Bioresource Technology 101 (2010) 26–33

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Anaerobic digestion of tannery waste: Semi-continuous and anaerobic

sequencing batch reactor processes
Gregor D. Zupančič *, A. Jemec
National Institute of Chemistry, Laboratory for Environmental Sciences and Engineering, Hajdrihova 19, P.O. Box 660, SI-1001 Ljubljana, Slovenia

a r t i c l e i n f o a b s t r a c t

Article history: Disposal of the vast amounts of tannery waste that are currently generated is a significant problem.
Received 14 May 2009 Anaerobic treatment of different types of tannery waste (fleshings, skin trimmings and wastewater
Received in revised form 10 July 2009 sludge) was investigated. The biochemical methane potential is the same at 37 °C or 55 °C and an assay
Accepted 14 July 2009
of this was shown to be an appropriate screening tool with which to estimate the susceptibility of a sub-
Available online 21 August 2009
strate to anaerobic digestion. The start-up procedure of a tannery waste thermophilic anaerobic digestion
in 100 days using seed from mesophilic digester processing municipal sludge is presented. The specific
Keywords:
methane production potential at 55 °C is estimated to be 0.617 m3 kg 1 of volatile suspended solids for
Anaerobic digestion
Biochemical methane potential
tannery waste sludge, 0.377 m3 kg 1 for tannery waste trimmings and 0.649 m3 kg 1 for tannery waste
Biogas production fleshings. Additional concerns such as chromium content, salinity and temperature fluctuations were also
Tannery waste addressed. Chromium content and salinity showed no adverse effects; however a reactor temperature
reduction of 4.4 °C led to a drop in biogas production of 25%, indicating a requirement to keep the tem-
perature constant at 55 °C.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction landfills and incineration. However, these two technologies are dis-
couraged as they fail to resolve the solid waste disposal problem in
The European Union is among the leading leather producers. A an ecologically acceptable manner (Dhayalan et al., 2007). Conse-
survey conducted in 2001 revealed that in 1998 there were around quently, new alternative re-uses of tannery waste, such as biodie-
3000 tanneries in Europe, with Italy, the major leather producer, sel production (Özgünay et al., 2007) and remanufacture into bone
producing 190 million m2 of leather per year (EC, 2001). It has been flour (Montoneri et al., 1994) are being developed. In light of recent
estimated, that between 400,000 and 900,000 tonnes of sludge (per developments in the renewable energy market and the increasing
fresh weight basis), the majority of which is deposited in landfills, cost of waste treatment the on-site anaerobic treatment of tannery
is generated annually in the EU from leather processing (EC, 2001). waste to produce biogas has become an attractive option for the
Dhayalan et al. (2007) reported that in addition, some 170,000 ton- tannery industry.
nes of tanned leather waste are generated annually. Disposal of Although considerable work has been done in treatment of tan-
waste generated in leather production is therefore a serious prob- nery wastewater (Song et al., 2003, 2004; Lefebvre et al., 2006),
lem and the importance of technological measures to combat envi- anaerobic treatment of tannery waste has received less attention
ronmental challenges from leather processing activities is now in recent years. There are some older reports on anaerobic diges-
increasingly recognized (Thanikaivelan et al., 2005). tion of tannery waste (Cenni et al., 1982; Tunick et al., 1985; Lalith-
The high chromium content in tannery waste prevents its use as a et al., 1994; Urbaniak, 2006), and reports on hydrolysis of tannery
a fertilizer and conventional handling of tannery waste involves waste (Raju et al., 1997) but few papers deal with the technology of
anaerobic digestion of tannery waste.
Tannery waste consists of wastewater, and solid waste fleshings
Abbreviations: ASBR, anaerobic Sequencing Batch Reactor; BMP, biochemical
and waste skin trimmings, the former two being composed mostly
methane potential; HRT, hydraulic retention time (days); MPR, methane production
rate (m3 of methane per m3 of digester per day); OLR, organic loading rate (kg of VSS of lipids and proteins. Gaseous efficiency from fats is estimated to
(TCOD) per m3 of digester per day); SBP, specific biogas productivity (m3 of biogas be higher than those of carbohydrates and proteins, therefore lipid-
per kg VSS inserted); SMP, specific methane productivity (m3 of methane per kg VSS rich waste can be regarded as a large potential renewable energy
inserted); TCOD, total chemical oxygen demand (g L 1 or mg g 1, degradation in %); source (Cirne et al., 2007). For example, 1250 L (68% CH4, 31%
TSS, total suspended solids (mass %, removal in %); VSS, volatile suspended solids
(mass %, removal in %).
CO2, 1% other) of biogas was estimated to be produced from 1 kg
* Corresponding author. Tel.: +386 1 4760 249; fax: +386 1 4760 300. (dry solids) of fat, while 790 L and 704 L of biogas was produced
E-mail address: gregor.zupancic@ki.si (G.D. Zupančič). from the same amount of carbohydrates and proteins, respectively

