w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0

Avai lab le a t www.sc iencedi rec t .com

journa l homepage : www.e lsev ie r . com/ loca te /wat res

Pre-treatment mechanisms during thermophilic–mesophilictemperature phased anaerobic digestion of primary sludge

Huoqing Ge, Paul D. Jensen, Damien J. Batstone*

AWMC, Advanced Water Management Centre, Environmental Biotechnology CRC, The University of Queensland, St Lucia, Brisbane,

QLD 4072, Australia

a r t i c l e i n f o

Article history:

Received 26 June 2009

Received in revised form

2 September 2009

Accepted 2 September 2009

Published online 8 September 2009

Keywords:

Temperature phased

anaerobic digestion

Thermophilic pre-treatment

Mesophilic pre-treatment

Primary sludge

* Corresponding author: Tel.: þ61 7 3346 905E-mail address: damienb@awmc.uq.edu.a

0043-1354/$ – see front matter ª 2009 Elsevidoi:10.1016/j.watres.2009.09.005

a b s t r a c t

Pre-treatment is used extensively to improve degradability and hydrolysis rate of material

being fed into digesters. One emerging process is temperature phased anaerobic digestion

(TPAD), which applies a short (2 day) 50–70 �C pre-treatment step prior to 35 �C digestion in

the main stage (10–20 days). In this study, we evaluated a thermophilic–mesophilic TPAD

against a mesophilic–mesophilic TPAD treating primary sludge. Thermophilic–mesophilic

TPAD achieved 54% VS destruction compared to 44% in mesophilic–mesophilic TPAD, with

a 25% parallel increase in methane production. Measurements of soluble COD and NH4þ-N

showed increased hydrolysis extent during thermophilic pre-treatment. Model based

analysis indicated the improved performance was due to an increased hydrolysis coeffi-

cient rather than an increased inherent degradability, suggesting while TPAD is suitable as

an intensification process, a larger main digester could achieve similar impact.

ª 2009 Elsevier Ltd. All rights reserved.

1. Introduction (a) Improve biogas production to offset energy demand

Waste organic solids are widely produced by domestic and

industrial wastewater treatment plants. Anaerobic digestion

is a common stabilisation method for treating these solids,

which is environmentally beneficial due to production of

renewable energy. However, degradability of the feed material

needs to be relatively high, to allow good solids destruction,

provide gas for heating and mixing, and prevent washout of

methanogens. Degradability is particularly poor in long-

sludge age activated sludge systems (Gossett and Belser, 1982).

Many long-sludge age systems are also smaller scale (<5 dry

tonnes solids produced per day), where high-capital options to

enhance degradability, such as sonication or thermal hydro-

lysis are not available (Barr et al., 2008). To address these

limitations in smaller plants, an anaerobic option should

(Batstone et al., 2008a):

1; fax: þ61 7 3365 4726.u (D.J. Batstone).

er Ltd. All rights reserved

(b) Increase solids destruction to reduce the volume of sludge

requiring ultimate disposal

(c) Increase hydrolysis rates to allow reduced digester size

and capital cost and

(d) Achieve pathogen free stabilised solids to expand reuse

options.

Temperature phased anaerobic digestion (TPAD) may

allow enhanced degradability and biogas production, as well

as pathogen destruction, at a relatively low capital cost. TPAD

consists of a pre-treatment stage operated under thermophilic

temperature (50–70 �C) and short hydraulic retention times

(HRT), followed by a main stage operated at lower mesophilic

temperature with a longer retention time. Pathogen destruc-

tion and hydrolytic and acidogenic conditions can be further

optimised in the pre-treatment process. In the following main

.

Table 1 – Characteristics of the primary sludge used inthis study.

Measure Primary sludge

TS (g L�1) 26.9� 2.9a

VS (g L�1) 20.7� 2.0

pH 5–6.5

COD (g L�1) 30.2� 3.2

VFA (g COD L�1) 0.6� 0.2

TKN (g N L�1) 1.3� 0.6

NH4þ-N (g L�1) 0.09� 0.02

a Indicates standard deviation across 5 different feed materials

used in the study over 6 months.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0124

stage, a longer retention time and a neutral pH favour meth-

anogenesis for maximum conversion of organic components

to methane.

