Soluble microbial products (SMPs) can act as a disinfection byproduct (DBP) precursor besides natural organic matter (NOM) when source water is polluted by biologically treated wastewater effluent that has SMPs as its main component. Influential factors of SMPs as a DBP precursor were investigated in this study. Model feed substrates were biologically incubated to simulate the biological treatment of wastewater organics, and the SMPs produced were chlorinated according to the standard procedure to study the DBP formation potential (DBPFP) of SMPs. Feed chemical type is a crucial factor affecting SMP production and the following DBP formation. SMPs from four kinds of model feed substrates with the same initial organic carbon concentration produced DBPs with a wide range from 196 to 684 μg L−1 and also different DBP formation properties. Different organic substrates would facilitate the growth of different microbial species, which produce SMPs with varied levels and chemical structure and subsequently different DBP formation characters. For the environmental factors, an anaerobic condition showed a significant effect, producing extremely high chloral hydrate up to about 2000 μg L−1, probably due to the production of volatile fatty acids. Different incubation conditions can not only bring about different levels of SMPs and DBPs, but also SMPs with different DBP formation feathers.

INTRODUCTION

The introduction of disinfection as a standard treatment technique for the drinking water supply caused a large drop in morbidity from infectious diseases, and is considered as one of the major public health advances in the last century. However, it also brings about the problem of the disinfection byproduct (DBP), with human health concerns (Grellier et al. 2015; Ma et al. 2015a, 2015b; Yang & Zhang 2016). The main organic DBP groups include trihalomethanes (THMs) and haloacetic acids (HAAs). The major DBP precursor that reacts with disinfectants to form DBPs is considered to be natural organic matter (NOM) (Bond et al. 2012; Tian et al. 2013; Zheng et al. 2016). However, the DBP precursors may also be attributed to water pollution caused by human activities. Water pollution is becoming one of the most serious public health crises of the world nowadays, especially in developing countries with rapid growth of population and economy. Researchers have found that organic pollutants in wastewater increase DBP precursors in the receiving water, resulting in more DBP formation in drinking water (Ding et al. 2013; Doederer et al. 2014; Jin et al. 2016). Therefore, there is a need to address wastewater-derived DBP problems in water supply.

Soluble microbial products (SMPs) are organic compounds that are released by microorganisms into water during substrate metabolism and microbial decay. SMPs are the majority of effluent organic matter (EfOM) in biologically treated wastewater (Zuthi et al. 2013; Xie et al. 2016). Hence, SMPs as a DBP precursor need to be addressed when investigating the effect of biologically treated wastewater effluent discharges on drinking water quality. It has been found that SMPs can result in an increase in DBP formation (Yang et al. 2011; Fisher et al. 2014; Ma et al. 2015a, 2015b). However, the study of SMPs as a DBP precursor is much limited. Our previous study demonstrated that SMPs have different properties to NOM as a DBP precursor (Liu et al. 2014). SMPs contain more biomolecules, e.g. proteins and polysaccharides, and have less aromatic structure compared with NOM, which is dominated by humic substances. Small molecules prevail in SMPs. Due to their varied chemical structure, SMPs have different reactivity with disinfectants and DBP speciation profile compared with NOM. Although SMPs are less reactive with disinfectants, more harmful DBPs can be formed by them. Thus, as another group of important DBP precursors derived from biologically treated wastewater, it is necessary to study how the biological processes affect SMP production during wastewater treatment and following DBP formation. It has been found that SMP production can be affected by several factors, like feed substrate, biomass, etc. (Kunacheva & Stuckey 2014). However, it is still unknown how these factors affect the sequential DBP formation by SMPs. The current work aims to study the effect of several factors on SMP production and the following DBP formation, including feed type, biomass concentration, nitrification, temperature, pH, and dissolved oxygen (DO).

