 36 (5), 1434e1469. Hao, X., Wei, J., van Loosdrecht, M.C.M., Cao, D., 2017. Analysing the mechanisms of sludge digestion enhanced by iron. Water Res. 117, 58e67. Hertkorn, N., Benner, R., Frommberger, M., Schmitt-Kopplin, P., Witt, M., Kaiser, K., Kettrup, A., Hedges, J.I., 2006. Characterization of a major refractory component of marine dissolved organic matter. Geochem. Cosmochim. Acta 70, 2990e3010. Hu, H., Liao, K., Shi, Y., Ding, L., Zhang, Y., Ren, H., 2018. Effect of solids retention time on effluent dissolved organic nitrogen in the activated sludge process: studies on bioavailability, fluorescent components, and molecular characteristics. Environ. Sci. Technol. 52 (6), 3449e3455. Huang, X., Mu, T., Shen, C., Lu, L., Liu, J., 2016. Effects of bio-surfactants combined with alkaline conditions on volatile fatty acid production and microbial community in the anaerobic fermentation of waste activated sludge. Int. Biodeterior. Biodegrad. 114, 24e30. Ishii, S.K., Boyer, T.H., 2012. Behavior of reoccurring PARAFAC components in fluorescent dissolved organic matter in natural and engineered systems: a critical review. Environ. Sci. Technol. 46 (4), 2006e2017. Ji, Z., Chen, Y., 2010. Using sludge fermentation liquid to improve wastewater shortcut nitrification-denitrification and denitrifying phosphorus removal via nitrite. Environ. Sci. Technol. 44 (23), 8957e8963. Jin, B., Wang, S., Xing, L., Li, B., Peng, Y., 2016. Long term effect of alkali types on waste activated sludge hydrolytic acidification and microbial community at low temperature. Bioresour. Technol. 200, 587e597. Kamjunke, N., von Tumpling, W., Hertkorn, N., Harir, M., Schmitt-Kopplin, P., Norf, H., Weitere, M., Herzsprung, P., 2017. A new approach for evaluating transformations of dissolved organic matter (DOM) via high-resolution mass spectrometry and relating it to bacterial activity. Water Res. 123, 513e523. Kujawinski, E.B., Behn, M.D., 2006. Automated analysis of electrospray ionization Fourier transform ion cyclotron resonance mass spectra of natural organic matter. Anal. Chem. 78 (13), 4363e4373. Li, D., Zhou, Y., Tan, Y., Pathak, S., Majid, M.B., Ng, W.J., 2016. Alkali-solubilized organic matter from sludge and its degradability in the anaerobic process. Bioresour. Technol. 200, 579e586. Li, J., Hao, X.D., van Loosdrecht, M.C.M., Yu, J.L., Liu, R.B., 2019. Adaptation of semicontinuous anaerobic sludge digestion to humic acids. Water Res. 161, 329e334. Li, X., Chen, H., Hu, L., Yu, L., Chen, Y., Gu, G., 2011. Pilot-scale waste activated sludge alkaline fermentation, fermentation liquid separation, and application of fermentation liquid to improve biological nutrient removal. Environ. Sci. Technol. 45 (5), 1834e1839. Li, X., Dai, X., Takahashi, J., Li, N., Jin, J., Dai, L., Dong, B., 2014a. New insight into chemical changes of dissolved organic matter during anaerobic digestion of dewatered sewage sludge using EEM-PARAFAC and two-dimensional FTIR correlation spectroscopy. Bioresour. Technol. 159, 412e420. Li, X., Peng, Y., Ren, N., Li, B., Chai, T., Zhang, L., 2014b. Effect of temperature on short chain fatty acids (SCFAs) accumulation and microbiological transformation in sludge alkaline fermentation with Ca(OH)2 adjustment. Water Res. 61, 34e45. Liu, H., Han, P., Liu, H., Zhou, G., Fu, B., Zheng, Z., 2018. Full-scale production of VFAs from sewage sludge by anaerobic alkaline fermentation to improve biological nutrients removal in domestic wastewater. Bioresour. Technol. 260, 105e114. Lu, D., Xiao, K.K., Chen, Y., Soh, Y.N.A., Zhou, Y., 2018. Transformation of dissolved organic matters produced from alkaline-ultrasonic sludge pretreatment in anaerobic digestion:From macro to micro. Water Res. 142, 138e146. Luo, J., Chen, Y., Feng, L., 2016. Polycyclic aromatic hydrocarbon affects acetic acid production during anaerobic fermentation of waste activated sludge by altering