There have been a number of studies evaluating TPAD

systems. Han et al. (1997) tested the effect of different solids

retention times for TPAD system (55 �C and 35 �C) compared

with conventional single-stage mesophilic (35 �C) digestion on

primary sludge and waste activated sludge. They showed that

the optimal solids retention time of across both stages of

a TPAD system ranged from 11 to 17 days, with volatile solids

(VS) destruction up to double in TPAD system compared to

single-stage anaerobic digestion. Skiadas et al. (2005) found

a VS destruction with TPAD system (70 �C, 2 day HRT and

55 �C) of 55% and 43% for primary and secondary sludge

respectively, higher than 43% and 6% achieved in the single-

stage thermophilic (55 �C) anaerobic digestion. Watts et al.

(2006) reported that lower thermophilic temperatures (47 �C

and 54 �C, 2 day HRT) treating waste activated sludge did not

offer higher VS destruction over single-stage mesophilic

(37 �C) anaerobic digestion. When the thermophilic tempera-

ture was increased to 60 �C, VS destruction was improved to

35%, compared with 24% in single-stage mesophilic anaerobic

digestion. They also observed increased gas production

consistent with the increased VS destruction.

These studies indicate enhanced treatment performance

for TPAD systems as compared to single-stage thermophilic or

mesophilic systems. However, rigorous analysis is missing, as

there is no direct parallel comparison of mesophilic–meso-

philic and thermophilic–mesophilic TPAD. There is also little

analysis of which conditions (temperature and pH) can opti-

mise eventual hydrolytic conversion. Finally, it has not been

established whether enhanced performance is due to

increased hydrolysis in the pre-treatment stage, increased

overall degradability, or a conditioning process (such as

a physical breakdown of sludge similar to that achieved during

thermal hydrolysis and sonication), that allows better perfor-

mance in the main stage. This paper addresses these limita-

tions on a particular feed (primary sludge) by operating parallel

thermophilic–mesophilic and mesophilic–mesophilic TPAD

systems, and detailed analysis of the pre-treatment process.

2. Materials and methods

2.1. Substrate

The substrate used in this study was primary sludge collected

from a large wastewater treatment plant in Brisbane,

Australia. The feed was screened with a 3 mm sieve and

diluted with tap water to a total solids (TS) concentration of 2–

3%. Feed batches were prepared at intervals of 1–2 months and

stored at below 4 �C. Regular analysis was performed to

determine the characteristics and consistency of the feed

material. The average characteristics of the primary sludge

feed are shown in Table 1.

2.2. Laboratory scale reactor systems

Two identical TPAD systems, as shown in Fig. 1 were used

throughout the study. Each system contained a 0.6L reactor

(HRT 2 days) for pre-treatment and a 4.0L reactor (HRT 13–14

days approx) as main methanogenic stage. The thermophilic

pre-treatment (TP) system and mesophilic pre-treatment (MP)

system were operated identically, except for the pre-treat-

ment stage, which was either 50–65 �C (TP1), or 35 �C (MP1).

The temperature in the pre-treatment stages was maintained

with temperature controlled water jackets, while temperature

in the main methanogenic stages was maintained using

submersed electrical heating elements. All reactors were

continually mixed using magnetic stirrer bars. Gas production

volumes and pH were recorded from each reactor and recor-

ded online by a process logic control system.

2.3. Start-up and operation

Each reactor was inoculated from a full-scale anaerobic

digester (35� 1 �C) in Brisbane, Australia. Reactors were fed at

intervals of 4 hours (6 times daily). During feed events,

approximately 50 mL of feed was pumped through the system

simultaneously using multi-head peristaltic pumps located

between the feed reservoirs and pre-treatment stages; pre-

treatment stages and methanogenic stages; and methano-

genic stages and the waste effluent drums.

The systems were operated for over 6 months. During this

time the temperature of TP1 was altered to create different

operating periods:

� Period 1: 50 �C (117 days)

� Period 2: 60 �C (20 days)

� Period 3: 65 �C (32 days)

� Period 4: 65 �C, pH 4.5 by dosing of 1 M HCl (14 days).

