This paper presents a review of methods that improve the production of volatile fatty acids (VFA) from excess sludge during the anaerobic digestion process. These methods are mainly divided into two approaches. The first approach is located in the pre-treatment methods, which change the properties of the substrates, such as thermal pre-treatment, alkaline pre-treatment, microwave pre-treatment and ultrasonic pre-treatment. The other approach is found in the fermentation process control methods, which influence the environment of anaerobic digestion for the production of VFA, such as pH, temperature, mixing, additives and solids retention time control. In the text recent research studies of each method are listed and analyzed in detail. Comparably, microwave and ultrasonic pre-treatment methods are considered emerging and promising technologies due to their efficiency and environmentally friendly characteristics. However, the microwave pre-treatment has high electricity demand, which might make the process economically unfeasible. In order to calculate optimal operation, further studies still need to be done.

## INTRODUCTION

During the processes of wastewater treatment, especially the activated sludge process, large amounts of excess sludge are generated and discharged. The cost of sludge treatment and disposal has become a big problem for wastewater treatment plants. It takes more than 40% of the plants’ total cost to treat the waste activated sludge (WAS) in China (Liu 2003; Ma et al. 2012) (waste activated sludge and excess sludge are synonyms). In the excess sludge there are plenty of microbial ﬂocs including microorganisms, extracellular polymeric substances (EPS), and inorganic matter. If proper disposal measures are not implemented, these materials can do harm to the environment. Conversely, excess sludge is a form of biomass, which contains abundant organic matters, so efforts can be attempted to make use of the organic matters present in the excess sludge (Yang et al. 2010; Zhu et al. 2011; Shehu et al. 2012).

In recent years, standards for effluent disposal have become increasingly strict. In order to remove nutrients by biological processes, extra carbon sources such as methanol, ethanol, sodium acetate and glucose need to be added into wastewater during the biological denitrification step, because carbon sources provided by the wastewater itself are typically not sufficient, especially when influent chemical oxygen demand (COD) is low (Tan et al. 2012). Nyberg et al. (1996) reported that based on the long-term experiences it has been shown that the nitrogen standards of 8 mg N/L in the effluent wastewater can be met with two carbon sources (methanol and ethanol). Very high specific nitrate utilization rates were measured in the system with the use of external carbon sources. Rates of around 10 mg N/(g-VSS·h) (VSS: volatile suspended solids) were reached with ethanol and around 3 mg N/(g-VSS·h) with methanol. A start-up with the addition of ethanol led to a direct response of the system while a start-up with methanol resulted in a much longer adaptation period before the full effect of the carbon source added was reached. However, by making use of volatile fatty acids (VFA) produced during the anaerobic digestion of excess sludge as carbon sources for wastewater treatment, the amount of excess sludge can be reduced and the cost of additional carbon sources can be saved as well.

The three-stage theory of anaerobic digestion was put forward in 1979. The anaerobic digestion process was divided into hydrolysis, acetogenesis, and methanogenesis (Madsen et al. 2011). McCarty & Smith (1986) put forward a four-stage theory of anaerobic digestion. Based on the three-stage theory, they made the acetogenesis phase more defined. The acetogenesis stage was divided into two parts, acidogenesis and acetogenesis. The breakdown structure of the four-stage process and the percentages of transferred substrates are illustrated in Figure 1 (Appels et al. 2011). The anaerobic treatment process of wastes, such as WAS, usually includes hydrolysis (the complex primary polymers of carbohydrates and proteins are converted to soluble organic compounds and further to soluble monomers by extracellular enzymes), acidogenesis (the hydrolysis products are fermented to various intermediate products such as VFA), acetogenesis (the VFA are converted into acetic acid, carbon dioxide and hydrogen by acetogenic bacteria) and methanogenesis (acetic acid and hydrogen are converted to methane and carbon dioxide by methanogenic bacteria). During the acidogenesis and acetogenesis process, VFA are produced. These VFA include acetic acid, propionic acid, iso-butyric acid, n-butyric acid, iso-valeric acid and n-valeric acid. Among these VFA, acetic acid is the major product (Wang et al. 1999; Ucisik & Henze 2008). As the anaerobic fermentation technology is an environmentally friendly method to manage excess sludge, pre-treatment methods to enhance bio-methane production of sludge have been studied by many investigators, and it has been reported in the literature that the two-phase anaerobic fermentation process was much more stable and efficient than the traditional single-phase anaerobic fermentation process (converts directly fermentative substrates to bio-methane) for bio-methane generation (Liu et al. 2006; Tan et al. 2012). Figure 2 illustrates the two-stage anaerobic digestion processes in two separate reactors. The first reactor shows that most of the input excess sludge converts to VFA during the hydrolysis, acidogenesis and acetogenesis processes. Then VFA are usually converted into methane through methanogenesis process as shown in the second reactor. This paper focuses on effects of VFA production, so attention should be paid to the first digestion reactor. Further, a consensus-simplified pathway for anaerobic digestion was described in the anaerobic digestion model no. 1 (Batstone et al. 2002). Whether as carbon source for nutrient removal in wastewater treatment plants or to produce methane in a stable way, short-chain fatty acids are more useful than long-chain fatty acids (LCFA) (Madsen et al. 2011; Tan et al. 2012). The optimal fermentation process control methods, i.e., pH, temperature, mixing, additives and solid retention time (SRT) control methods, can promote LCFA to turn into short-chain fatty acids (Wang et al. 1999).

Figure 1

Metabolic pathways for substrate degradation during the anaerobic digestion.

Figure 1

Metabolic pathways for substrate degradation during the anaerobic digestion.

