Upgrading of anaerobic digestion of waste activated sludge by a hyper-thermophilic–mesophilic temperature-phased process with a recycle system

Abstract

In order to upgrade the conventional mesophilic anaerobic digestion (MD) of waste activated sludge (WAS), a hyper-thermophilic (70 °C, stage I)–mesophilic (35 °C, stage II) temperature-phased anaerobic digestion system with a recycle system (TPAD-R) was constructed, with MD as a control. The accumulation of solids increased with shorter hydraulic retention time (HRT). Compared with MD, TPAD-R improved the reduction of solids by over 10%, and the removal rate of protein in TPAD-R increased by more than 20%. In stage I of TPAD-R an amount of organic matter was solubilized, in a range between 10 g L⁻¹ and 20 g L⁻¹, as soluble chemical oxygen demand (COD), and played an important role in enhancing hydrolysis and acidification. The specific hydrolysis and acidification rates of stage I reached their maximum values after a HRT of 10 days, 2.367 g COD per g VS per day and 1.120 g COD per g VS per day, respectively. Consequently, the methane yield in TPAD-R, primarily produced in stage II, was also improved, 29% higher than that in MD for a HRT of 10 days. Besides compensating for energy loss, TPAD-R obtained a higher net energy than MD, achieving a net energy which was 3.3 kJ g⁻¹ VS more than that in MD for a HRT of 10 days. The TPAD-R proved to be efficient in upgrading the MD of WAS.

---

1. Introduction

Sewage sludge, as a byproduct in wastewater treatment, is generated in wastewater treatment plants (WWTPs). Its disposal has become a growing problem. The anaerobic digestion process of sewage sludge has been widely used for decades to reduce the sludge volume and obtain energy in the form of biogas.¹ Compared with primary sludge, the anaerobic digestibility of waste activated sludge (WAS) is about half, possibly due to the recalcitrant nature of various cellular components,^2–4 and the organic fraction of WAS is only about 30% to 45% digestible in conventional mesophilic anaerobic treatment (MD).^2,5,6 Moreover, with the increased requirement for nutrient removal, in order to achieve enhanced nitrification and de-nitrification, a number of WWTPs have removed the primary sedimentation process. As a result, only WAS is produced in these plants, resulting in the production of greater volumes of WAS, and biological sludge shows a lower tendency to biodegrade and thus lower biogas yields, compared to the activated sludge treatment for carbon removal only,^7,8 especially if long retention times are applied in the activated sludge process.⁶ Nevertheless, the ultimate biodegradability of WAS is reasonably high.² Therefore, there is a need to develop an innovative approach to upgrade the anaerobic digestion of WAS in order to achieve improved biodegradation and biogas production.

In the treatment of WAS, a waste with a high solids content, anaerobic digestion is generally limited by the rate of hydrolysis of particulates to soluble substrates.⁹ Therefore, the critical issue for upgrading the MD of WAS is to enhance the hydrolysis. Many pre-treatment methods, such as mechanical, thermal, chemical, biological, and a combination of these,^10–12 have proved effective in making suspended substrates more accessible for the anaerobic bacteria, optimizing the methanogenic potential.³ Among these pre-treatment methods, thermal pre-treatment is suitable for not only the improvement of stabilization, but also the enhancement of dewatering of the sludge and the reduction of the number of pathogens at relatively low cost.¹¹ However, higher temperature pre-treatment has high energy requirements and operating difficulties. Thus, thermal pre-treatment below 100 °C is accepted as appropriate.¹³ Pre-treatment at 70 °C prior to the MD showed positive effects on the enhanced solubilisation of WAS and the methane production in the subsequent MD.^13,14 However, substantial acidification in the pre-treatment stage can possibly occur, and biogas production is nearly impossible in that stage.

A recycling of the effluent from the end stage to the front stage can supplement the alkalinity to the pre-treatment step, and reduce the serious acidification.¹⁵ Also, the introduction of the recycle system resulted in a diluted influent and microorganisms being supplemented from the end stage.^16,17 Thus, it is possible that biogas production becomes available in the front stage, and further process upgrading is achieved. The process feasibility in treating high-solid waste has been demonstrated.^16,18 Unfortunately, research utilizing a two-stage process with a recycle system to degrade WAS can hardly be found. In addition, the studies in the past, on the basis of a thermophilic–mesophilic two-stage process, focused on introducing the recycle system from the end stage to a thermophilic stage, and little attention was paid to hyper-thermophilic–mesophilic temperature-phased anaerobic digestion.

Based on the above, a hyper-thermophilic–mesophilic temperature-phased anaerobic digestion system with a recycle system (TPAD-R) was constructed, with MD as a control, to compare the performance and the degradation characteristics of TPAD-R with MD under different hydraulic retention times (HRTs) for treatment of concentrated WAS, and to evaluate the feasibility of utilizing TPAD-R to upgrade the treatment of WAS.

2. Experimental section

2.1 Substrate

The substrate fed into the system was the concentrated WAS, obtained from the Senshio purification centre, Tagajo, Miyagi, Japan. The sludge fetched from the WWTP was soon transferred and stored in a 4 °C refrigerator to guarantee the stable characteristics of the substrate. The characteristics of the WAS in the steady state under the conditions of Run 1, Run 2 and Run 3 are shown in Table 2. In the composition of the WAS, protein was the main component, accounting for about 40%.

2.2 Experiment systems

Laboratory-scale TPAD-R and MD systems were constructed. The MD system had a working volume of 5 litres. For TPAD-R, the front hyper-thermophilic stage (stage I) and the latter mesophilic stage (stage II) had working volumes of 3 litres and 12 litres, respectively. Thus the total volume of the two-stage system was 15 litres. The schematic diagram of the designed system is illustrated in Fig. 1. The whole TPAD-R system consisted of three parts: a feeding tank, two continuously stirred tank reactors, and a gas measuring unit. Each reactor had its own gas measuring unit. The main part of the gas measuring unit was the wet gas meters (WNK-0.5, Shinagawa Corporation, Japan). The temperature of the feeding tank was controlled by a water jacket and a cooler. The temperatures of the two systems were maintained by the water jacket and heaters. Each reactor was equipped with a thermometer, fully submerged in the reactants, to indicate the temperature in the reactors. Reactors were fed semi-continuously using peristaltic pumps. The MD was fed WAS from a feeding tank. A recycle system was introduced from stage II to stage I of the TPAD-R. Stage I of the TPAD-R was fed with WAS from the same feeding tank as for MD and the effluent from stage II of TPAD-R, and stage II was fed the effluent from stage I.