The TP system had been operated for 64 days before the MP

system commenced operation. The temperature of MP1, TP2

and MP2 were held constant at 35 �C during all periods. After

Period 4 the acid dosing was stopped and the pH in TP1

returned to its natural level of 6.8. Only data from Day 75 was

used in comparative analysis (i.e., after stabilisation of both

digesters).

2.4. Analysis

Gas production was measured using tipping bucket gas

meters and continuously logged. Gas meters were regularly

recalibrated and switched between reactors to prevent

Pretreatment 0.6LTP1 = 50-65°C

MP1 = 35°C

Main Digester 4LTP2=35°CMP2=35°C

PLCF

Gas meterF

Gas meter

Feed reservior Effluent drum

Feedpump

Effluentpump

Digester pump

Heatingcoil

Water jacket temperature

control

Gas to exhaust

pH

pH

Fig. 1 – Schematic diagram of TPAD systems.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0 125

systematic errors. Biogas composition (H2, CH4, CO2) was

analysed by a Perkin Elmer loop injection gas chromatography

(GC), as described by Tait et al. (2009). The pH in each reactor

was measured daily with a calibrated glass body probe (TPS,

Brisbane, Australia).

Liquid samples were collected from each reactor three

times per week. Analyses were performed for TS, VS, volatile

fatty acid (VFA), chemical oxygen demand (COD), total Kjel-

dahl nitrogen (TKN) and ammonium–nitrogen (NH4þ-N).

Analytical methods were as for Standard Methods (APHA,

1998). COD was measured on Merck Method for total (TCOD)

and soluble fractions (SCOD), using an SQ 118 Photometer

(Merck, Germany). For measurement of SCOD, VFA and NH4þ-

N, the liquid samples were centrifuged at 2500g for 20 min and

filtered through a syringe filter (0.22 mm PES membrane) prior

to analysis. VFA concentrations were measured by GC (Agi-

lent, FID with polar capillary column). NH4þ-N and TKN were

measured using a Lachat Quik-Chem 8000 Flow Injection

Analyser (Lachat Instrument, Milwaukee).

2.5. Calculation of VS destruction

VS destruction was calculated using the Van Kleeck equation,

which assumes that the amount of fixed solids is conserved

during digestion (Switzenbaum et al., 2003). It can be

expressed as

VS destruction% ¼ VSfraci � VSfrac0

VSfraci ��VSfraci � VSfrac0

� (1)

Where VSfraci and VSfrac0 are volatile fractions (VS/TS) in the

influent and effluent solids.

VS destruction was also calculated based on the gas flow,

expressed as

VS destructiongas% ¼�CODCH4=fraci

�

VSi(2)

Where CODCH4is daily CH4 production as g COD d�1

fraci is COD/VS ratio of influent, measured as 1.47� 0.02

(95% confidence in mean over 197 measurements)

VSi is volatile solids loading rate as g VS d�1.

2.6. Mathematical analysis

2.6.1. Model implementationThe IWA Anaerobic Digestion Model No. 1 (ADM1) (Batstone

et al., 2002a) was used. The reference Aquasim 2.1d version

was used (Reichert, 1994) with inputs as described below.

Initial conditions were based on a steady state, adjusted for

measured initial conditions (organic solids, organic acids,

ammonia, TKN, etc).

2.6.2. Model inputsDefining inputs well is important to achieve reliable model

predictions. In this case, inputs were divided into particulate

inerts, carbohydrates, proteins, lipids, organic acids and

ammonia, based on a modified form of the COST ASM1-ADM1

interface (Nopens et al., 2009). The main difference is that the

inert fraction was mapped in terms of an overall degradability

parameter. Other fractions were based on VS, COD, TKN,

organic acids, and NH4þ-N measurements as in the standard

interface. There were 170 input changes over 180 days used in

the model.

2.6.3. Parameter estimation and analysisEstimation of parameter value and confidence in value are

critical to assess difference between two systems. The main

parameters compared were degradability extent ( fd) and

apparent first order hydrolysis rate coefficient (khyd) (Pavlos-

tathis and Giraldo-Gomez, 1991), based on the method of

Batstone et al., (2003, 2008b) used to estimate parameter

confidence regions for a two-parameter system. A 95% confi-

dence limit was used, with appropriate F-values for 2

parameters and the number of degrees of freedom (approx.