MATERIALS AND METHODS

SMP sample production

Model organic substrates were dissolved in ultrapure water as the feed solutions. Biodegradation incubation was carried out in a batch reactor placed in a temperature-controlled incubator at 20 °C. The activated sludge from a membrane bioreactor was dosed as the seed biomass into the bio-reactors with an initial suspended solids (SS) concentration of 2 mg L−1. N, P, and other trace nutrients were added to the organic solutions. Aeration was conducted by air pumps to provide oxygen to the water, and the water pH was controlled at about 7 with a phosphate buffer. After incubation, the solution was filtered through 0.45-μm membranes (Millipore) to remove any suspended solids to obtain SMP samples before the subsequent experiments.

Feed type effect

Four different types of model organic substrates, including glucose, starch (a polysaccharide), glycine (an amino acid), and bovine serum albumin (BSA) (a protein), were dissolved in ultrapure water to make feed solutions with the same initial dissolved organic carbon (DOC) concentration of 80 mg L−1. These model organic substrates were selected because polysaccharides, proteins, and amino acids are the main organic components of wastewater. The feed substrate solutions, in triplicates, were incubated for biodegradation for 5 days according to the process described above. Preliminary tests monitored the DOC concentration of the feed substrate solutions every day for 10 days. DOC dropped from the initial level to a constant level after 2 days (for bovine serum albumin) or 4 days (for glucose, glycine, starch) of incubation. After then, the DOC level kept almost constant. According to the preliminary tests, after 5 days of biodegradation incubation, the model feed substrates could be degraded completely, and SMP remained. After the incubation, solutions were collected and filtered, on which the DOC, UV absorbance at 254 nm (UV254), and DBP formation potential (DBPFP) tests were conducted. The filtered samples were analyzed for their organic component contents, including carbohydrates or polysaccharides, proteins, and humic-like substances. Proteins and humic-like substances were analyzed with a UV/VIS spectrophotometer (UV/VIS Lambda 25, Perkin Elmer) following the modified Lowry method (Li & Yang 2007) using bovine serum albumin (Sigma) and humic acid (Fluka) respectively as the standards. The carbohydrates were measured with the UV/VIS spectrophotometer following the phenol-sulfuric acid method (Li & Yang 2007), with glucose as the standard. In addition, the microbial species of the biomass in the incubated solution were analyzed using the molecular technique. The DNA mixture of the microbial cells was extracted, amplified by PCR (polymerase chain reaction), and separated by the method of DGGE (denaturing gradient gel electrophoresis) as described by Li et al. (2013).

Other factors

Biodegradation conditions that may affect the SMP production and the resulting DBP formation were also investigated, including biomass concentration, nitrification, temperature, pH, and DO level, with glucose as the feed substrate. As a control biodegradation condition for comparison, the biodegradation incubation of 200 mg L−1 glucose was conducted under aerobic conditions at pH 7 and 20 °C, with an initial biomass concentration of 2 mg L−1 for 5 days. The parameters above were varied to test their effects on SMP production and DBP formation properties. A higher biomass concentration of 10 mg L−1 was tested, the nitrification inhibitor (HACH) was added to stop the nitrification process, and a higher incubation temperature of 30 °C was also applied. Besides the neutral pH of around 7 without any adjustment, the feed pH was adjusted to 10 with 1 N NaOH (BDH) solution to give an abnormal pH. In addition, an anaerobic condition for biodegradation was tested. For each change of the incubation condition, the biodegradation experiment was conducted in triplets with three parallel samples to ensure the reliability and accuracy of the results. Preliminary tests were also conducted to make sure after 5 days of biodegradation incubation that the feed substrates were depleted, and SMP remained. After a 5-day incubation under different conditions, the solutions were collected and filtered, followed by organic characterization and the DBPFP test.

Determination of the DBPFP

DBPFP tests were carried out on the filtered water samples upon chlorine disinfection in accordance with Standard Methods (APHA 2012), in which excess free chlorine was added to test the reactivity of organic matter. After chlorination incubation of the DBPFP test, the following main groups of organic DBPs were analyzed: THMs and HAAs (the most predominant and commonly regulated DBP groups), trihaloacetaldehydes (the third largest group of organic DBPs in chlorinated water), halopropanones (commonly detected in chlorinated water after the previous three groups), and nitrogenous DBPs (N-DBPs) including haloacetonitriles and trihalonitromethanes (at lower concentrations but imposing higher health risks). The volatile DBP groups, including THMs, trihaloacetaldehydes, halopropanones, and N-DBPs, were analyzed according to EPA Method 551.1, and HAAs were analyzed according to EPA Method 552.3.