Fig. 1 Schematic diagrams of the MD and TPAD-R systems constructed in this study.

2.3 Operational conditions

The work in this study was conducted on the basis of the previous study with the front stage at 55 °C. Both of the systems had attained steady state under the previous conditions. The whole operation period was divided into three stages, Run 1, Run 2 and Run 3 with different HRTs. The corresponding HRTs of Run 1, Run 2 and Run 3 were 30 days, 20 days and 10 days, respectively. The HRTs mentioned here were for the whole system, regardless of the two-stage process with or without a recycle system. The real HRT for each reactor in the TPAD-R was calculated and is shown in Table 1. The temperatures of stage II of the TPAD-R and MD were always maintained at 35 °C. The temperature of stage I was always kept at 70 °C. The water bath for the feeding tank was set at 4 °C using a cooler. All of the reactors were operated in a semi-continuous manner by feeding and withdrawing substrate 5–20 times per day, depending on the flow-rate of the influent. The recycle ratio from stage II of TPAD-R to stage I of TPAD-R was 1 : 1. The operational conditions in the experiment are shown in Table 1.

Table 1 Operational conditions for the MD and TPAD-R systems, and in stage I and stage II of TPAD-R for the different runs

	Run 1				Run 2				Run 3
	MD	TPAD-R			MD	TPAD-R			MD	TPAD-R
		I	II	I + II		I	II	I + II		I	II	I + II
Reactor volume (L)	5	3	12	15	5	3	12	15	5	3	12	15
Temperature (°C)	35	70	35	—	35	70	35	—	35	70	35	—
HRT (days)	30	3	12	30	20	2	8	20	10	1	4	10
TS loading (g per L per day)	1.6	12.2	2.6	1.6	2.2	20.1	4.0	2.2	5.0	42.5	10.0	5.0
VS loading (g per L per day)	1.2	8.5	1.7	1.2	1.7	12.5	2.7	1.66	3.80	30.6	6.6	3.80
COD loading (g per L per day)	1.8	13.6	3.4	1.8	2.7	19.4	4.6	2.7	5.8	48.4	11.4	5.8

---

2.4 Sampling and analytical methods

A steady state, defined as a relatively constant VS reduction (5%), was obtained after at least 3 hydraulic turnovers in the reactors. Samples, together with substrate, were taken from the sampling ports of each reactor or substrate tank every three or four days. Data were collected in the steady state at least three times. The volume of biogas, measured using the wet gas meter, was noted every day. Measurements of pH, total solids (TS), volatile solids (VS), COD, NH₄⁺–N, and PO₄³⁻–P were conducted following standard methods.¹⁹

Gas samples were taken using a 0.5 mL syringe equipped with metal hub needles to measure the gas composition. A volume of 0.4 mL of biogas was injected into a gas chromatograph (GC-8A, Shimazu Corporation, Japan) to measure the biogas composition. The temperatures of the injector, detector and column were set at 100 °C, 100 °C and 70 °C, respectively. The carrier gas was argon and its pressure was controlled at 100 kPa. The GC column used for the analysis detected the relative proportions of nitrogen, methane, and carbon dioxide.

The carbohydrate, protein and lipid content, as the main components of COD in WAS, was also analyzed. The measurement of carbohydrate and protein content utilized spectrophotography. The carbohydrate content was measured with a phenol-sulfate examination method.²⁰ The protein content was measured with the traditional Folin-phenol method.²¹ A methanol–chloroform extraction and weight method was used to measure the lipid content.²²

A GC with a flame ionization detector (FID) was utilized to detect volatile fatty acid (VFA) components. Six acids, namely acetic, propionic, iso-butyric, non-butyric, iso-valeric, and non-valeric acid were analyzed. When the total VFA was calculated, each individual VFA was converted to a value of acetic acid, namely HAc. The conversion ratios for acetic, propionic, butyric, and valeric acid were 1, 0.811, 0.682 and 0.588, respectively. To analyze the VFA, the filtrate after centrifugation was filtered through 0.45 μm filters. 0.5 mL filtrate was collected in a 1.5 mL GC vial, and 0.5 mL, 0.1 mol L⁻¹ HCl solution was added to adjust the pH to acid. An Agilent 6890 GC with a FID and equipped with a 30.0 m × 530 μm × 1.00 μm DB-WAXetr column was utilized to analyze the content of the VFA. Nitrogen was the carrier gas and the flow-rate was 2 mL min⁻¹. The injection port and the detector were maintained at 250 °C. The oven of the GC was programmed to begin at 125 °C and hold here for 5 minutes, then to increase the temperature at a rate of 2.5 °C min⁻¹ to 180 °C, and to hold at 180 °C for 5 minutes. The sample injection volume was 1.0 μL.

2.5 Calculations

The hydrolysis, acidogenesis and methanogenesis were calculated from a COD balance in the anaerobic digestion process. The total COD conversion ratios of hydrolysis, acidogenesis, and methanogenesis were defined, using SCOD, COD_VFA and COD_CH4, as described in eqn (1).¹⁴where massSCOD is the SCOD weight (g); massCOD_CH4 is the CH₄ weight calculated as COD (g); massCOD_VFA is the VFA weight calculated as COD (g); massTCOD is the T-COD weight (g); eff. is effluent; and inf. is influent.

The hydrolysis of the particulate organic matter in each system was characterized using the VS removal and PCOD hydrolysis rates. To obtain the specific rate, the rate of PCOD hydrolysis was divided by the mass of the VS concentration in the reactor as shown in the equation. Because of the difficulty in determining the active biomass concentration, the specific hydrolysis rates were normalized on the basis of the total VS.^14,23

The heat requirements of the digesters consist of that to raise the temperature of the incoming sludge to that of the digestion tank and that to compensate for the heat losses through the walls, floor, and roof of the digester.^24,25 The heat requirements are estimated using eqn (4).

q = CQΔT₁ + UAΔT₂ (4)

where q is the heat loss, J per day; C is the specific heat of the feedstock, J g⁻¹ °C⁻¹; Q is the flow rate to be added, g per day; ΔT₁ is the temperature difference between the incoming sludge and the digestion tank, °C; U is the overall coefficient of heat transfer, J per m⁻² per day per °C; A is the cross-sectional area through which the heat loss is occurring, m²; and ΔT₂ is the temperature drop across the surface in question (°C). Thus, the net energy obtained in the digestion process is calculated using eqn (5).