Figure2

Two-stage anaerobic digestion process.

Figure2

Two-stage anaerobic digestion process.

In this paper, emphasis will be placed on the effects of pre-treatment and fermentation process control methods to enhance the production of VFA during the anaerobic digestion of excess sludge. These pre-treatment methods include thermal pre-treatment, alkaline pre-treatment, microwave pre-treatment and ultrasonic pre-treatment methods. The process control methods considered include pH, temperature, mixing, additives and solids retention time control. The purpose of this paper is to attempt to analyze the effects of each method on VFA production from excess sludge by anaerobic digestion.

## FACTORS IMPACTING THE VFA PRODUCTION

Factors that impact the production of VFA from excess sludge are primarily divided into two parts. The first factor is pre-treatment methods, which can remarkably improve the hydrolysis process by converting larger organic matter into smaller organic matter. After that, the following acidogenesis and acetogenesis processes can be conducted more readily and efficiently. The other factor is process control methods. By changing the pH, temperature, mixing, additives, or SRT, the environment of anaerobic digestion is changed, and consequently, so is the rate and degree of anaerobic digestion for the production of VFA.

### Pre-treatment for improving VFA production

During the conventional anaerobic digestion process, SRTs are typically 20–30 days, which is a long period (Appels et al. 2008). Pre-treatments could speed up these processes, particularly the hydrolysis process, which is considered as the rate-limiting step of anaerobic digestion and VFA production (Salsabil et al. 2010). Pre-treatment methods can be classified as mechanical pre-treatment, thermal pre-treatment, chemical pre-treatment and biological pre-treatment (Guo et al. 2008; Liu et al. 2008; Salsabil et al. 2010; Saha et al. 2011). All of these methods result in the disintegration of flocs and the breaking of sludge cells, causing suspended and intracellular materials to be released and dissolved into the water phase. Larger insoluble particles become smaller soluble particles, enhancing the hydrolysis process and causing the release of soluble substrates, which can be utilized by the acidogenic bacteria.

The functions of thermal, alkaline, microwave and ultrasonic pre-treatment to enhance VFA production during anaerobic digestion will be focused on because of the fact that they have been commonly used to enhance the hydrolysis process in recent years.

#### Thermal pre-treatment for improving VFA production

Thermal pre-treatments can be used to improve VFA production during the anaerobic digestion of excess sludge. After the heated internal pressure of the cells has increased, which is due to the rise in intracellular water evaporation, cell walls will begin to rupture (Appels et al. 2010). Thus, soluble organic matters are released into the water phase, which may increase the following VFA production. A wide range of temperature (60–270 °C) was applied to enhance the VFA production of excess sludge during the anaerobic digestion process, but studies were usually focused on a range of 60–180 °C (Ferrer et al. 2008), which has been determined to be a relatively low value.

Related studies of the effects of thermal pre-treatment on improving VFA generation are shown in Supplementary Table S1 (available online at http://www.iwaponline.com/wst/072/280.pdf). Ferrer et al. (2008) investigated the effect of 70 °C pre-treatment on the efficiency of thermophilic anaerobic digestion of excess sludge. They found that volatile dissolved solids increased by almost 10 times compared with the blank test after 9 hours of 70 °C pre-treatment, and this short period (9 h) of pre-treatment was enough to promote the hydrolysis process and enhance VFA production. Gavala et al. (2003) studied the difference between VFA production in a mesophilic digester and a thermophilic digester after the 70 °C thermal pre-treatment. Their experiments showed that the concentration of the VFA in the mesophilic digester was lower than in the thermophilic digester, and a similar conclusion was obtained by other researchers (Kim et al. 2002; Song et al. 2004). During the 70 °C pre-treatment, Lu et al. (2008) found an interesting phenomenon, namely that microbes in pre-treated sludge become more active than in raw sludge, which would benefit the anaerobic digestion process. Low temperature pre-treatment (70 °C) enhanced thermal solubilization of particulate materials, as well as enzyme activities; thus hydrolysis was significantly improved, and so was the VFA production. Morgan-Sagastume et al. (2011) pre-treated waste sludge with high-pressure thermal hydrolysis (130–180 °C at 6–12 bar), and found that it increased the VFA yield by 2–5 times and the VFA production rate by 4–6 times after fermentation. Bougrier et al. (2007) discovered that soluble carbohydrates and soluble proteins were increased dramatically after 135 °C pre-treatment, which would benefit VFA generation in the acetogenesis process. In another experiment, Bougrier et al. (2008) found that VFA concentration increased from 7.33 mg/g-VS (volatile solids) in raw sludge to 87.92 mg/g-VS after pre-treatment of 170 °C. That is to say, thermal pre-treatment can increase VFA yield in sludge, and also increase the soluble carbohydrates and proteins (Bougrier et al. 2007). Thus, VFA production would be greatly improved during fermentation for the abundant available substrates. The above studies also showed that the higher the thermal pre-treatment temperature, the shorter the pre-treatment time would be. However, higher temperature also requires more energy input, which also needs to be taken into account in a practical project.

#### Alkaline pre-treatment for improving VFA production

Flocs in excess sludge consist of sludge particles, and, through the use of EPS, small sludge particles can join to form flocs. These EPS include protein, saccharides, lipids and other macromolecular organic matters. Alkaline pre-treatment can destroy the structure of proteins, saccharides and lipids by chemical degradation and ionization of the hydroxyl radical (Li et al. 2008,, 2012). In that way insoluble and intracellular materials are released and the continuing acidogenesis process can proceed more easily.