Analytical methods

UV254 and DOC of the organic content were measured for each water sample after filtration. UV254 is an index of aromatic structures, which are closely related to the DBPFP of a water sample. A UV-visible spectrophotometer (UV/VIS Lambda 25, Perkin Elmer) with a 1-cm cuvette cell was used to determine the UV254. DOC was determined by a TOC (total organic carbon) analyzer (IL550, Lachat) using the catalytic combustion-infrared method. Accordingly, the specific UV absorbance (SUVA) indicating the aromatic structure density of the organics in water was calculated with the UV254 divided by DOC. Similarly, the DBPFP yield indicating the chlorination reactivity of the organics in the water sample was determined from its DBPFP divided by DOC.

RESULTS AND DISCUSSION

Feed type effect

The four types of model organic substrates behaved differently during the 5-day biodegradation incubation. Their DOC all decreased from the same initial value of 80 mg L−1. DOC for glucose and starch solutions decreased at a relatively steady rate and reached a constant value after about 4 days of incubation. The DOC of glycine had almost no change on the first 2 days and then depleted on the next 2 days. For BSA, most of the organic carbon was utilized rapidly just within 2 days. The four model organic substrates also performed differently on the UV254 profile. UV254 of glucose, starch, and glycine, which had no UV absorbance initially without aromatic structures in them, increased after the biodegradation incubation started, indicating the production of SMPs. The initial UV absorbance of BSA decreased for structure decomposition by bacteria and then increased also due to SMP release. Biomass of the four model organics behaved relatively similarly, which first increased for organic metabolism and then decreased for bacteria decay. However, different feeds resulted in different biomass levels, which is in an order of BSA > glucose ∼ starch > glycine (Table 1).

Table 1

Parameters of feed substrate solutions after 5-day incubation

 GlucoseStarchGlycineBSA
DOC (mg/L) 7.6 ± 0.9 9.5 ± 0.4 3.9 ± 0.2 4.8 ± 0.5 
UV254 (m−13.5 ± 0.4 1.1 ± 0.1 1.4 ± 0.0 3.2 ± 0.3 
Biomass (mg/L) 41.2 ± 2.8 39.8 ± 4.0 27.9 ± 2.1 51.0 ± 3.8 
Turbidity (NTU) 16.9 ± 3.1 20.6 ± 5.5 14.0 ± 1.1 20.5 ± 2.4 
Organic component (mg/L) 
 Polysaccharids 3.7 ± 0.4 9.1 ± 1.0 1.1 ± 0.2 2.2 ± 0.2 
 Proteins 4.3 ± 0.2 0.5 ± 0.0 3.0 ± 0.2 3.0 ± 0.3 
 Humic-like substances 0.9 ± 0.2 0.7 ± 0.1 1.0 ± 0.1 4.4 ± 1.0 
 GlucoseStarchGlycineBSA
DOC (mg/L) 7.6 ± 0.9 9.5 ± 0.4 3.9 ± 0.2 4.8 ± 0.5 
UV254 (m−13.5 ± 0.4 1.1 ± 0.1 1.4 ± 0.0 3.2 ± 0.3 
Biomass (mg/L) 41.2 ± 2.8 39.8 ± 4.0 27.9 ± 2.1 51.0 ± 3.8 
Turbidity (NTU) 16.9 ± 3.1 20.6 ± 5.5 14.0 ± 1.1 20.5 ± 2.4 
Organic component (mg/L) 
 Polysaccharids 3.7 ± 0.4 9.1 ± 1.0 1.1 ± 0.2 2.2 ± 0.2 
 Proteins 4.3 ± 0.2 0.5 ± 0.0 3.0 ± 0.2 3.0 ± 0.3 
 Humic-like substances 0.9 ± 0.2 0.7 ± 0.1 1.0 ± 0.1 4.4 ± 1.0 