P_net = P_output − q (5)

where P_net is the net energy production of the system, J per day; and P_output is the energy output, namely all of the produced energy, J per day.

3. Results and discussion

3.1 Operational performance of the process

The time profiles of the pH, TS and VS, gas production rate, gas composition, NH₄⁺–N, and VFA in MD and stage I and stage II of TPAD-R are shown in Fig. 2. According to Fig. 2, except for Run 3, the time profiles of MD in Run 1 and Run 2 show a similar performance characteristic, and every item in Run 1 and Run 2 was maintained in an appropriate range for anaerobic digestion. This indicates that MD with a HRT of 20 days seemed to be more applicable for WAS digestion, due to the short retention time and lack of VFA accumulation. When the HRT was shortened to 10 days, on the one hand, there was no substantial change happening to the pH, which was about 7.0, and the gas production rate obviously increased due to improved organic loading, from 0.45 L per L per day in the HRT to 1.06 L per L per day; on the other hand, the solids concentration in the reactor increased, in addition to the accumulation of about 1600 mg L⁻¹ as HAc of VFA.

Fig. 2 Time profiles of the pH, TS, VS, biogas production rate and composition, NH₄⁺–N and VFA in MD, and stage I and stage II of TPAD-R.

There were analogous results indicated in TPAD-R like MD with a HRT of 30 days and 20 days. Conversely, in Run 3 although the solids concentration in stage II increased compared to that in Run 1 and Run 2, the VFA concentration remained in a low range, about 700 mg L⁻¹ as HAc, and most of it could be consumed. It was inferred that the optimal HRT for TPAD-R to treat WAS should be between 10 days and 20 days. Whichever conditions were utilized, the pH in stage I and stage II of TPAD-R was relatively higher than that in MD. As to the higher pH in stage I, it was possible that the recycle system from stage II with high alkalinity and pH played an important part. Under the conditions of a HRT of 20 days, the pH in stage II of TPAD-R was especially high, attaining 7.61. It was believed that the decrease in alkalinity was mainly caused by acid formation in the anaerobic digestion, while most of the acidification in TPAD-R was accomplished in stage I. Thus, in the following stage II alkalinity utilization was reduced. In addition, the pH increase in stage I of TPAD-R as a result of the alkalinity being supplemented from the recycle system led to the improved pH in the influent of stage II. Because of the dilution from the recycle system and the higher decomposition level of TPAD-R, the solids concentration in each reactor of TPAD-R was lower than for MD. As a consequence, the gas production rate of TPAD-R was improved. Or rather, a higher gas production rate was observed in stage I, about 1.7 times faster than that in stage II, while in stage II the longer retention time (4 times longer than that in stage I) and higher methane content (above 70%) than stage I resulted in most of the methane in TPAD-R being produced in stage II. Through the time profile of the VFA in each reactor of TPAD-R, it could be seen that with the biogas produced in stage I, the VFA concentrations were also in a high range, 2500–4000 mg L⁻¹ as HAc. This phenomenon was also reported in the previous study, which pointed out that increasing the temperature to more than 60 °C will often lead to an increased concentration of VFA in the completely mixed tank reactors.³ Another investigation into the effects of different thermophilic temperatures on the pretreatment of sewage sludge has shown that acidogenesis had the highest product formation when the temperature was at 70 °C or 75 °C, under the conditions of which a higher acidogenic activity was maintained and the methanogenic activity was inhibited.²⁶ The VFA produced in stage I was almost consumed in the following stage II of TPAD-R. NH₄⁺–N is usually formed in anaerobic treatment from the degradation of proteins or urea and its inhibitory concentrations may be approached in highly concentrated municipal waste sludge.^27,28 It has been reported that when the NH₄⁺–N concentration exceeds 3000 mg L⁻¹, then the ammonium ion itself becomes quite toxic regardless of the pH and the process can be expected to fail.²⁸ In this study the NH₄⁺–N concentration in both reactors of TPAD-R was maintained in the range of 2000–2500 mg L⁻¹, suggesting that the NH₄⁺–N concentration in the reactors was appropriate to not cause inhibition, although when the HRT was changed to 10 days NH₄⁺–N immediately accumulated for a time. The averages of the experimental results in the steady states of Run 1, Run 2 and Run 3 are summarized in Table 2.

Table 2 Averages of the experimental results of the digested sludge and biogas in every reactor in the steady state^a