Tan et al. (2012) pre-treated excess sludge for 24 hours at pH 11 and 60 °C. They then utilized mesophilic digestion for 6 days. The results showed that VFA accumulation was 1.63 times higher than it was without pre-treatment (Tan et al. 2012). The detailed information is listed in Supplementary Table S1. Kim et al. (2003) used NaOH, KOH, Mg(OH)2 and Ca(OH)2 to regulate the alkalinity of WAS at pH 12. They found that, after ambient temperature pre-treatment, the order of efficacy in sludge solubilization is NaOH > KOH > Ca(OH)2 > Mg(OH)2. Bibasic alkali agents were only partially dissolved, so monobasic agents were more efficient. Their experiments also showed that soluble COD (SCOD) reached the peak value when 7 g/L NaOH was added to adjust pH. However, too much NaOH led to a decrease in SCOD (Kim et al. 2003). Further, studies pointed out that concentrations of Na+ or K+ that are too high may lead to subsequent inhibition of VFA production (Carrère et al. 2010). Doğan & Sanin (2009) compared alkaline pre-treatment at pH 12 with pH 10 and found that the higher pH caused more carbohydrate and protein releases. More available carbohydrate and protein can result in higher concentration of SCOD, which can be converted to VFA more easily.

Alkaline pre-treatment, compared with other pre-treatment methods, utilizes a simple device and operation system. It also has high efficiency, but it is less economically attractive due to the need for additional chemicals. Alkaline pre-treatment is seldom used alone and is typically combined with other treatment methods, such as thermal pre-treatment, microwave pre-treatment or ultrasonic pre-treatment.

#### Microwave pre-treatment for improving VFA production

The effects of microwave pre-treatment can be divided into two parts: one is the thermal effect caused by the temperature increase during the microwave treatment process (Menéndez et al. 2002; Tang et al. 2010; Jiang & Ma 2011), and the other is the athermal effect caused by polarizing macromolecules. Due to the microwaves, the hydrogen bonds can be broken (Eskicioglu et al. 2007; Sólyom et al. 2011). Thanks to these two effects, EPS can be disrupted and more soluble organic matters can be released than without pre-treatment (Appels et al. 2013). Thus the hydrolysis process as well as VFA production would be improved.

Toreci et al. (2009) compared different microwave intensities for WAS pre-treatment. Their study indicated that, for long durations of anaerobic digestion (10–20 days) sludge retention time, lower microwave intensity was better than higher microwave intensity for VFA production. Meanwhile, for short durations of anaerobic digestion (5 days), higher microwave intensity is better. According to their experiments, the highest VFA production was reached with low microwave intensity at 10 days sludge retention time. Experiments of Yu et al. (2009) indicated that proteins were released more quickly than polysaccharides after microwave pre-treatment. Furthermore, as contact time became longer (20–140 s), proteins and polysaccharides in the water phase continued to increase at 900 W microwave exposure. Guo et al. (2008) pre-treated sludge with microwaves for 2 min at a power of 560 W and, after mesophilic digestion, the concentration of VFA reached 1.28 g/g-VSS. The above studies of the effects of microwave pre-treatment on VFA production are shown in Supplementary Table S1.

Microwave pre-treatment of excess sludge is a wide and promising field to study, and microwave power, intensity, contact time and temperature are all factors that should be taken into account. If any of these experimental conditions is changed, the resulting VFA production would be different, and sometimes even contradictory phenomena could be observed. Compared to conventional heating, microwave pre-treatment has the added advantage of rapid heating with less energy loss while transmitting energy (Eskicioglu et al. 2007), which leads to energy and cost savings. The microwave pre-treatment method is also more environmentally friendly than conventional heating methods.

#### Ultrasonic pre-treatment for improving VFA production

The ultrasonic pre-treatment method is based on the cavitation phenomena (Pilli et al. 2011), which can form huge shear force (Tiehm et al. 2001) and release the cytoplasm membrane from cell walls. This makes the cell walls become thinner and easier to break, thus allowing intracellular substances to be released into the aqueous phase more easily (Tomei et al. 2008). Also the ultrasound can break down flocs and cut large organic particles into smaller particles. Ultrasonic pre-treatment disintegrates both cellular and extracellular organic matters of excess sludge (Braguglia et al. 2012). More easily digested particles are then released, thus allowing the rate of hydrolysis and VFA production to be accelerated.

Yan et al. (2010) studied the effects of ultrasonic energy on VFA accumulation. They found that the maximum VFA accumulation, two times higher than without ultrasonic pre-treatment, was attained by using an ultrasonic energy density of 1.0 kW/L at a frequency of 20 kHz and fermentation duration of 3 days. Their studies also showed that enzymes related to VFA production such as phosphotransacetylase, phosphotransbutyrylase, acetate kinase and butyrate kinase had the highest activity rate at that ultrasonic energy density. Laﬁtte-Trouqué & Forster (2002) pre-treated excess sludge with ultrasound at 23 kHz for 90 s and found that, in the first 8 hours of anaerobic digestion, organic matter in the water phase was increased by 354% compared to untreated sludge. Studies by Zhuo et al. (2012) indicated that, after ultrasonic pre-treatment at a frequency of 20 kHz and energy density of 1.0 kW/L, soluble organic matters released into the water phase during that process were used by acidogenic bacteria quickly in the initial 1 day of fermentation. Quarmby et al. (1999) compared the effect of VFA production at different ultrasonic intensities and found that, in the second day of digestion, the VFA concentration reached the peak. At the peak, VFA in the sludge pre-treated by high ultrasonic intensity (365 W/min) was two times higher than without pre-treatment, while sludge pre-treated by low ultrasonic intensity (111 W/min) increased 0.6 times. The above studies of the effects of ultrasonic pre-treatment on VFA production are shown in Supplementary Table S1.