0960-8524/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2009.07.028
 G.D. Zupančič, A. Jemec / Bioresource Technology 101 (2010) 26–33 27

(Urbaniak, 2006). Indeed high gaseous efficiency was previously The processed particle size was 1–4 mm. Waste sludge was used
found for tannery waste, varying between 950–1120 L from 1 kg directly with no preprocessing.
organic dry solids, which was predetermined by high content of All the methods were kept under continuous statistical control.
fat (Urbaniak, 2006). Control charts were created from the results obtained in the anal-
Anaerobic biomethane formation is a complex process, in which ysis of RMs (laboratory working reference standards). In addition,
organic compounds are mineralised to biogas. It consists of several our laboratory also participated in proficiency tests (AQUACHECK,
phases, such as hydrolysis, acidogenesis, acetogenesis and metha- WRc plc, UK) and good performance in all determinations was ob-
nation, carried out by different groups of microorganisms, which tained. The measurement uncertainty of the measured TCOD, TSS
partly stand in syntrophic interrelation and depend on different and VSS was 8% and uncertainty of biogas measurements was 3%
requirements in the environment (Deublein and Steinhauser, (95% confidence limit). Measurement uncertainty was evaluated
2008). For instance, lipids are first hydrolyzed by acidogenic bacte- according to the principles of Guide to the Expression of Uncer-
ria to glycerol and free long-chain fatty acids; furthermore glycerol tainty in Measurement (BIPM-IEC-IFCC-ISO-IUPAC-IUPAP-OIML,
is converted to acetate, while fatty acids convert to acetate, propi- 1995).
onate and hydrogen. Finally, methanogenic bacteria which utilize
methanol, acetate or hydrogen and carbon dioxide, produce meth- 2.2. Experimental setup, parameters and conditions
ane (Cirne et al., 2007; Deublein and Steinhauser, 2008).
The aim of this study was to investigate the potential for anaer- Experiments were conducted in three parts. In the first, bio-
obic digestion of different types of tannery waste, including flesh- chemical methane potential assays were conducted with the goal
ings, skin trimmings and tannery wastewater sludge. We focused of determining the biogas potential of each substrate at two differ-
on two techniques for anaerobic digestion, conventional semi-con- ent temperatures (37 °C and 55 °C). In the second part, a laboratory
tinuous and more recent, ASBR (Anaerobic Sequencing Batch Reac- scale pilot reactor was used, a start-up procedure was established,
tor) process. The ASBR operates in a cyclic batch mode with four and conventional semi-continuous experiments of anaerobic
distinct phases per cycle. The four phases are: filling, reacting, set- digestion were continued. In the third part, the ASBR process was
tling and release. The advantage of ASBR process according to re- utilized and compared to the conventional process.
ports in the literature (Lee et al., 2001; Wang et al., 2002) is
better TCOD removal as well as higher biogas production com- 2.2.1. Biochemical methane potential (BMP assay)
pared to conventional digestion systems. ASBR systems are also The potential of each of the individual substrate, fleshings, skin
popular largely due to possible elimination of equalization tanks trimmings and wastewater sludge, to be anaerobically degraded to
and secondary clarifiers as well as relatively simple operations methane was evaluated according to ISO 11734:1994 and OECD
(Zhang and Dugba, 2000; Ioannis and Bagley, 2002; Zupančič 311:2006. Two experiments with different temperatures (37 °C
et al., 2007). The suitability of these two techniques will be evalu- and 55 °C) were performed. Digested sludge to be used as an inoc-