	Items		Run 1				Run 2				Run 3
			WAS	MD	TPAD-R		WAS	MD	TPAD-R		WAS	MD	TPAD-R
					I	II			I	II			I	II
a *‘T’ is ‘Total’. **‘S’ is ‘Soluble’. ***‘HAc’ is ‘acetic acid’. ****‘N.D.’ is ‘not detected’.
Digested sludge	pH		5.80 ± 0.04	7.06 ± 0.07	7.37 ± 0.09	7.55 ± 0.02	5.82 ± 0.01	7.22 ± 0.02	7.50 ± 0.03	7.61 ± 0.02	5.92 ± 0.05	7.03 ± 0.03	7.14 ± 0.03	7.28 ± 0.05
	TS (%)		4.66 ± 0.13	3.19 ± 0.11	3.16 ± 0.14	2.67 ± 0.06	4.30 ± 0.07	3.25 ± 0.07	3.17 ± 0.02	2.73 ± 0.06	5.00 ± 0.05	3.99 ± 0.09	3.82 ± 0.10	3.50 ± 0.07
	VS (%)		3.46 ± 0.20	2.16 ± 0.12	2.08 ± 0.18	1.64 ± 0.12	3.32 ± 0.03	2.17 ± 0.05	2.15 ± 0.04	1.69 ± 0.02	3.80 ± 0.16	2.75 ± 0.04	2.64 ± 0.10	2.32 ± 0.08
	T*-COD (g L^−1)		53.3 ± 0.9	31.3 ± 1.0	40.4 ± 2.6	28.4 ± 2.1	53.7 ± 0.8	33.6 ± 0.9	37.1 ± 0.6	26.9 ± 0.6	58.1 ± 0.9	42.1 ± 1.1	45.7 ± 1.3	38.6 ± 1.1
	SCOD (g L^−1)		7.2 ± 0.7	3.3 ± 0.1	11.4 ± 0.5	4.9 ± 0.2	6.9 ± 0.4	4.1 ± 0.4	15.3 ± 0.5	6 ± 0.8	2.7 ± 0.5	4.5 ± 0.6	19.6 ± 0.4	10.1 ± 0.3
	T-carbohydrate (g L^−1)		5.5 ± 0.7	3.1 ± 0.1	3.9 ± 0.4	2.8 ± 0.2	4.8 ± 0.3	3.6 ± 0.6	3.6 ± 0.2	3.0 ± 0.1	7.9 ± 0.6	5.6 ± 0.2	5.6 ± 0.4	5.0 ± 0.3
	S**-carbohydrate (g L^−1)		0.6 ± 0.0	0.3 ± 0.0	1.0 ± 0.1	0.6 ± 0.0	0.5 ± 0.0	0.5 ± 0.0	1.4 ± 0.2	0.8 ± 0.1	0.2 ± 0.0	0.5 ± 0.0	1.8 ± 0.1	1.1 ± 0.1
	T-protein (g L^−1)		16.8 ± 1.3	11.7 ± 0.5	10.8 ± 1.0	8.2 ± 0.4	14.5 ± 1.0	10.2 ± 0.9	7.1 ± 0.7	5.7 ± 0.5	14.6 ± 1.0	9.9 ± 0.4	7.1 ± 0.7	6.9 ± 0.6
	S-protein (g L^−1)		1.3 ± 0.0	0.9 ± 0.1	2.4 ± 0.2	1.0 ± 0.1	1.1 ± 0.1	1.4 ± 0.1	2.5 ± 0.2	1.4 ± 0.2	0.3 ± 0.0	1.1 ± 0.0	2.5 ± 0.2	1.6 ± 0.1
	T-lipid (g L^−1)		4.0 ± 0.2	2.8 ± 0.5	3.4 ± 0.2	3.2 ± 0.3	4.9 ± 0.3	3.4 ± 0.2	4.3 ± 0.1	3.5 ± 0.1	4.9 ± 0.4	3.3 ± 0.2	4.3 ± 0.1	3.7 ± 0.2
	S-lipid (g L^−1)		3.1 ± 0.1	1.6 ± 0.4	1.8 ± 0.5	1.8 ± 0.2	1.8 ± 0.1	1.8 ± 0.4	1.9 ± 0.2	1.8 ± 0.2	1.0 ± 0.1	1.2 ± 0.4	2.4 ± 0.3	1.7 ± 0.1
	VFA (mg HAc*** L^−1)		1420 ± 280	N.D.****	2590 ± 170	N.D.	1840 ± 230	N.D.	2790 ± 290	N.D.	850 ± 190	1640 ± 220	3840 ± 230	702 ± 40
	Alkalinity (mg CaCO3 L^−1)		1020 ± 100	4190 ± 160	4920 ± 380	6860 ± 110	930 ± 70	4850 ± 80	5760 ± 160	7570 ± 120	660 ± 70	3780 ± 90	4150 ± 120	5530 ± 320
	NH4^+–N (mg L^−1)		430 ± 40	1280 ± 170	2100 ± 110	2220 ± 60	330 ± 20	1700 ± 50	2350 ± 90	2580 ± 60	260 ± 50	1200 ± 100	2420 ± 100	2100 ± 80
Biogas	Production rate (L per L per day)		—	0.45 ± 0.01	0.63 ± 0.03	0.39 ± 0.04	—	0.58 ± 0.01	1.15 ± 0.01	0.63 ± 0.01	—	1.06 ± 0.02	2.08 ± 0.02	1.19 ± 0.06
	Composition	CH4 (%)	—	60.6 ± 0.2	36.7 ± 1.1	70.9 ± 0.2	—	62.8 ± 0.3	35.8 ± 1.1	72.4 ± 0.2	—	60.0 ± 0.4	36.4 ± 0.4	70.1 ± 1.5
		CO2 (%)	—	38.9 ± 0.5	62.1 ± 1	28.3 ± 0.7	—	36.7 ± 0.3	63.0 ± 1.5	27.5 ± 0.2	—	39.4 ± 0.7	62.4 ± 0.5	29.4 ± 1.3
		H2S (ppm)	—	1180 ± 20	2010 ± 10	600 ± 0	—	1340 ± 110	2360 ± 110	1300 ± 80	—	1520 ± 120	1860 ± 170	1060 ± 140

---

3.2 Removal and solubilisation

In the process of anaerobic fermentation, solids and organics are degraded and biogas is produced accordingly. Indeed, TS and VS reduction is one of the important factors indicating stabilization of the sludge. The results of the removal rates of the solids and organic matter in the MD, TPAD-R systems and in stage I and stage II of TPAD-R are shown in Table 3. It was reported that conventional anaerobic digestion of activated sludge could exhibit VS reduction of up to only 40%.² This VS removal level was consistent with the results of MD in this study, 37.6% in Run 1, 34.6% in Run 2 and 27.6% in Run 3. The solids reduction in either MD or TPAD-R decreased with a shorter HRT nearly linearly, from 31.5% to 20.2% for MD and from 42.6% to 30.0% for TPAD-R, which indicated that a long retention time did benefit solids reduction. Compared with TS reduction, when the HRT became longer than 20 days, the extent to which VS reduction increased, about 3%, was less than that for when the HRT was changed from 10 days to 20 days, over 7%. Besides, in the whole experiment process, the level of solids reduction in TPAD-R was consistently over 10% higher than that in MD in terms of either TS or VS, which exhibits the superiority of utilizing TPAD-R to degrade the solids fraction of WAS. It seemed that in stage I, with a partial solids reduction, the amount of original solids was only converted to being more biodegradable and was not removed from the reactor, but it resulted in higher solids reduction percentages in the following stage II.