With the advantages of energy efficiency and environment friendliness, the ultrasonic pre-treatment method is one of the most promising emerging technologies. However, there are not many studies about the effects of ultrasonic pre-treatment methods on VFA production of excess sludge. Thus, in the future, research in related areas should pay attention to ultrasonic pre-treatment factors that may have an effect on excess sludge for VFA production, such as sonication frequency, density, intensity and treatment time.

### Fermentation process control for improving VFA production

Important operational parameters, such as pH, temperature, mixing, additives and SRT, should be taken into account during VFA production from anaerobic fermentation of excess sludge. Optimal choices for these parameters can significantly improve VFA production, meanwhile reducing cost and saving energy. For this reason, this section focuses on operational parameters during anaerobic fermentation of excess sludge. It also tries to determine the optimal operational conditions for VFA production, which could be used in future laboratory and pilot experiments or full-scale practices.

#### pH

The effect of pH from 4 to 11 on VFA production of excess sludge was investigated by Chen et al. (2007). The results of their studies are shown in Supplementary Table S2 (available online at http://www.iwaponline.com/wst/072/280.pdf). Their experiments indicated that the concentration of VFA on the eighth day of fermentation at pH 4.0, 7.0 and 10.0 was, respectively, 32.78, 77.85 and 250.39 mg/g-VSS, while VFA in the blank test was only 58.58 mg/g-VSS. At pH 11 VFA production was limited during the first 12 days of fermentation, which might be due to the toxic effect caused by large amount of Na+ (Mouneimne et al. 2003). However, when the acidogens adapted to the toxic environment, the VFA concentration increased dramatically. It was noted in the studies of Yu et al. (2008) that VFA production of excess sludge at pH 10 was two to 34 times higher than that at pH 5.5, whether in mesophilic reactors or thermophilic reactors. Zhang et al. (2009a) investigated the effect of different pH under mesophilic and thermophilic conditions. They found that under alkaline condition, with pH increasing, more fermentation time was needed to reach the maximum VFA production in mesophilic fermentation. Although at pH 10 the maximum VFA production was recorded, 0.32 g/g-VSS, compared to other pH conditions, it required twice the fermentation time than at pH 9 with the amount of 0.30 g/g-VSS. In thermophilic fermentation, at pH 10 or 11, VFA accumulation was not significantly improved, but at pH 8 it reached the maximum VFA production.

Recent studies from different researchers indicated that VFA production was much higher in an alkaline condition than acidic condition (Elefsiniotis & Oldham 1994; Moon et al. 2004), though few studies reported the final pH or pH evolution of WAS during fermentation. The result is contrary to the beliefs that acidic condition would be the best choice to improve the acidogenesis process (Yu et al. 2008; Zhang et al. 2009a, 2009b). However, former studies were focused on pH less than 7, and alkaline conditions were seldom considered (Yu et al. 2003). Furthermore, variety in substrates led to diverse conclusions (Elefsiniotis & Oldham 1994; Moon et al. 2004). VFA production was significantly improved in an alkaline condition, because alkaline condition facilitate the breakage of the sludge matrix, thus increasing the effective contact between extracellular organic matters and enzymes, which also creates a favorable environment for microbes to accumulate VFA (Yu et al. 2008). Comparing to acidic conditions, improvement of hydrolysis was another reason for enhanced VFA accumulation at alkaline pH.

#### Temperature

Cai et al. (2009) compared VFA production of excess sludge at 21 ± 1 °C, 35 ± 1 °C and 60 ± 1 °C. In their studies it was noted that the higher the temperature, the more the VFA accumulation. They also found that high temperature shortened fermentation time for VFA production in an alkaline condition, because higher temperature shortened the lag-stage for biological adaptation. Yuan et al. (2011) conducted experiments at 4 °C, 14 °C and 24.6 °C, and noted the same phenomena as Cai et al. (2009), i.e. with the decrease in temperature, fermentation time increased. They also pointed out that low temperature would slow down, but not inhibit, the activity of microbes, which resulted in less VFA accumulation. Higher temperature, which can promote release of intercellular organic materials, is an advantage for bacterial hydrolysis during fermentation. Because of sufficient organic compounds, the production of VFA was greater in higher temperature than lower temperature. Studies by Feng et al. (2009) showed that the total VFA concentration of a continuous-flow fermentation system was 0.19 g/g-VSS at 30 °C with SRT of 12 days, which was 2.7 times higher than at 10 °C (total VFA concentration was only 0.071 g/g-VSS). However, VFA production did not signiﬁcantly increase at temperature above 30 °C. Zhuo et al. (2012) set experiments at a series of different temperatures ranging from 10 to 55 °C and came to a different conclusion from that of Cai et al. (2009). It was indicated in their studies that, at 37 °C, VFA accumulation reached the maximum amount compared to 55 °C after 3 days fermentation. All these results of studies are listed in Supplementary Table S2.

Different results might come from different fermentation time. Cai et al. (2009) obtained data every 2 days for 30 days, while Zhuo et al. (2012) got their results in a span of 7 days. Cai et al. (2009) might miss some phenomena because of their long interval time and Zhuo et al. (2012) did not observe later days of fermentation because of their short experiment duration. Thus, in order to find out the optimal temperature, more detailed and comprehensive research should be done regarding the influence of temperature on VFA production of excess sludge in the future.