158, F¼ 2.996). A modified version of Aquasim 2.1d was used to

determine the parameter surfaces. Gas flow was used as

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0126

a measured variable, with sum of squared errors (c2) as an

objective function.

3. Results

3.1. Overall performance of TPAD systems

There are three methods of calculating VS destruction; Van

Kleeck VS destruction, which uses inlet mineral solids as

a reference, mass balance VS destruction, which uses inlet

organic solids as a reference, and gas flow VS destruction,

which uses inlet flow and COD as references. In this study,

Van Kleeck VS destruction was consistent with gas flow VS

destruction, but was higher than mass balance VS destruc-

tion. This may indicate incomplete mixing in the methano-

genic stage. Therefore, Van Kleeck and gas flow VS destruction

were used as main indicators of performance. Mineral solids,

NH4þ-N, and other analyses supported the use of Van Kleeck

and gas flow VS destruction measures.

The TP and MP systems both achieved VS destruction

greater than 38% (Fig. 2, top), the required value specified in

the 40 Code of Federal Regulations (CFR) Part 503 regulations

to minimize vector attraction (US EPA, 1994). Statistical anal-

ysis (student t-test, a¼ 0.05) confirmed that VS destruction in

)%(

noitcurtsedSV

0

20

40

60

80

100% VS destruction in TP system% VS destruction in MP system

Time in op0 20 40 60 80

yadL(

noitcudorpenahte

M1-)

0

1

2

3

Methane production in TP systemMethane production in MP system

Peri

Fig. 2 – VS destruction (top) and daily methane production (bott

(% VS destruction is based on the primary sludge feed characte

the TP system was significantly greater than that in the MP

system from Day 75 to Day 183. However, varying the ther-

mophilic pre-treatment temperature from 50 �C to 65 �C did

not have a significant impact on VS destruction. Additionally,

VS destruction was not improved under acidic pre-treatment

conditions. A summary of the average VS destruction during

each period is shown in Fig. 3.

Thermophilic pre-treatment enhanced VS destruction,

resulting in higher conversion of organic components to

methane. This was reflected in the approximately 25% higher

methane production from the TP system compared to the MP

system, as shown in Fig. 2 (bottom). The increase was

confirmed as a statistically significant improvement by the

student t-test analysis (a¼ 0.05). In both systems, the

methane production from pre-treatment stage was negligible

compared to that in the methanogenic stage. Methane

accounted for 72% and 69% of biogas composition in TP2 and

MP2 respectively with carbon dioxide being the other major

component during all operating periods.

The methane production increase was not observed when

the thermophilic pre-treatment temperature was increased to

60 �C and 65 �C. Methane production results were reflected in

apparent VS destruction. Methane production in both systems

was lowest during Period 4, it is not clear if this was due to

variations in the feed or operational conditions.

eration (days)100 120 140 160 180 200

od 1 Period 2 Period 3 Period 4

om) during each period in the TP system and MP system

ristics and Van Kleeck equation).

VS d

estru

ctio

n (%

)

0

20

40

60

80

100VS destruction in TP systemVS destruction in MP systemApparent VS destruction on methane flow in TP systemApparent VS destruction on methane flow in MP system

Period 1 Period 2 Period 3 Period 4

Fig. 3 – Average VS destruction and apparent VS

destruction on methane flow during each period in the TP

system and MP system (Error bars are 95% confidence in

mean VS destruction and methane production).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1Degradability

Hyd

roly

sis

rate

Degrades more

Deg

rade

s fa

ster

TP system

MP system

Fig. 5 – Confidence regions for khyd (dL1) and fd (degradable

fraction) using gas flow as an objective function in the TP

system (172 measurements) and MP system (108

measurements).

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0 127

Fig. 4 shows the biogas production for a 24 h period from

TP2 and MP2, indicating the increase in performance was

across the feed cycle. During each feed event, the methane

production declined with the substrate consumed in both

methanogenic stages, and TP2 demonstrated a faster

response to feed than MP2.