The final DOC concentrations of SMPs from the four model feed solutions after 5 days differed considerably, with values of 7.6, 9.5, 3.9, and 4.8 mg L−1, respectively. Final UV254 values of glucose and BSA solutions (3.5 and 3.2 m−1) were higher than those of starch and glycine (1.1 and 1.4 m−1) (Table 1). Sequentially, their SUVA values were also different from each other. It can be concluded despite the same initial feed concentration, different types of model feed organics would result in different SMP production to different levels with different chemical properties. This is consistent with the findings of other researchers that feed types could affect the SMP production results (Shen et al. 2012).

The varied organic carbon degradation performance of the four model organic substrates resulted in varied DBP formation profiles (Figure 1(a)). DBPFPs of glucose, starch, and glycine solutions firstly increased for production of new DBP precursors by biodegradation, such as SMPs as well as some intermediate products, and then decreased for the following breakup of the biodegradable part of these precursors. The overall DBPFPs, however, increased compared with the initial condition, confirming that SMPs can act as a DBP precursor with much higher chlorination reactivity compared with the feed substrates. The high initial DBPFP of the BSA solution was reduced by organic degradation and then increased, probably also for the SMP production.
Figure 1

(a) DBPFP of the solutions as a function of the degradation time (5 days) for the four model organic substrates; (b) DBPFP yield and (c) DBP species fraction of SMPs after 5-day biodegradation with different feed substrates.

Figure 1

(a) DBPFP of the solutions as a function of the degradation time (5 days) for the four model organic substrates; (b) DBPFP yield and (c) DBP species fraction of SMPs after 5-day biodegradation with different feed substrates.

The different DOC levels and properties of SMPs from the four feed types brought about varied DBP formation characteristics. The DBPFPs of glucose and starch based SMPs were similar, with values of 251 and 294 μg L−1 respectively, which were higher than that of glycine at 196 μg L−1. The SMPs formed from BSA had an extremely high DBPFP of 684 μg L−1. The different biomass concentrations after the incubation for the four feed chemicals were also in the order of BSA > glucose ∼ starch > glycine. This suggests the significance of biomass on SMP production and the resulting DBP formation. Different organic chemicals have their different biomass yields and biotransformation efficiencies. The mineralization of BSA may yield more energy than the degradation of other feed organics. This resulted in a higher bacterial yield, thus higher levels of biomass production and SMP accumulation. Biomass is a critical factor affecting the SMP production and its DBP formation potential. Thus, the organic type would regulate SMP production through biomass during the biodegradation process, which eventually affects the DBPFP of the solution.

The DBPFP yield varied in a wide range from 31 to 143 μg mg−1 DOC, while starch-based SMPs had the lowest DBPFP and BSA-based SMPs had the highest value (Figure 1(b)). As mentioned above, the SUVA values also varied. As for the DBPFP speciation profiles, the DBP species detected from SMPs included chloroform (CF) for THMs, dichloroacetic acid (DCAA) and trichloroacetic acid (TCAA) for HAAs, chloral hydrate (CH) for trihaloacetaldehydes, trichloropropanone (TCP) for halopropanones, dichloroacetonitrile (DCAN) for haloacetonitriles, and trichloronitromethane (TCNM) for trihalonitromethanes. DCAN and TCNM are both nitrogen-containing DBPs (Figure 1(c)). SMPs derived from glucose and starch behaved similarly. For the three types of most abundant DBP groups, THMs (CF) dominated. HAAs and trihaloacetaldehydes (CH) had similar levels of formation, while among the HAA group formation of DCAA and TCAA was similar. Compared with the two types of carbohydrates, glycine-based SMPs had more HAAs and TCP and less CH formation. For BSA, less CF was formed. Significant amounts of DCAA and TCAA formation were observed, and HAAs became the most prevalent group (Figure 1(c)). Apparently, SMPs from BSA contained more HAA-forming materials. The N-DBPs are considered to be more toxic than the regulated DBPs, such as THMs and HAAs (Bond et al. 2012). However, the N-DBP fraction by SMPs with the N-based feeds including glycine and BSA, had no obvious difference from the other two model organics. The different DBP speciation profiles imply SMPs from the four kinds of feed substrates had varied kinds of DBP precursors.