#### Mixing

Mixing increases the contact surface area between the substrates and microbes; thus substrates are more evenly available to microbes during fermentation. Studies by Perot et al. (1988) showed that when a large production of VFA is required, agitation speed must be greatly increased to the range of 350–650 rpm. According to the study of Perot et al. (1988) in order to obtain maximum VFA production, mixing speed should be chosen based on certain anaerobic digestion conditions. Perot et al. (1988) pointed out the optimal agitation speed should be 545 rpm when fermentation pH was 6.8 at a temperature of 50 °C. Yuan et al. (2011) studied the effect of mixing during fermentation of excess sludge to produce VFA and found that mixing at a speed of 50 rpm not only improved VFA production, but also shortened fermentation time. The above studies are all listed in Supplementary Table S2. Research has seldom been conducted in the area of the effect of mixing on VFA production of excess sludge, and the optimal agitation speed is not yet known. However, from the view of theory, mixing can improve VFA production of excess sludge; so further detailed studies related to the area mentioned above should be pursued. It should be emphasized that although there is significant difference between the mixing in laboratory equipment and an industrial one, we cannot give a definite conclusion about the mixing uding industrial equipment because the most current studies of mixing were conducted in a laboratory scale.

Additives, such as sodium dodecylbenzenesulfonate (SDBS), β-cyclodextrin (β-CD) and anthraquinone-2,6-disulfonate (AQDS), which can change surface properties and improve dissolubility of excess sludge, is an emerging field to study in the area of enhancing VFA production during anaerobic digestion of excess sludge.

Studies by Jiang et al. (2007) showed that surfactants enhanced VFA production during anaerobic digestion of excess sludge. In the presence of SDBS at a concentration of 0.02 g/g dry sludge, after 6 days fermentation, VFA accumulation reached 0.24 g/g-VSS, which was 7.66 times higher than in the absence of SDBS. Jiang et al. (2007) explained that it was because of SDBR's biological effects, which made EPS solubilize and able to dissolve into aqueous phase. In addition, SDBS can also inhibit the methanogenesis process, so the VFA production was improved significantly. The related investigations indicated that at lower SDBS dosage, such as 0.01 g/g, there was only slight decrease of methane production with no significant lag-time appearing, whereas the lag-phase of methane generation increased from 2 days at SDBS 0.01 g/g to 8 days at SDBS 0.02 g/g, meaning that the activity of methanogens was inhibited by SDBS. It can be seen that the decrease of methane always coincides with the increase of VFA production, which indicates that one main reason for improved VFA accumulation in WAS fermentation was the decreased VFA consumption by methanogens. Zhang et al. (2009b) investigated VFA accumulation during excess sludge anaerobic fermentation in semi-continuous reactors and found that SDBS increased VFA accumulation to 0.14 g/g-VSS, which was 4.66 times higher than the blank test. Another study by Zhang et al. (2010) showed that, no matter whether in ambient, mesophilic or thermophilic fermentation, VFA accumulations were enhanced remarkably at SDBS of 0.02 g/g. Studies by Yang et al. (2012a) indicated that VFA production was greatly increased in the presence of β-CD. At 0.2 g-(β-CD)/g-(dried solids), the peak production of VFA per gram of β-CD per gram of dried solids reached 4.2 g, and at 0.3 g/g-(β-CD), the maximum production rate of VFA was 0.2 g/g-VSS, about four times that of the blank sample. Another study by Yang et al. (2012b) showed that AQDS also had the function of enhancing VFA production. With the presence of AQDS, amino acid degradation was accelerated and methanogenic activities were inhibited, which led to VFA production enhancement. All the conditions and results are summarized in Supplementary Table S2.

Studies by Jiang et al. (2007), Zhang et al. (2009b) and Yang et al. (2012a, b) pointed out new ways for researchers to study. However, except for SDBS ($2.0/100 g), β-CD ($3.7/100 g) and AQDS (\$1.0/100 g) there might be other cheap and environmentally friendly additives which could be used to enhance VFA production during anaerobic digestion of excess sludge. Finding new and effective additives for VFA production would be a wide and promising field to study in the future.

#### Solids retention time

It was found in the studies of Yuan et al. (2011, 2009) that VFA yields increased with SRT. They fermented excess sludge with an SRT of 5, 7 and 10 days and found that the highest VFA accumulation was 0.14 g/g-VSS with a fermentation time of 10 days. It was noted in the studies of Jiang et al. (2007) that the SRT of VFA production was 6–15 days. Studies by many researchers indicated that pre-treatment could shorten SRT of sludge to 2–3 days (Yan et al. 2010; Morgan-Sagastume et al. 2011). Studies by Chen et al. (2007), Cai et al. (2009) and Zhang et al. (2009b) showed that the average SRT was 5–11 days, but under strong alkaline conditions SRT was extended. Feng et al. (2009) noted in their studies that the optimal SRT of VFA production of excess sludge at pH 10 and 20 °C was 12 days. Studies by Zhang et al. (2009b) showed that SRT of VFA production during mesophilic fermentation was shorter than thermophilic fermentation, whereas Zhuo et al. (2012) came to the conclusion that mesophilic fermentation and thermophilic fermentation had the same SRT, and fermentation at 10 °C had longer SRT. The difference between results might be due to the difference in substrates. In the studies of Yuan et al. (2011), it was found that mixing could shorten SRT for VFA production. Studies by Jiang et al. (2007) indicated that the addition of surfactants could influence SRT of VFA production of excess sludge. Supplementary Table S2 shows all the detailed information.

Conclusion can be drawn that the SRT of VFA production of excess sludge according to the literatures is usually within 15 days, but longer SRT leads to bigger digesters and higher implementation costs. Besides, fermentation temperature, pH, mixing and additives have great influences on SRT, so the specific SRT should be different and related to its operational parameters and the recommended SRT may be less than 5 days.