3.2. Model based analysis

Fig. 5 shows the 95% confidence regions for degradability ( fd)

(x-axis) and apparent hydrolysis rate (khyd) ( y-axis) in both

systems using complete gas flow over 180 days as objective

functions. In the TP system, khyd values were between 0.20–

0.51 d�1, with fd of 0.56–0.64. In the MP system, the confidence

region was right-unbounded in fd, indicating that a degrad-

ability upper limit could not be determined. Therefore, there

was statistical overlap between the two fd values, but hydro-

lysis was significantly faster in the TP system.

3.3. Pre-treatment mechanism

SCOD in TP1 was higher than that formed in MP1 for all

periods, increased with temperature increase, and dropped

Time in operation (h)

Biog

as p

rodu

ctio

n ra

te (L

day

-1)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

0 4 8 12 16 20 24

Biogas production in TP2Biogas production in MP2

Fig. 4 – Biogas production for a 24 h period from TP2 and

MP2.

under acidic conditions (pH 4.5). It is also important to note

that only 5.5% of the organic material was solubilised during

thermophilic pre-treatment, while the final release was

considerably more. Organic acids did not follow this trend, as

organic acid concentrations were lower in TP1 as compared to

MP1. This indicates that the material is being solubilised to

a greater extent at thermophilic conditions, but not subse-

quently converted to organic acids. The main organic acid

produced in TP1 was acetate, while propionate was the main

VFA produced in MP1, as shown in Table 2. Other VFAs were

also measured (iso-butyrate, butyrate, iso-valerate, valerate

and hexanoate), at significantly lower levels than acetate and

propionate.

Increasing the thermophilic pre-treatment temperature

from 50 �C to 60 �C resulted in an increase in acetate concen-

tration. However, acetate did not increase further with

temperature increase to 65 �C and dropped under acidic

conditions. Propionate concentration was lower in TP1 than

MP1. Propionate did not appear influenced by temperature,

and dropped significantly under acidic conditions. The total

VFA concentration dropped by approximately half with pH

decrease, indicating that low pH may be responsible for

inhibition of fermentation or hydrolysis. Although VFA

concentrations were high in both TP1 and MP1, in the meth-

anogenic stages (TP2 and MP2) the concentrations were very

low (<100 mg COD L�1).

NH4þ-N is another key intermediate released from

fermentation of protein or other nitrogenous organic

compounds. Generally, but especially during periods 1–3,

NH4þ-N concentration was higher in TP1 than in MP1 (Fig. 6),

indicating enhanced protein fermentation under thermo-

philic conditions. Again, this was not influenced by tempera-

ture. NH4þ-N release decreased significantly under acidic

conditions, which was consistent with SCOD and VFA

concentration, indicating the low pH has a negative impact on

fermentative activity. The final concentration of NH4þ-N in TP2

and MP2 was similar at each period, indicating the thermo-

philic pre-treatment does not substantially influence protein

degradation extent.

Table 2 – Summary of solubilisation performance in TP1 and MP1 according to VFAs.

Acetate (mg COD L�1) Propionate (mg COD L�1) Total VFA (mg COD L�1) SCOD (mg COD L�1)

Period 1 TP1 990 (230)a 560 (90) 2470 (400) 3030 (500)

MP1 730 (210) 1130 (380) 2500 (540) 2630 (370)

Period 2 TP1 1250 (90) 700 (70) 2870 (170) 3520 (170)

MP1 900 (80) 1680 (220) 3300 (360) 3120 (210)

Period 3 TP1 1240 (70) 650 (50) 2800 (140) 4220 (560)

MP1 540 (150) 1630 (140) 2970 (220) 3270 (360)

Period 4 TP1 670 (100) 200 (70) 1300 (230) 3510 (230)

MP1 400 (70) 1380 (50) 2600 (50) 3380 (140)

a Indicates standard deviation across different measurements over each period.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0128