Table 3 Removal rates of solids and organic matter in the MD and TPAD-R systems, and in stage I and stage II of TPAD-R for the different runs

Removal rate (%)	Run 1				Run 2				Run 3
	MD	TPAD-R			MD	TPAD-R			MD	TPAD-R
		I	II	I + II		I	II	I + II		I	II	I + II
TS	31.5	13.8	15.5	42.6	24.4	9.8	13.9	36.5	20.2	10.1	8.4	30.0
VS	37.6	18.4	21.2	52.5	34.6	14.2	21.4	49.1	27.6	13.7	12.1	38.9
T-COD	41.3	1.1	29.7	46.7	37.4	7.9	27.5	49.9	27.5	5.5	15.5	33.6
T-carbohydrate	43.6	6.0	28.2	50.0	24.2	8.3	16.0	37.4	29.3	13.0	11.2	37.0
T-protein	30.4	13.6	24.1	50.9	29.7	29.7	19.7	60.7	32.2	34.0	2.8	52.7
T-lipid	30.0	5.6	5.9	18.7	30.6	0.0	18.6	28.6	32.7	0.0	14.0	24.5

---

With the VS degradation, different solid organic matter was firstly solubilized and further degraded and converted into biogas. Actually, a large number of microbial species are capable of utilizing organic substrates such as carbohydrates, proteins, and lipids to produce VFA and other soluble carbon compounds via a variety of anaerobic metabolic pathways. But since complex biopolymers cannot penetrate the cell membrane, bacteria excrete enzymes that hydrolyze the particulate substrate into small transportable molecules.^29–31 In general, polysaccharides are converted into simple sugars, protein is hydrolyzed to amino acids, and lipid metabolism results in the production of fatty acids and a variety of other organic compounds.³² Thus, the organic matter reduction and solubilization could always reflect the degradation process of anaerobic digestion. In the hydrolysis, the rate-limiting step of WAS digestion,⁹ organic matter is firstly solubilized to become small transportable components. On one hand, the concentration of the solubilized organic matter indicates the hydrolysis capability, even the digestion capability of reactors for WAS; on the other hand, with the solubilization, soluble organic matter is rapidly metabolized into biogas. Consequently, there is a balance between formation and consumption of organic matter.

The reduction and solubilization of COD and its major fractions, carbohydrate, protein and lipid, are shown in Tables 3 and 2, respectively. Except for the results of total lipid reduction between MD and TPAD-R, the other organic matter reduction in TPAD-R was always higher than that in MD. The differences between the two systems for a HRT of 20 days were highest, and the improved percentages of COD, carbohydrate and protein were 12.5%, 13.2% and 31.0%, respectively. Moreover, under the conditions of a HRT of 20 days, while the lipid reduction in TPAD-R remained lower than that in MD, the values between the two systems were close to each other, about 30%. Actually, the COD reduction in stage I was always over 10% lower than that in stage II under whichever HRT condition was used, but the solubilized COD maintained a higher level, more than 2 times higher than that in stage II, which further proved that stage I in TPAD-R played the main role in the degradation of solids into soluble compounds. The experiment with 70 °C pretreatment to degrade WAS showed that solubilization was improved during pretreatment.¹⁴ Another study with pretreatment at 70 °C showed the dissolution of up to 40% of the particulate COD in waste sludge.³³ Furthermore, with the HRT shortened, the concentrations of soluble COD in each reactor increased, and with a HRT of 10 days they accumulated most, 4.5 g L⁻¹ for MD, 19.6 g L⁻¹ for stage I and 10.1 g L⁻¹ for stage II. From the effluents of MD and stage II, the conversions of soluble COD for a HRT of 20 days were more complete and appropriate for TPAD-R optimization, while for a HRT of 20 days the soluble COD in stage II, 6.0 g L⁻¹, remained lower and close to the effluent of MD, 4.1 g L⁻¹. For carbohydrate, the results of organic matter reduction and solubilization were consistent with those of COD. From the results of the WAS characterisation shown in Table 2, protein was always the primary component of the organic matter in WAS, about 15 g L⁻¹ in the original WAS. Stage I still played an important role in protein solubilization, resulting in more soluble protein in the effluent of stage I than in stage II. The HRT seemed not to be influential to the protein solubilization under the experimental conditions in this study, or only a short retention time was needed to solubilize the protein. However, with a short HRT of 10 days for the TPAD-R system, the protein degradation in the following stage II was substantially affected, with only 2.8% removed. Therefore, for TPAD-R the retention time of 4 days in stage II was inappropriate. Lipid exhibited different degradation characteristics. The effects of TPAD-R on lipid degradation seemed weak, and stage II showed a stronger lipid removal ability than stage I. As reported, only carbohydrate and nitrogenous materials were extensively fermented in the acid stage, and lipid was fermented concurrently with the volatile acids in the methanogenic stage.^9,34 The results of lipid reduction in Table 3 also show that almost no lipid was degraded in stage I of TPAD-R.

3.3 Biogas production

In the process of anaerobic digestion, the end product of organic decomposition is biogas. As shown in Table 2, methane and carbon dioxide were the most substantial components of biogas. The methane production in the systems and each reactor of TPAD-R for the different runs is shown in Table 4. In terms of the methane production rate, it always increased with improved organic loading, either for MD or TPAD-R. Under the conditions of the HRT being shortened from 30 days to 10 days, the methane production rates for TPAD-R and MD increased from the same starting point of 0.27 L per L per day to 0.82 L per L per day and 0.64 L per L per day, respectively. There was no substantial difference between TPAD-R and MD under the conditions of Run 1 and Run 2, but when the HRT was changed to 10 days, the methane production rates in each stage of TPAD-R were apparently higher than that in MD. Consequently, with a HRT of 10 days the methane production rate in TPAD-R, 0.82 L per L per day, became higher than that in the single-stage system, 0.64 L per L per day. In addition, the higher methane production rate in TPAD-R led to a higher methane yield than that in MD, and particularly in Run 3 methane yields for TPAD-R and MD were 0.22 L per g VS added and 0.17 L per g VS added, respectively. It is worth pointing out that the longer retention time in stage II than stage I was one reason for the resulting higher methane yield, while the greater amount of biodegradable substances available in stage II also contributed to it. However, in terms of the methane recovery rate, the performance of TPAD-R was more than 10% less than MD. The methane recovery rate in stage II was still at least 3 times higher than that in stage I. From the VS reduction values observed in Table 3, in stage II of TPAD-R, nearly 2 times more VS was degraded in Run 1 and Run 2 than in Run 3, so the methane recovery rates, namely the methane production per unit of VS degraded, in Run 1 and Run 2 exhibited lower values than that in Run 3.