## MICROORGANISMS RELATED TO VFA PRODUCTION

Anaerobic digestion is a biological decomposition process, so the production of VFA is closely related to species, activity and population of microorganisms. In an appropriate environment (suitable temperature, pH and so on) certain microbes could, to a great extent, enhance the rate and the amount of VFA produced.

### Key bacteria involved in VFA production

Shin et al. (2010) detected the shift of bacteria during anaerobic digestion of excess sludge and various bacteria related to VFA production were found, such as the Fusibacter-related, the Dethiosulfatibacter aminovorans-related, Clostridium aminobutyricum and the Clostridium-like organisms. In the study of Kim et al. (2013) it was found that Clostridium-related populations were the dominant acidogens. It was reported in literature that carbohydrates were used by the Fusibacter-related organisms to produce acetate and butyrate (Basso et al. 2009), while amino acids were utilized by C. aminobutyricum and the Clostridium-like organisms (Shin et al. 2010). Dethiosulfatibacteraminovorans-related organisms could ferment various organic compounds and produce acetate and propionate (Takii et al. 2007). Ariesyady et al. (2007) investigated bacteria in a full-scale anaerobic sludge digester and found 393 clones (90 operational taxonomic units) belonging to the domain bacteria, among which 21% belonged to Firmicutes. Furthermore, they also found various glucose-utilizing bacteria such as Actinobacteria, Bacteroidetes, Alphaproteobacteria, Chloroﬂexi, Spirochaeta and TM 7 candidate phylum. Studies by Tan et al. (2012) indicated that albumin and Gram-negative bacteria, such as cocci, bacilli and most of the filamentous bacteria, could remarkably enhance the VFA production of excess sludge.

### Detection technologies for microorganisms

Detection of microorganisms in anaerobic digestion for VFA production is a complex study. Common technologies used to analyze microorganisms are polymerase chain reaction (PCR)-based 16S rRNA gene cloning, ﬂuorescent in situ hybridization and dot blot hybridization. Recently, a new technology called tag-pyrosequencing was invented and Forrest et al. (2012) have used it to identify the bacterial communities in the fermentations. Comparing to (PCR)-based 16S rRNA gene cloning, tag-pyrosequencing does not need a specific primer and the information about community composition.

Although many species of bacteria and their function mechanisms are already known to us, many still remain to be discovered. Recently studies which related to VFA production usually focused on mechanical process control; little attention has been paid to microorganisms and their function. Thus, detailed researchs should be done to investigate types and functions of microbes which could improve VFA production of excess sludge. In addition, new technology such as tag-pyrosequencing should be brought in to detect microorganisms.

## CONCLUSION

The functions of thermal, alkaline, microwave and ultrasonic pre-treatment to enhance VFA production during anaerobic digestion was focused on because of the fact that they have been commonly used to enhance the hydrolysis process in recent years. The higher the thermal pre-treatment temperature, the shorter the pre-treatment time would be. However, higher temperature also requires more energy input, which also needs to be taken into account in a practical project. Alkaline pre-treatment, compared with other pre-treatment methods, utilizes a simple device and operation system. However, comparing to thermal and alkaline pre-treatment methods, microwave and ultrasonic pre-treatment methods, due to their advantages of efficiency and environment friendliness, are kinds of emerging and promising technologies, but the full-scale practices of microwave and ultrasound technologies have been tried and failed. Further studies of the pilot-scale or full-scale practices of microwave and ultrasound technologies should be investigated. During the anaerobic digestion process for VFA production, recent studies showed that VFA production in an alkaline condition was much higher than that in an acid condition, which was contrary to former studies. In order to find out the optimal temperature, more detailed and comprehensive research should be done regarding the influence of temperature on VFA production of excess sludge in the future. From the view of theory, mixing can improve VFA production of excess sludge, so further detailed studies related to this area should be pursued. Recent studies also found that additives have great impact on enhancing VFA production, and additives would be a promising area to study. Also, fermentation temperature, pH, mixing and additives have great influences on SRT, so the specific SRT should be different and related to its operational parameters. Finally, anaerobic digestion is a biological decomposition process, so the production of VFA is closely related to species, activity and population of microorganisms. It is necessary to compare the different detection technologies to analyze the key microorganisms involved in VFA production.

## ACKNOWLEDGEMENTS

This work was financially supported by the National Hi-Tech Research and Development Program of China (863 Program) (2012AA063502), the National Science Foundation of China (51408419), and the 12th National Five-Year-Plan of Key Science and Technology (2012BAJ21B01).