4. Discussion

4.1. Overall performance of TPAD systems

Primary sludge is a large and unstabilised stream common in

wastewater treatment plants. An increase in VS destruction in

real plants translates to better sludge dewaterability (Novak,

2006) and lower overall costs of disposal, as this is generally

weight based. A corresponding higher methane production in

the TP system can be used to produce heat and power for the

whole treatment process and offset the higher energy demand

required by thermophilic temperature. Since most of the heat

used is excess heat from cogeneration, the heating demand

required at 55 �C can be provided from methane production. It

is important that VS destruction and methane production were

not improved with increased thermophilic pre-treatment

N c

once

ntra

tion

(mg

L-1)

0

500

1000

1500

2000

Period 1 Period

TKN in MP1 TKN

NH4+NH4

+-N in MP1

TKN in TP1 TKNNH4

+NH4+-N in TP1

0

500

1000

1500

2000

Fig. 6 – Concentrations of TKN and NH4D-N during each period i

confidence in mean TKN and NH4D-N).

temperature, suggesting that the temperature may be selected

to optimise pathogen destruction rather than VS destruction.

4.2. Model analysis of TPAD systems

The estimates of apparent khyd based on the gas flow showed

a greater hydrolysis rate during thermophilic pre-treatment,

which was 67% higher than during mesophilic pre-treatment.

However, the degradability extent in the TP system was not

increased compared to the MP system, indicating that ther-

mophilic pre-treatment influences speed of degradation

rather than extent of degradation. This is similar to other

lower impact pre-treatment methods such as sonication,

which alter physical properties of the substrate to enhance

hydrolysis rates (Tiehm et al., 2001). In contrast, high-inten-

sity pre-treatment methods such as thermal hydrolysis

increase both rate and extent (Neyens and Baeyens, 2003).

2 Period 3 Period 4

in MP2

-N in MP2

in TP2-N in TP2

n the TP system (top) and MP system (Error bars are 95%

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0 129

There are a wide range of hydrolysis constants reported in

the literature for primary sludges (e.g., 0.2–0.6 d�1 in the review

of Batstone et al., 2002b). However, the best comparison is

probably to that of Siegrist et al. (2002) who reported a hydro-

lysis rate of 0.25 d�1 at mesophilic conditions compared to

0.4 d�1 at thermophilic conditions. This was for a thermophilic

methanogenic reactor. Our results indicate comparative

performance can be obtained simply by conditioning, rather

than operating the main digester at thermophilic conditions.

Since hydrolysis rate rather than extent is increased, the

increase in performance can be accomplished by either add-

ing a thermophilic pre-treatment stage or increasing the main

digester size. If considering the footprint and capital invest-

ment of anaerobic digestion process, the addition of a ther-

mophilic pre-treatment stage will benefit the design due to the

smaller process vessels compared to mesophilic pre-treat-

ment process or conventional mesophilic digestion. In terms

of solids destruction, a larger main digester could achieve the

same performance as adding a thermophilic pre-treatment

stage. However, the thermophilic pre-treatment process

enables pathogen destruction to achieve the ultimate solids

hygienisation required for land application and agricultural

use (Sung and Santha, 2003).

4.3. Pre-treatment mechanisms

Analysis of the pre-treatment reactors as assessed by NH4þ-N

and SCOD confirmed hydrolysis (solubilisation) in TP1 was

improved compared to MP1, however this did not translate to

increased conversion to organic acids. There was a consider-

able component of SCOD which could not be attributed to

organic acids.

Digestion intermediates such as glucose, pyruvate, succi-

nate, lactate, and ethanol (Elefsiniotis and Oldham, 1994) were

not detected in TP1. However, anaerobic organisms are able to

directly take up and utilise partially hydrolysed organics

including oligosaccharides and long-chain fatty acids (Lynd

et al., 2002). Significantly, all hydrolysates produced during the

TPAD processes were biologically degradable, exhibited by the

significant reduction in SCOD concentrations in the effluent of

the methanogenic stages (<500 mg L�1) compared to the pre-

treatment effluents and raw sludge feed (1000–2000 mg L�1).

Variations in the biological processes occurring in TP1 and

MP1 (5.5% solubilisation compared to 5.1% solubilisation) were

minimal compared to the increased methane production and

VS destruction (25% and 20%) observed between thermo-

philic–mesophilic TPAD and mesophilic–mesophilic TPAD. It

is clear that key mechanisms active during thermophilic pre-

treatment affected biological availability of the substrate

during downstream processes. However, the specific nature of

these mechanisms is not clear.