Table 4 Methane production in the MD and TPAD-R systems, and in stage I and stage II of TPAD-R for the different runs

Methane production	Run 1				Run 2				Run 3
	MD	TPAD-R			MD	TPAD-R			MD	TPAD-R
		I	II	I + II		I	II	I + II		I	II	I + II
Production rate (L per L per day)	0.27	0.23	0.28	0.27	0.36	0.41	0.46	0.45	0.64	0.76	0.83	0.82
Yield (L per g VS added)	0.23	0.03	0.16	0.23	0.22	0.03	0.17	0.27	0.17	0.02	0.13	0.22
Recovery rate (L per g VS degraded)	0.63	0.15	0.75	0.44	0.63	0.23	0.79	0.55	0.61	0.18	1.04	0.55

---

3.4 Dewaterability

The sludge-dewatering tests were conducted via a flocculation experiment. In the experiment polyamidine, a polymer flocculant used in WWTPs, was utilized. The dosages of flocculant can determine the reagent consumption costs for dewatering of the digestate. The dewatering filtrate is returned into the wastewater treatment process to be treated again. Consequently, the returned filtrate increases the loading of the wastewater treatment. The fewer pollutants that are included in the dewatering filtrate, the fewer the effects on the WWTP, caused by improved loading. As a result, the reagent dosage and water quality of the filtrate in the digestate can be the indicators of the dewaterability of the effluents from anaerobic digestion systems. In fact, pollutant concentrations vary with the optimal dosages of the flocculant, under which the maximum values of the pollutant exist. The optimal dosages of flocculant and COD, NH₄⁺–N, and VFA of the dewatering filtrate under the optimal dosages of the flocculant are listed in Table 5. In terms of the dosage of reagent needed and the pollutant content in the filtrate, under any of the conditions used in this study, the dewaterability of the MD process showed better characteristics than TPAD-R, which is demonstrated by the lower dosages of the flocculant and lower concentrations of COD and NH₄⁺–N in MD. The dosages of the flocculant for the effluent from TPAD-R were approximately 1.3–1.4 times higher than that from MD. Accordingly, TPAD-R had a slightly weaker capability than the MD process with regard to dewaterability.

Table 5 Dewaterability of the digestates from the MD and TPAD-R systems for the different runs

		Run 1		Run 2		Run 3
		MD	TPAD-R	MD	TPAD-R	MD	TPAD-R
Dosage of flocculant (mg per g TS)		58.1	79.2	56.6	79.3	62.7	81.4
Characteristics of the dewatering filtrate	COD (mg L^−1)	600	770	440	1050	970	1870
	NH4^+–N (mg L^−1)	1200	1510	1160	1850	980	1350
	VFA (mg HAc L^−1)	N.D.	N.D.	N.D.	N.D.	330	610

---

3.5 Mechanism analysis

Based on the experiment results, the COD mass balance was calculated and analyzed. The summarized results are shown in Fig. 3. The calculated results demonstrated that the errors between influent and effluent were controlled within 5%, which proved the system stability and the reliability of the experimental results. Under every experimental condition in this study, compared with MD, TPAD-R exhibited higher solids reduction capability indicated by the particulate COD (PCOD) reduction, and methane production capability. In terms of the methane production in TPAD-R, the amount produced in stage I was always much less than that in stage II, with the differences of 28.6% in Run 1, 30.1% in Run 2 and 25.4% in Run 3. In other words, the introduction of a hyper-thermophilic system enhanced the solids reduction, and improved the total methane production. In addition, from Fig. 3 it can be seen that the methane percentage for MD and stage II of TPAD-R was nearly the same, approximately 36% in Run 1, 38% in Run 2 and 32% in Run 3, and the improvement is derived from stage I of TPAD-R, which, to an extent, shows the effect of the recycle system. It is possible that the methane producing archaea in stage I were replenished from the recycle system resulting in more methane production in stage I of TPAD-R. Furthermore, the SCOD in the effluent of TPAD-R increased with a shortened HRT, from 9.2% to 17.4%. In particular, under a HRT of 10 days, the difference in the SCOD percentages between MD and TPAD-R attained about 10%.

Fig. 3 Mass balances in the MD and TPAD-R systems, and in stage I and stage II of TPAD-R for the different runs.

As shown in Table 6, either hydrolysis, acidogenesis, or methanogenesis in TPAD-R was higher than those in MD, indicating that digestion was advanced in TPAD-R compared to MD. Or rather, in TPAD-R hydrolysis mainly happened in stage I rather than stage II. In particular, in stage II under a HRT of 10 days, there was almost no hydrolysis. It is safe to say that to a great extent stage I in TPAD-R served as the hydrolysis stage. In fact, with the hydrolysis in stage I, part of the soluble COD was also converted into VFA, leading to acidification, and in stage II the abundance of available biodegradable influent caused more methane production. Moreover, in eqn (1) the COD conversion ratio of acidogenesis included the item of the mass of COD_CH4, which possibly resulted in the acidogenesis ratios between stage I and stage II being nearly the same. As a matter of fact, the results of the methanogenesis ratios shown in Table 6 also show that the methane producing capability of stage II was much higher than that of stage I.