ated and the beneficial impact of anaerobic treatment of tannery ulum was collected from an anaerobic digester at a local municipal
waste will be discussed. wastewater treatment plant. At least 5 days prior to the experi-
ment, it was predigested without the addition of any nutrients at
37 °C and 55 °C, depending on the temperature used in subsequent
BMP assay. The properties of this digested sludge were as follows:
2. Methods TSS was 9.42 g L 1 and TCOD was 10.791 mg L 1. The methane po-
tential was followed using an OxiTopÒ OC110 system (WTW-Wis-
2.1. Characterisation of tannery waste senschaftlich-Technische Werkstätten GmbH, Germany). The total
volume of the glass vessel was 1160 ml. The test mixture consisted
Leather processing is characterized by large amounts of solid of 100 ml of predigested sludge, 20 ml of sodium (Na2HPO4) and
and liquid waste. Pretanning processes are those in which most potasium (KH2PO4) buffer pH = 7.0 (final concentration 20 mM),
of the solid waste is produced. Leather production produces vast 380 ml of de-oxygenated water and tannery waste substrates.
amounts of wastewater, which is usually treated on-site and also The pH of the mixture was kept between 7.0 and 7.5 and did not
produces excess sludge, which is the main component in the mass change during the experiment. Glucose was used as a reference
balance of tannery waste in our case amounting to 50% of the en- substance. The experiments were run until the stabile methane/
tire organic load. The other two viable substrates for biogas pro- biogas pressure was reached in the vessels, which on average
duction produced in tannery industry are waste fleshings and was up to 30 days.
waste skin trimmings, which are solid waste, composed mostly The organic load (mg COD of tannery substrate per g TSS of di-
of lipids and fats, with water content of approximately 50% or less. gested sludge inoculum) varied with the temperature and sub-
They originate from the process of animal skin preparation for the strate. At 37 °C the organic loadings were: 130, 220, 155 and 150
tanning process. In this process all adherent flesh and fatty tissue is and at 55 °C: 53, 90, 83 and 50 for skin trimmings, fleshings, and
removed from animal skin and rough edges and impurities of skin tannery wastewater sludge and reference substance (glucose),
are also removed, becoming waste. Waste fleshings in our case add respectively.
up to 40% of the organic load, waste skin trimmings contributing Biogas sample was taken from the glass vessel at the end of
up to 10% of the organic load in the mass balance of tannery waste. each experiment and the percentages of CH4, CO2 and H2S of the
Waste leather can be successfully recycled into production of biogas produced were determined using GC–MS, (Agilent Technol-
leather goods and thus is rarely used in anaerobic digestion. ogies, 6890N, USA) to calculate the corresponding methane
Different types of tannery waste were collected at a local tan- production.
nery, where the ratio of produced wastes were 50% waste sludge,
40% waste fleshings and 10% skin trimmings. The waste was char- 2.2.2. Start-up procedure and semi-continuous process
acterised by analysis, with Standard Methods (APHA-AWA-WEF, When conducting anaerobic digestion of tannery waste, start-
2005), of the total suspended solids (TSS), volatile suspended solids up and biomass adaptation procedure are vital. We began our
(VSS) and total chemical oxygen demand (TCOD). Substrates were experiments by adopting the biomass from a local mesophilic di-
collected on weekly basis and stored at 4 °C. Waste fleshings and gester processing municipal sludge to conditions intended for
trimmings were preprocessed using an industrial meat cutter. digesting the tannery waste digestion. When starting we instantly
28 G.D. Zupančič, A. Jemec / Bioresource Technology 101 (2010) 26–33