From a biological perspective, possible mechanisms

include stimulated growth of the microbial population or

production of extracellular hydrolytic enzymes which are

then passed downstream into the methanogenic reactors.

Increased microbial concentrations or enzyme activities could

explain the increases in apparent hydrolysis rates, without an

increase in sludge degradability (as determined from model

simulations).

From a non-biological perspective, increased disintegra-

tion of the sludge may have reduced particle size and

increased the surface area available to the microbial

community. Hydrolysis is a surface process and rates may be

improved by increasing the surface area of feed particles

(McAllister et al., 1994; Lynd et al., 2002). Furthermore, effluent

from TP1 may have contained increased colloidal substrates

that are readily degradable, but not measured as SCOD.

Further investigations into these mechanisms are required.

As a final note, decreased pH did not enhance hydrolysis.

Control of pH in full scale sludge fed systems is inherently

difficult to manipulate due to buffering from NH4þ-N release.

5. Conclusion

The following conclusions can be drawn from this study:

� Thermophilic–mesophilic TPAD achieved 20% and 25%

higher VS destruction and methane production respec-

tively, compared to mesophilic–mesophilic TPAD.

Increasing thermophilic pre-treatment temperature from

50–65 �C had no further impacts.

� Higher SCOD was produced during thermophilic pre-treat-

ment over mesophilic pre-treatment, and further increased

by increasing the thermophilic pre-treatment temperature

from 50 �C to 65 �C. Higher NH4þ-N was released during

thermophilic pre-treatment, but did not increase at

increased temperatures. Both SCOD and NH4þ-N decreased

under acidic pre-treatment conditions (pH 4.5).

� Model based analysis indicated that the improved perfor-

mance was due to an increased hydrolysis rate (0.1� 0.05 d�1

to 0.3� 0.15 d�1), rather than overall degradability.

Acknowledgements

This work was funded by the Queensland State Government,

under the Smart State Research-Industry Partnerships Program

(RIPP), Meat and Livestock Australia, and Environmental

Biotechnology Cooperative Research Centre (EBCRC), Australia

as P23 ‘‘Small-medium scale organic solids stabilization’’.

Huoqing Ge and Paul Jensen are recipients of an EBCRC post-

graduate scholarship and postdoctoral award, respectively. We

thank Beatrice Keller, and the AWMC Analytical Services Labo-

ratory for conducting organic acid and nitrogen analysis.

r e f e r e n c e

APHA, 1998. Standard Methods for the Examination of Water andWastewater, twentieth ed. American Public HealthAssociation, Washington, DC, USA.

Barr, K.G., Solley, D.O., Starrenburg, D.J., Lewis, R.G., 2008.Evaluation, selection and initial performance of a large scalecentralised biosolids facility at Oxley Creek WaterReclamation Plant, Brisbane. Water Science and Technology57 (10), 1579–1586.

w a t e r r e s e a r c h 4 4 ( 2 0 1 0 ) 1 2 3 – 1 3 0130

Batstone, D.J., Keller, J., Angelidaki, I., Kalyuzhnyi, S.,Pavlostathis, S.G., Rozzi, A., Sanders, W.T.M., Siegrist, H.,Vavilin, V., 2002a. Anaerobic Digestion Model No. 1 (ADM1),IWA Task Group for Mathematical Modelling of AnaerobicDigestion Processes. IWA Publishing, London, UK.

Batstone, D.J., Keller, J., Angelidaki, I., Kalyuzhny, S.,Pavlostathis, S.G., Rozzi, A., Sanders, W.T.M., Siegrist, H.,Vavilin, V., 2002b. Anaerobic Digestion Model No. 1 (ADM1)Scientific and Technical Report No. 13, IWA Task Group forMathematical Modelling of Anaerobic Digestion Processes.IWA Publishing, London, UK.

Batstone, D.J., Pind, P.F., Angelidaki, I., 2003. Kinetics ofthermophilic, anaerobic oxidation of straight and branchedchain butyrate and valerate. Biotechnology andBioengineering 84 (2), 195–204.