The SMP properties in relation to DBP formation, such as SUVA, DBPFP yield, and DBP speciation, varied among the four model organic substrates (Figure 1(b) and 1(c)). It is apparent that different organic substrates result in SMPs with different chemical compositions. In order to investigate the issue further, concentrations of several organic components of SMPs were investigated, including polysaccharides, proteins, and humic-like substances. Although SMPs show complicated organic composition, the three important microbial-related groups may partially reflect the structure information of SMPs. Apparently the chemical composition of the four kinds of SMPs varied (Table 1). SMPs with carbohydrate (glucose and starch) feed contained more carbohydrates in the resulting solutions compared with the other two chemicals. The two nitrogenous organics, glycine and BSA, did not result in significantly more organic nitrogen, such as proteins. Glycine-based solution had the lowest total concentration of the three organic groups. SMPs with BSA feed had the highest humic-like substance concentration at 4.4 mg L−1. The humic-like substances might be attributed to the production of some SMPs with humic feathers such aromatic structure with strong reactivity with chlorine. This may explain the high DBP formation by SMPs from BSA.

It is possible that the different chemical composition of SMPs with the four kinds of feed substrates is caused by the fact that different organic substrates would facilitate the growth of different microbial species, resulting in SMPs with different chemical compositions and structural features. In order to investigate the issue further, the microbial species of biomass grown on the different feed substrates were analyzed with the PCR-DGGE method (Figure 2). The DGGE images differed obviously among the biomass produced from the four model chemicals, indicating different amounts of bacterial species and dominant species for the different samples. Different feed substrates could lead to largely varied microbial structures in the biodegradation suspensions even after just 5 days of incubation. The different bacterial population diversities for the four feeds were also reflected by other microbial properties, including morphology, color, and turbidity. In general, the microbial population would vary with the feed substrates, which in turn produces SMPs at different levels with different chemical structures and different DBP formation characteristics.
Figure 2

DGGE images for the biomass grown on different feed substrates (from left to right: glucose, starch, glycine, BSA).

Figure 2

DGGE images for the biomass grown on different feed substrates (from left to right: glucose, starch, glycine, BSA).

Other factors

SMP production and further DBP formation may perform differently under different degradation conditions. Thus, the effect of experimental biodegradation conditions was also investigated. For the control biodegradation condition test, DOC was reduced from 80 to around 2 mg L−1 after 5 days under the aerobic condition at pH 7 and 20 °C with an initial biomass of 2 mg L−1 (Figure 3(a)). With an increase in seed biomass content, nitrification process inhibition, and an increase in temperature, apparently an even better organic reduction was achieved with a final DOC of less than 2 mg L−1. However, under a less favorable condition, such as too high a pH, a lower level of organic degradation was observed with a final DOC of around 6 mg L−1. Without DO supply by aeration, the organic reduction was rather marginal, with a final DOC of 29 mg L−1 under the anaerobic condition. The UV absorbance at 254 nm was rather comparable to each other for the kinds of reaction conditions from 1.5 to 2.0 m−1 (Figure 3(a)). In relation to the DOC residue level, the SUVA values of the final glucose degradation solutions varied between 0.8 and 1.1 L m−1 mg−1 for the control, biomass, nitrification, and temperature cases. For the pH and anaerobic cases, SUVA values were much lower at 0.4 and 0.1 L m−1 mg−1, respectively.
Figure 3

(a) DOC and SUVA and (b) DBPFP and DBPFP yield of the SMPs in the solutions produced from glucose under different biodegradation conditions.

Figure 3

(a) DOC and SUVA and (b) DBPFP and DBPFP yield of the SMPs in the solutions produced from glucose under different biodegradation conditions.