Table 6 Conversion ratios, rates and specific rates of hydrolysis, acidogenesis and methanogenesis in the MD and TPAD-R systems, and in stage I and stage II of TPAD-R for the different runs

		Run 1				Run 2				Run 3
		MD	TPAD-R			MD	TPAD-R			MD	TPAD-R
			I	II	I + II		I	II	I + II		I	II	I + II
Conversion ratios (%)	Hydrolysis	34.2	21.0	10.7	45.2	39.9	34.2	5.2	53.8	36.0	36.6	0.1	55.6
	Acidogenesis	33.8	12.9	14.5	40.7	34.6	15.0	16.6	44.1	36.5	15.5	11.4	41.4
	Methanogenesis	36.9	4.8	23.8	43.4	38.7	5.8	28.1	47.6	31.3	4.5	20.9	40.3
Reaction rates (g COD per L per day)	Hydrolysis	0.526	2.440	0.258	0.695	0.955	5.877	0.141	1.288	1.996	15.363	0.008	3.079
	Acidogenesis	0.573	1.700	0.436	0.689	0.871	2.889	0.663	1.109	2.072	7.268	1.118	2.348
	Methanogenesis	0.656	0.657	0.800	0.771	1.040	1.177	1.303	1.278	1.816	2.163	2.383	2.339
Specific reaction rates (g COD per g VS per day)	Hydrolysis	0.024	0.117	0.016	0.040	0.044	0.273	0.008	0.072	0.220	2.367	0.000	0.127
	Acidogenesis	0.027	0.082	0.027	0.040	0.040	0.134	0.039	0.062	0.228	1.120	0.039	0.097
	Methanogenesis	0.030	0.032	0.049	0.045	0.048	0.055	0.077	0.072	0.200	0.333	0.083	0.097

---

The COD conversion rates and specific conversion rates of hydrolysis, acidogenesis and methanogenesis, were further analyzed using eqn (2) and (3). The calculated results are also shown in Table 6. According to the results, it’s clear that in stage I of TPAD-R, especially under a HRT of 10 days, the hydrolysis and acidogenesis rate and specific rate were considerably increased. The hydrolysis rate and specific hydrolysis rate of stage I in Run 3 were 15.4 g COD per L per day and 2.37 g COD per g VS per day, respectively. Thus, for pre-fermentation at the temperature of 70 °C, in order to enhance the hydrolysis and acidogenesis of WAS anaerobic digestion, the retention time of 1 to 2 days had been adequate. Although the specific hydrolysis and acidogenesis rates in TPAD-R for a HRT of 10 days, caused by the accumulation of solids, were 0.093 g COD per g VS per day and 0.131 g COD per g VS per day lower than those in MD, respectively, which possibly resulted from the weak hydrolysis and acidogenesis rate and specific rate in stage II, because of the enhancement in stage I both the hydrolysis and acidogenesis rates in TPAD-R were improved to 3.079 g COD per g VS per day and 2.348 g COD per g VS per day, respectively. Furthermore, the introduction of the recycle system resulted in stage I of TPAD-R having a certain methanogenesis rate and specific rate. Consequently, the methanogenesis rate of the whole TPAD-R system was always faster than that of MD. In particular, for a HRT of 10 days, the methanogenesis rate in TPAD-R was 0.423 g COD per L per day faster than that in MD.

3.6 Energy balance in the process

Biogas produced as a result of the anaerobic digestion may be used as fuel source. The energy from biogas combustion may be partly used for heating the digester to maintain its temperature and the surplus can be utilized elsewhere.

The calculated results show that the energy produced for every system under all conditions could always create a surplus energy output when the energy balance was obtained. The heat requirements increased with shorter HRTs. The heat requirements for MD increased from 15.0 kJ per g VS degraded in Run 1 to 18.6 kJ per g VS degraded, while the heat requirements for TPAD-R increased from 10.7 kJ per g VS degraded in Run 1 to 13.2 kJ per g VS degraded. The net energy obtained for MD decreased with shorter HRTs, and in Run 3 3.2 kJ per g VS degraded was left. The net energy for TPAD-R fluctuated in a range between 5.6 kJ per g VS degraded and 7.6 kJ per g VS degraded. Overall, TPAD-R achieved a higher net energy than MD. In particular, in Run 3 TPAD-R obtained 3.3 kJ per g VS degraded more than MD.

4. Conclusions

In this experiment, conducted by introducing a hyper-thermophilic process (70 °C, stage I) prior to the mesophilic digestion (35 °C, stage II) and a recycle system from stage II to stage I to treat waste activated sludge (WAS), with conventional mesophilic digestion (MD) as a control, it has been proven that the designed process (temperature-phased anaerobic digestion with a recycle system, TPAD-R) was effective in upgrading the mesophilic digestion of WAS. The solids reduction in MD was lower than 40%, while TPAD-R showed an enhanced solids reduction capability, improving the solids reduction by over 10% for every hydraulic retention time (HRT) studied, 30 days, 20 days and 10 days. The solids reduction percentages increased with prolonged HRTs. Protein, accounting for about 40% of the WAS, was removed most among all of the organic matter in TPAD-R, attaining 60.7% reduction with a HRT of 20 days. By comparison with MD, the methane yield in TPAD-R was also higher, 29% higher than that in MD with a HRT of 10 days, which was mainly produced in stage II of TPAD-R. In stage I of TPAD-R, organic matter was abundantly solubilized, more than 10 g L⁻¹, resulting in improved hydrolysis and acidification rates. The specific hydrolysis and acidification rates of stage I with a HRT of 10 days were 2.367 g COD per g VS per day and 1.120 g COD per g VS per day, respectively. TPAD-R obtained a higher net energy than MD, 3.3 kJ g⁻¹ VS degraded more than MD with a HRT of 10 days.

Nomenclature

WWTP	Wastewater treatment plant
WAS	Waste activated sludge
HRT	Hydraulic retention time
TS	Total solids
VS	Volatile solids
VFA	Volatile fatty acid
PCOD	PCOD particulate-COD
FID	Flame ionization detector
MD	Conventional mesophilic anaerobic digestion
TPAD-R	Temperature-phased anaerobic digestion with a recycle system