## REFERENCES

REFERENCES
Appels
L.
Baeyens
J.
Degrève
J.
Dewil
R.
2008
.
Prog. Energy Combust. Sci
.
34
,
755
781
.
Appels
L.
Degrève
J.
Bruggen
B. V.
Impe
J. V.
Dewil
R.
2010
.
Bioresour. Technol
.
101
,
5743
5748
.
Appels
L.
Lauwers
J.
Degrève
J.
Helsen
L.
Lievens
B.
Willems
K.
Impe
J. V.
Dewil
R.
2011
.
Renew. Sustain. Energy Rev
.
15
,
4295
4301
.
Appels
L.
Houtmeyers
S.
Degrève
J.
Impe
J. V.
Dewil
R.
2013
.
Bioresour. Technol
.
128
,
598
603
.
H. D.
Ito
T.
Okabe
S.
2007
.
Water Res
.
41
,
1554
1568
.
Basso
O.
Lascourreges
J. F.
Borgne
F. L.
Goff
C. L.
Magot
M.
2009
.
Res. Microbiol
.
160
,
107
116
.
Batstone
D. J.
Keller
J.
Angelidaki
I.
Kalyuzhnyi
S. V.
Pavlostathis
S. G.
Rozzi
A.
Sanders
W. T. M.
Siegrist
H.
Vavilin
V. A.
2002
Anaerobic Digestion Model No. 1
.
Sci. Tech. Report 13
,
IWA Publishing
,
London
,
UK
.
Bougrier
C.
Delgenès
J. P.
Carrèrè
H.
2007
.
Biochem. Eng. J.
34
,
20
27
.
Bougrier
C.
Delgenès
J. P.
Carrèrè
H.
2008
.
Chem. Eng. J
.
139
,
236
244
.
Braguglia
C. M.
Gagliano
M. C.
Rossetti
S.
2012
.
Bioresour. Technol
.
110
,
43
49
.
Cai
M. M.
Chua
H.
Zhao
Q. L.
Sin
N. S.
Ren
J.
2009
.
Bioresour. Technol
.
100
,
1399
1405
.
Carrère
H.
Dumasa
C.
Battimelli
A.
Batstone
D. J.
Delgenèsa
J. P.
Steyera
J. P.
Ferrer
I.
2010
.
J. Hazard. Mater
.
183
,
1
15
.
Chen
Y. G.
Jiang
S.
Yuan
H. Y.
Zhou
Q.
Gu
G. W.
2007
.
Water Res
.
41
,
683
689
.
Doğan
I.
Sanin
F. D.
2009
.
Water Res
.
43
,
2139
2148
.
Elefsiniotis
P.
Oldham
W. K.
1994
.
J. Chem. Technol. Biotechnol
.
60
,
89
96
.
Eskicioglu
C.
Terzian
N.
Kennedy
K. J.
Droste
R. L.
Hamoda
M.
2007
.
Water Res
.
41
,
2457
2466
.
Feng
L. Y.
Wang
H.
Chen
Y. G.
Wang
Q.
2009
.
Bioresour. Technol
.
100
,
44
49
.
Ferrer
I.
Ponsá
S.
Vázquez
F.
Font
X.
2008
.
Biochem. Eng. J
.
42
,
186
192
.
Forrest
A. K.
Hollister
E. B.
Gentry
T. J.
Wilkinson
H. H.
Holtzapple
M. T.
2012
.
Bioresour. Technol
.
118
,
343
349
.
Gavala
H. N.
Yenal
U.
I. V.
Westermann
P.
Ahring
B. K.
2003
.
Water Res
.
37
,
4561
4572
.
Guo
L.
Li
X. M.
Bo
X.
Yang
Q.
Zeng
G. M.
Liao
D. X.
Liu
J. J.
2008
.
Bioresour. Technol
.
99
,
3651
3658
.
Jiang
J.
Ma
X. Q.
2011
.
Appl. Therm. Eng
.
31
,
3897
3903
.
Jiang
S.
Chen
Y. G.
Zhou
Q.
Gu
G. W.
2007
.
Water Res
.
41
,
3112
3120
.
Kim
M.
Alan
Y. H.
Speece
R. E.
2002
.
Water Res
.
36
(
17
),
4369
4385
.
Kim
J.
Park
C.
Kim
T. H.
Lee
M.
Kim
S.
Kim
S. W.
Lee
J.
2003
.
J. Biosci. Bioeng
.
95
(
3
),
271
275
.
Kim
J.
Lee
S.
Lee
C.
2013
.
Bioresour. Technol
.
129
,
495
505
.
Laﬁtte-Trouqué
S.
Forster
C. F.
2002
.
Bioresour. Technol
.
84
,
113
118
.
Li
H.
Jin
Y. Y.
Mahar
R. B.
Wang
Z. Y.
Nie
Y. F.
2008
.
Bioresour. Technol
.
99
,
5140
5144
.
Li
H.
Li
C. C.
Liu
W. J.
Zou
S. X.
2012
.
Bioresour. Technol
.
123
,
189
194
.
Liu
D.
Zeng
R. J.
Anelidaki
I.
2006
.
Water Res
.
40
,
2230
2236
.
Liu
X. L.
Liu
H.
Chen
J. H.
Du
J. C.
Chen
J.
2008
.
Waste Manage
.
28
,
2614
2622
.
Lu
J.
Gavala
H. N.
I. V.
Z.
Ahring
B. K.
2008
.
J. Environ. Manage
.
88
,
881
889
.
Ma
H.
Zhang
S.
Lu
X.
Xi
B.
Guo
X.
Wang
H.
Duan
J.
2012
.
Bioresour. Technol
.
116
,
441
447
.
M.
Holm-Nielsen
J. B.
Esbensen
K. H.
2011
.
Renew. Sustain. Energy Rev
.
15
,
3141
3155
.
McCarty
P. L.
Smith
D. P.
1986
Anaerobic wastewater treatment
.
Environ. Sci. Technol
.
20
,
1200
1206
.
Menéndez
J. A.
Inguanzo
M.
Pis
J. J.
2002
.
Water Res
.