The results of organic biodegradation also affected the DBPFPs of the solutions (Figure 3(b)). Proper biodegradation under the control and similar conditions had a total DBPFP of 262 μg L−1 or lower. More initial biomass might result in better organic reduction and less SMPs. A higher temperature is more favorable to bioactivity and the metabolism of the SMPs (Xie et al. 2013). In general, nonetheless, DBP formation by SMPs produced by organic degradation may not be affected significantly by the variation in these parameters.

For the other less favorable pH and anaerobic conditions, higher DBPFPs were recorded. The solution with an initial pH of 10 had a total DBPFP of 314 μg L−1, which was considerably higher than that of the control condition. More SMP may be produced for organic degradation at abnormal pH conditions. More SMPs are supposed to be released against the environmental stress. Under the anaerobic condition, the final solution after 5-day incubation had an extremely high DBPFP at 2308 μg L−1. This is consistent with the high level of organic residue left after anaerobic degradation. Anaerobic organic digestion could produce plenty of volatile fatty acids (Kondaveeti & Min 2015), resulting in substantial DOC and DBPFPs.

The DBPFP yields of SMPs under abnormal pH and anaerobic conditions were 56 and 81 μg mg−1 DOC (Figure 3(b)). These values were relatively lower than the DBPFP yields under other reaction conditions, which ranged from 119 to 138 μg mg−1 DOC. Although an abnormal pH and an anaerobic reactor could cause more organic carbon and therefore DBPFPs, the organics owned weaker reactivity with chlorine. When considering DBPFP speciation, abnormal pH showed a relatively higher DCAA formation ratio, indicating that under this condition SMPs contained more DCAA precursors. An anaerobic reactor could form extremely high CH, up to 1966 μg L−1. Volatile fatty acids at a significant level are supposed to be a precursor. The DBP species and DBPFP yield, together with SUVA results, suggest different incubation conditions could not only bring about different levels of SMPs, but also SMPs with different chemical properties. Among all the biodegradation conditions, DO level was found to be the most significant factor that affected SMPs' production and subsequent DBP formation, due to abundant short-chain fatty acid release. Although they could produce significant CH, their DBP formation capacity was very marginal.

CONCLUSIONS

SMPs can play important roles as a DBP precursor in source water owing to more and more serious water pollution, especially in developing countries. The feed chemical type was found to be a crucial factor that may influence the SMP production and following DBP formation. The mechanism can be attributed to different organic substrates facilitating the growth of different microbial species, which in turn produce SMPs with varied levels and chemical structures and subsequent different DBP formation characteristics. Among the environmental influential factors, anaerobic condition showed the most significant effect, due to the production of volatile fatty acids. Different incubation conditions can not only bring about different levels of SMPs and DBPs, but also SMPs with different DBP formation properties.

The study of DBP formation properties by SMPs is limited, although SMPs may become more and more important as a DBP precursor in source water for the worsening water pollution situation. The current study results imply the wastewater's organic property is critical, and some kinds of pollutants may be transformed into SMPs at a high level and reactivity towards chlorination. The anaerobic biodegradation of wastewater will also benefit DBP production derived from wastewater discharges in source water. Under these conditions, discharges of wastewater should be evaluated carefully, and further treatment may be needed for the control of the DBP problem. The current study can give useful information about the issue of the DBP problem derived from wastewater discharges and water pollution.

REFERENCES

REFERENCES
American Public Health Association
2012
Standard Methods for the Examination of Water and Wastewater, 22nd edn. American Public Health Association
,
Washington, DC
,
USA
.
Ma
D.
Gao
B.
Wang
Y.
Yue
Q.
Li
Q.
2015a
Factors affecting trihalomethane formation and speciation during chlorination of reclaimed water
.
Water Science and Technology
72
(
4
),
616
622
.
Yang
M. T.
Zhang
X. R.
2016
Current trends in the analysis and identification of emerging disinfection byproducts
.
Trends in Environmental Analytical Chemistry
10
,
24
34
.