S. Ma et al. / Water Research 164 (2019) 114924 activity and viability of acetogen. Environ. Sci. Technol. 50 (13), 6921e6929. Ma, S.J., Ma, H.J., Hu, H.D., Ren, H.Q., 2019. Effect of mixing intensity on hydrolysis and acidification of sewage sludge in two-stage anaerobic digestion: characteristics of dissolved organic matter and the key microorganisms. Water Res. 148, 359e367. Maizel, A.C., Remucal, C.K., 2017. The effect of advanced secondary municipal wastewater treatment on the molecular composition of dissolved organic matter. Water Res. 122, 42e52. Maspolim, Y., Zhou, Y., Guo, C., Xiao, K., Ng, W.J., 2015. The effect of pH on solubilization of organic matter and microbial community structures in sludge fermentation. Bioresour. Technol. 190, 289e298. Michael-Kordatou, I., Michael, C., Duan, X., He, X., Dionysiou, D.D., Mills, M.A., FattaKassinos, D., 2015. Dissolved effluent organic matter: characteristics and potential implications in wastewater treatment and reuse applications. Water Res. 77, 213e248. Ni, B.J., Rittmann, B.E., Yu, H.Q., 2011. Soluble microbial products and their implications in mixed culture biotechnology. Trends Biotechnol. 29 (9), 454e463. Osborne, D.M., Podgorski, D.C., Bronk, D.A., Roberts, Q., Sipler, R.E., Austin, D., Bays, J.S., Cooper, W.T., 2013. Molecular-level characterization of reactive and refractory dissolved natural organic nitrogen compounds by atmospheric pressure photoionization coupled to Fourier transform ion cyclotron resonance mass spectrometry. Rapid Commun. Mass Spectrom. 27 (8), 851e858. Song, J., Jin, P., Jin, X., Wang, X.C., 2019. Synergistic effects of various in situ hydrolyzed aluminum species for the removal of humic acid. Water Res. 148, 106e114. Sun, F., Xiao, K., Zhu, W., Withanage, N., Zhou, Y., 2018. Enhanced sludge solubilization and dewaterability by synergistic effects of nitrite and freezing. Water Res. 130, 208e214. Valle, J., Gonsior, M., Harir, M., Enrich-Prast, A., Schmitt-Kopplin, P., Bastviken, D., Conrad, R., Hertkorn, N., 2018. Extensive processing of sediment pore water dissolved organic matter during anoxic incubation as observed by high-field mass spectrometry (FTICR-MS). Water Res. 129, 252e263. Wang, B.B., Liu, X.T., Chen, J.M., Peng, D.C., He, F., 2018. Composition and functional group characterization of extracellular polymeric substances (EPS) in activated sludge: the impacts of polymerization degree of proteinaceous substrates.

9

Water Res. 129, 133e142. Wang, X., Li, Y., Zhang, Y., Pan, Y.R., Li, L., Liu, J., Butler, D., 2019. Stepwise pH control to promote synergy of chemical and biological processes for augmenting shortchain fatty acid production from anaerobic sludge fermentation. Water Res. 155, 193e203. Wu, H., Gao, J., Yang, D., Zhou, Q., Liu, W., 2010. Alkaline fermentation of primary sludge for short-chain fatty acids accumulation and mechanism. Chem. Eng. J. 160 (1), 1e7. Xiao, K., Chen, Y., Jiang, X., Seow, W.Y., He, C., Yin, Y., Zhou, Y., 2017. Comparison of different treatment methods for protein solubilisation from waste activated sludge. Water Res. 122, 492e502. Yekta, S.S., Gonsior, M., Schmitt-Kopplin, P., Svensson, B.H., 2012. Characterization of dissolved organic matter in full scale continuous stirred tank biogas reactors using ultrahigh resolution mass spectrometry: a qualitative overview. Environ. Sci. Technol. 46 (22), 12711e12719. Yu, G.H., He, P.J., Shao, L.M., He, P.P., 2008. Toward understanding the mechanism of improving the production of volatile fatty acids from activated sludge at pH 10.0. Water Res. 42 (18), 4637e4644. Zhang, P., Chen, Y., Zhou, Q., 2009. Waste activated sludge hydrolysis and shortchain fatty acids accumulation under mesophilic and thermophilic conditions: effect of pH. Water Res. 43 (15), 3735e3742. Zhang, Q., Vlaeminck, S.E., DeBarbadillo, C., Su, C., Al-Omari, A., Wett, B., Pumpel, T., Shaw, A., Chandran, K., Murthy, S., De Clippeleir, H., 2018. Supernatant organics from anaerobic digestion after thermal hydrolysis cause direct and/or diffusional activity loss for nitritation and anammox. Water Res. 143, 270e281. Zhang, Y., Wang, X.C., Cheng, Z., Li, Y., Tang, J., 2016. Effect of fermentation liquid from food waste as a carbon source for enhancing denitrification in wastewater treatment. Chemosphere 144, 689e696. Zheng, X., Su, Y., Li, X., Xiao, N., Wang, D., Chen, Y., 2013. Pyrosequencing reveals the key microorganisms involved in sludge alkaline fermentation for efficient short-chain fatty acids production. Environ. Sci. Technol. 47 (9), 4262e4268. Zhu, L., Qi, H.Y., Lv, M.L., Kong, Y., Yu, Y.W., Xu, X.Y., 2012. Component analysis of extracellular polymeric substances (EPS) during aerobic sludge granulation using FTIR and 3D-EEM technologies. Bioresour. Technol. 124, 455e459.