36
,
3261
3264
.
Moon
H.
Wang
H.
Jang
N. J.
Hyun
S. H.
Kim
I. S.
2004
.
J. Biotechnol
.
111
,
297
309
.
Morgan-Sagastume
F.
Pratt
S.
Karlsson
A.
Cirne
D.
Lant
P.
Werker
A.
2011
.
Bioresour. Technol
.
102
,
3089
3097
.
Mouneimne
A. H.
Carrère
H.
Bernet
N.
Delgenès
J. P.
2003
.
Bioresour. Technol
.
90
(
1
),
89
94
.
Nyberg
U.
B.
Aspegren
H.
1996
.
Water Sci. Technol
.
33
(
12
),
109
116
.
Perot
C.
Sergent
M.
Richard
P.
Phan Tan Luu
R.
Millot
N.
1988
.
Environ. Technol. Lett
.
9
,
741
752
.
Pilli
S.
Bhunia
P.
Yan
S.
LeBlanc
R. J.
Tyagi
R. D.
Surampalli
R. Y.
2011
.
Ultrason. Sonochem
.
18
,
1
18
.
Quarmby
J.
Scott
J. R.
Mason
A. K.
Davies
G.
Parsons
S. A.
1999
.
Environ. Technol
.
20
,
1155
1161
.
Sachs
J.
Meyer
U.
Rys
P.
Feitkenhauer
H.
2003
.
Water Res
.
37
,
973
982
.
Saha
M.
Eskicioglu
C.
Marin
J.
2011
.
Bioresour. Technol
.
102
,
7815
7826
.
Salsabil
M. R.
Laurenta
J.
Casellasa
M.
Dagot
C.
2010
.
J. Hazard. Mater
.
174
,
323
333
.
Shehu
M. S.
Manan
Z. A.
Alwi
S. R. W.
2012
.
Bioresour. Technol
.
114
,
69
74
.
Shin
S. G.
Lee
S.
Lee
C.
Hwang
K.
Hwang
S.
2010
.
Bioresour. Technol
.
101
,
9461
9470
.
Sólyom
K.
Mato
R. B.
Pérez-Elvira
S. I.
Cocero
M. J.
2011
.
Bioresour. Technol
.
102
,
10849
10854
.
Song
Y. C.
Kwon
S. J.
Woo
J. H.
2004
Mesophilic and thermophilic temperature co-phase anaerobic digestion compared with single-stage mesophilic and thermophilic digestion of sewage sludge
.
Water Res
.
38
(
7
),
1653
1662
.
Takii
S.
S.
Tamaki
H.
Ueno
Y.
Sekiguchi
Y.
Ibe
A.
Matsuura
K.
2007
.
Int. J. Syst. Evol. Micr
.
57
,
2320
2326
.
Tan
R.
Miyanaga
K.
Uy
D.
Tanji
Y.
2012
.
Bioresour. Technol
.
118
,
390
398
.
Tang
B.
Yu
L. F.
Huang
S. S.
Luo
J. Z.
Zhuo
Y.
2010
.
Bioresour. Technol
.
101
,
5092
5097
.
Tiehm
A.
Nickel
K.
Zellhorn
M.
Neis
U.
2001
.
Water Res
.
35
,
2003
2009
.
Tomei
M. C.
Braguglia
C. M.
Mininni
G.
2008
.
Bioresour. Technol
.
99
,
6119
6126
.
Toreci
I.
Kennedy
K. J.
Droste
R. L.
2009
.
Water Res
.
43
,
1273
1284
.
Wang
Q.
Kuninobu
M.
Ogawa
H. I.
Kato
Y.
1999
.
Biomass Bioenergy
16
,
407
416
.
Yan
Y. Y.
Feng
L. Y.
Zhang
C. J.
Wisniewski
C.
Zhou
Q.
2010
.
Water Res
.
44
,
3329
3336
.
Yang
X. Y.
Wang
X.
Wang
L.
2010
.
Bioresour. Technol
.
101
,
2580
2584
.
Yang
X.
Du
M. A.
Lee
D.
Wan
C. L.
Zheng
L. N.
Li
G. Y.
Chang
J.
2012a
,
Bioresour. Technol
.
110
,
688
691
.
Yang
X.
Du
M. A.
Lee
D.
Zheng
L. N.
Li
G. Y.
Wan
F.
2012b
.
Bioresour. Technol
.
103
,
494
497
.
Yu
H. Q.
Zheng
X. J.
Hu
Z. H.
Gu
G. W.
2003
High-rate anaerobic hydrolysis and acidogenesis of sewage sludge in a modified upflow reactor
.
Water Sci. Technol
.
48
,
69
75
.
Yu
G. H.
He
P. J.
Shao
L. M.
He
P. P.
2008
.
Water Res
.
42
,
4637
4644
.
Yu
Q.
Lei
H. Y.
Yu
G. W.
Feng
X.
Li
Z. X.
Wu
Z. C.
2009
.
Chem. Eng. J
.
155
,
88
93
.
Yuan
Q.
Sparling
R.
Oleszkiewicz
J. A.
2009
.
Water Res
.
43
,
5180
5186
.
Yuan
Q.
Sparling
R.
Oleszkiewicz
J. A.
2011
.
Chemosphere
82
,
603
607
.
Zhang
P.
Chen
Y. G.
Zhou
Q.
2009a
.
Water Res
.
43
,
3735
3742
.
Zhang
P.
Chen
Y. G.
Huang
T. Y.
Zhou
Q.
2009b
.
Chem. Eng. J
.
148
,
348
353
.
Zhang
P.
Chen
Y. G.
Zhou
Q.
2010
.
Bioresour. Technol
.
101
,
6902
6909
.
Zhu
H. G.
Parker
W.
Conidi
D.
Basnar
R.
Seto
P.
2011
.
Bioresour. Technol
.
102
,
7086
7092
.
Zhuo
G. H.
Yan
Y. Y.
Tan
X. G.
Dai
X. H.
Zhou
Q.
2012
.
J. Biotechnol
.
159
,
27
31
.