References

1. S. Ghosh, J. R. Conrad and D. L. Klass, Anaerobic acidogenesis of wastewater sludge, J.- Water Pollut. Control Fed., 1975, 47, 30–45.
2. S. Ghosh, Pilot-scale demonstration of two-phase anaerobic digestion of activated sludge, Water Sci. Technol., 1991, 23, 1179–1188.
3. J. B. van Lier, A. Tilche, B. K. Ahring, H. Macarie, R. Moletta, M. Dohanyos, L. W. H. Pol, P. Lens and W. Verstraete, New perspectives in anaerobic digestion, Water Sci. Technol., 2001, 43, 1–18.
4. S. I. Zack and G. P. Edwards, Gas production from sewage sludge, Sewage Works J., 1929, 1, 160–186.
5. N. Aoki and M. Kawase, Development of high-performance thermophilic two-phase digestion process, Water Sci. Technol., 1991, 23, 1147–1156.
6. J. M. Gossett, M. ASCE and R. L. Belser, Anaerobic digestion of waste activated sludge, J. Environ. Eng. Div., 1982, 108, 1101–1120.
7. D. Bolzonella, P. Pavan, P. Battistoni and F. Cecchi, Mesophilic anaerobic digestion of waste activated sludge: influence of the solid retention time in the wastewater treatment process, Process Biochem., 2005, 40, 1453–1460.
8. D. Bolzonella, L. Innocenti and F. Cecchi, Biological nutrient removal wastewater treatments and sewage sludge anaerobic mesophilic digestion performances, Water Sci. Technol., 2002, 46, 199–208.
9. J. A. Eastman and J. F. Ferguson, Solubilization of particulate organic carbon during the acid phase of anaerobic digestion, J.- Water Pollut. Control Fed., 1981, 53, 352–366.
10. J. Ariunbaatar, A. Panico, G. Esposito, F. Pirozzi and P. N. L. Lens, Pretreatment methods to enhance anaerobic digestion of organic solid waste, Appl. Energy, 2014, 123, 143–156.
11. J. A. Müller, Prospects and problems of sludge pre-treatment processes, Water Sci. Technol., 2001, 44, 121–128.
12. H. Carrère, C. Dumas, A. Battimelli, D. J. Batstone, J. P. Delgenès, J. P. Steyer and I. Ferrer, Pretreatment methods to improve sludge anaerobic degradability: a review, J. Hazard. Mater., 2010, 183, 1–15.
13. H. N. Gavala, U. Yenal, I. V. Skiadas, P. Westermann and B. K. Ahring, Mesophilic and thermophilic anaerobic digestion of primary and secondary sludge. Effect of pre-treatment at elevated temperature, Water Res., 2003, 37, 4561–4572.
14. H. Q. Ge, P. D. Jensen and D. J. Batstone, Temperature phased anaerobic digestion increases apparent hydrolysis rate for waste activated sludge, Water Res., 2011, 45, 1597–1606.
15. M. Romli, P. F. Greenfield and P. L. Lee, Effect of recycle on a two-phase high-rate anaerobic wastewater treatment system, Water Res., 1994, 28, 475–482.
16. Y. Y. Li, T. Noike, O. Mizuno and Y. Okuno, A new two-phase process for waterless methane fermentation treating the organic fraction of MSW, J. Environ. Eng. Manage., 2006, 16, 297–302.
17. C. F. Chu, Y. Y. Li, K. Q. Xu, Y. Ebie, Y. Inamori and H. N. Kong, A pH- and temperature-phased two-stage process for hydrogen and methane production from food waste, Int. J. Hydrogen Energy, 2008, 33, 4739–4746.
18. D. Y. Lee, Y. Ebie, K. Q. Xu, Y. Y. Li and Y. Inamori, Continuous H₂ and CH₄ production from high-solid food waste in the two-stage thermophilic fermentation process with the recirculation of digester sludge, Bioresour. Technol., 2010, 101, S42–S47.
19. APHA, Standard methods for the examination of water and wastewater, American Public Health Association, Washington, D. C., 22nd edn, 2012.
20. M. DuBois, K. A. Gilles, J. K. Hamilton, P. A. Rebers and F. Smith, Colorimetric method for determination of sugars and related substances, Anal. Chem., 1956, 28, 350–356.
21. O. H. Lowry, N. J. Rosebrough, A. L. Farr and R. J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem., 1951, 193, 265–275.
22. E. G. Bligh and W. J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol., 1959, 37, 911–917.
23. K. H. Schmit and T. G. Ellis, Comparison of temperature-phased and two-phase anaerobic co-digestion of primary sludge and municipal solid waste, Water Environ. Res., 2001, 73, 314–321.
24. A. H. Igoni, M. J. Ayotamuno, C. L. Eze, S. O. T. Ogaji and S. D. Probert, Designs of anaerobic digesters for producing biogas from municipal solid-waste, Appl. Energy, 2008, 85, 430–438.
25. P. J. Mills, Minimisation of energy input requirements of an anaerobic digester, Agric. Wastes, 1979, 1, 57–66.
26. J. Q. Lu and B. K. Ahring, Presented in part at Anaerobic Digestion Solid Waste 2005 Conference Proceedings, Copenhagen, 2005.
27. J. J. Lay, Y. Y. Li and T. Noike, The influence of pH and ammonia concentration on the methane production in high-solids digestion processes, Water Environ. Res., 1998, 70, 1075–1082.
28. P. L. McCarty, Anaerobic waste treatment fundamentals—Part III: toxic materials and their control, Public Works, 1964, 95, 91–94.
29. S. G. Pavlostathis and E. Giraldo-Gomez, Kinetics of anaerobic treatment, Water Sci. Technol., 1991, 24, 35–39.
30. P. Elefsiniotis and W. K. Oldham, Substrate degradation patterns in acid-phase anaerobic digestion of municipal primary sludge, Environ. Technol., 1994, 15, 741–751.
31. I. Metcalf Eddy, G. Tchobanoglous, F. L. Burton and H. D. Stensel, Wastewater Engineering: Treatment and Reuse, McGraw-Hill, New York, 4th edn, 2002.
32. L. Appels, J. Baeyens, J. Degrève and R. Dewil, Principles and potential of the anaerobic digestion of waste-activated sludge, Prog. Energy Combust. Sci., 2008, 34, 755–781.
33. D. Bolzonella, P. Pavan, M. Zanette and F. Cecchi, Two-phase anaerobic digestion of waste activated sludge: effect of an extreme thermophilic prefermentation, Ind. Eng. Chem. Res., 2007, 46, 6650–6655.
34. Y. Miron, G. Zeeman, J. B. van Lier and G. Lettinga, The role of sludge retention time in the hydrolysis and acidification of lipids, carbohydrates and proteins during digestion of primary sludge in CSTR systems, Water Res., 2000, 34, 1705–1713.

---