Batstone, D.J., Darvodelsky, P., Keller, J., (2008a). Trends inbiosolids handling technologies: economics andenvironmental factors. In: AWA biosolids specialty conferenceIV, Adelaide, Australia.

Batstone, D.J., Tait, S., Starrenburg, D., 2008b. Estimation ofhydrolysis parameters in full-scale anaerobic digesters.Biotechnology and Bioengineering 102 (5), 1513–1520.

Elefsiniotis, P., Oldham, K.W., 1994. Influence of pH on the acid-phase anaerobic digestion of primary sludge. Journal ofChemical Technology and Biotechnology 60 (1), 89–96.

Gossett, J.M., Belser, R.L., 1982. Anaerobic-digestion of wasteractivated sludge. Journal of the Environmental EngineeringDivision-Asce 108 (6), 1101–1120.

Han, Y., Sung, S., Dague, R.R., 1997. Temperature-phasedanaerobic digestion of wastewater sludges. Water Science andTechnology 36 (6-7), 367–374.

Lynd, L.R., Weimer, P.J., van Zyl, W.H., Pretorius, I.S., 2002.Microbial cellulose utilization: fundamentals andbiotechnology. Microbiology and Molecular Biology Reviews 66(3), 506–577.

McAllister, T.A., Bae, H.D., Jones, G.A., Cheng, K.J., 1994. Microbialattachment and feed digestion in the rumen. Journal ofAnimal Science 72 (11), 3004–3018.

Neyens, E., Baeyens, J., 2003. A review of thermal sludge pre-treatment processes to improve dewaterability. Journal ofHazardous Materials 98 (1–3), 51–67.

Nopens, I., Batstone, D.J., Copp, J.B., Jeppsson, U., Volcke, E.,Alex, J., Vanrolleghem, P.A., 2009. An ASM/ADM modelinterface for dynamic plant-wide simulation. Water Research43 (7), 1913–1923.

Novak, J.T., 2006. Dewatering of sewage sludge. DryingTechnology 24 (10), 1257–1262.

Pavlostathis, S.G., Giraldo-Gomez, E., 1991. Kinetics ofanaerobic treatment. Water Science and Technology 24 (8),35–59.

Reichert, P., 1994. Aquasim – a tool for simulation and data-analysis of aquatic systems. Water Science and Technology 30(2), 21–30.

Siegrist, H., Vogt, D., Garcia-Heras, J.L., Gujer, W., 2002.Mathematical model for meso- and thermophilic anaerobicsewage sludge digestion. Environmental Science &Technology 36 (5), 1113–1123.

Skiadas, I.V., Gavala, H.N., Lu, J., Ahring, B.K., 2005. Thermal pre-treatment of primary and secondary sludge at 70 �C prior toanaerobic digestion. Water Science and Technology 52 (1–2),161–166.

Sung, S., Santha, H., 2003. Performance of temperature-phasedanaerobic digestion (TPAD) system treating dairy cattlewastes. Water Research 37 (7), 1628–1636.

Switzenbaum, M.S., Farrell, J.B., Pincince, A.B., 2003.Relationship between the Van Kleeck and mass-balancecalculation of volatile solids loss. Water EnvironmentResearch 75 (6), 377–380.

Tait, S., Tamis, J., Edgerton, B., Batstone, D.J., 2009. Anaerobicdigestion of spent bedding from deep litter piggery housing.Bioresource Technology 100 (7), 2210–2218.

Tiehm, A., Nickel, K., Zellhorn, M., Neis, U., 2001. Ultrasonic wasteactivated sludge disintegration for improving anaerobicstabilization. Water Research 35 (8), 2003–2009.

US Environmental Protection Agency, 1994. A Plain English Guideto the EPA Part 503 Biosolids Rule. In: EnvironmentalProtection Agency 8322/R-93-003, September Washington, DC,USA.

Watts, S., Hamilton, G., Keller, J., 2006. Two-stage thermophilic–mesophilic anaerobic digestion of waste activated sludge froma biological nutrient removal plant. Water Science andTechnology 53 (8), 149–157.