raised the temperature to the thermophilic level (55 ± 1.0 °C) and (50% waste sludge, 40% waste fleshings and 10% skin trimmings,
started to feed the digester first with waste sludge from tannery percentage relative to VSS). Biogas collected from the reactor was
wastewater treatment plant with a low OLR of approximately lead through biogas absorption unit, where CO2 and H2S were ab-
0.5 kgVSS m 3 d 1 (corresponding to 0.70 ± 0.10 kgTCOD m 3 d 1). sorbed and subsequently measured by an Agilent ADM 2000 Flow-
After 45 days we added sterilized and hydrolyzed tannery waste meter. The detailed experimental setup is shown in Fig. 1.
(at 0.3 MPa and 130 °C) in sludge/waste ratio which would later For process evaluation, parameters such as organic matter deg-
be applied in the digestion process (50% waste sludge, 40% waste radation kinetics (total suspended solids – TSS, volatile suspended
fleshings and 10% skin trimmings, percentages are relative to solids – VSS and total chemical oxygen demand – TCOD) were ana-
VSS). We used this thermal pretreatment to make waste more eas- lyzed according to Standard Methods (APHA-AWA-WEF, 2005),
ily degradable in this start-up and adaptation phase. specific methane productivity (SMP), methane production rate
After 100 days of start-up we began semi-continuous (MPR) and biogas composition was determined by GC–MS (Agilent
experiments. We performed two experiments: in the first, the Technologies, 6890N, USA).
average OLR was 3.02 ± 0.02 kgVSS m 3 d 1 (corresponding to
6.27 ± 0.03 kgTCOD m 3 d 1) and HRT (hydraulic retention time) 2.2.3. ASBR process
was from 12 to 17 days, varying according to sludge concentration. In the third of the experiments we modified an ASBR process to
In the second experiment the average OLR was 4.37 ± match the tannery waste production. The ASBR process is batch in
0.02 kgVSS m 3 d 1 (corresponding to 9.11 ± 0.03 kgTCOD m 3 d 1) nature with cycle duration of 7 days; the filling phase continued
and HRT was from 8 to 12 days. for 5 days with the reaction phase already in progress. This lasted
All of the processing, including the start-up, was semi-continu- 2 extra days and on day 7 the reactor settle and the release phases
ous in nature, feeding and removing a batch of the substrate once took place. These were relatively short compared to other phases.
daily. The process was conducted in the thermophilic temperature The exchange volume was from 22% to 39% of total reactor volume.
range due to high lipid content in the substrate. Fleshings and skin We chose to operate the process in a batch manner for two distinct
trimmings liquefy in excess of 50 °C and are therefore much easier reasons. First, the degradation of lipids proceeds more slowly than
to suspend in the reactor solution and are easier available for the sludge itself and in a batch operation the sludge has more time to
bacteria to produce biogas. No thermal pretreatment is necessary degrade completely. Secondly, we hooped to adopt the process of
when using the thermophilic process. A portion of the reactor out- anaerobic digestion to the production cycle in a week without a
let is taken and solid waste is quickly melted and washed into the need of inserting a buffer tank for continuous feed.
reactor. The only pretreatment necessary, which is essential to suc- We performed five experiments with different OLR’s ranging
cessful digestion, is cutting and mincing of the solid tannery waste from 6.20 kgTCOD m 3 d 1 to the maximum of 11.79 kgTCOD m 3 d 1
to 1–4 mm size in an industrial meat cutter. The substrate feed ra- (corresponding to an OLR of 2.86–4.91 kgVSS m 3 d 1). The lowest
tio of waste sludge–fleshings–skin trimmings was always constant OLR, in experiment 1, was slightly lower in comparison to that in
and matching the quantities in a conventional tannery production the semi-continuous experiment 1. The substrate feed ratio of

Fig. 1. Experimental setup.

 G.D. Zupančič, A. Jemec / Bioresource Technology 101 (2010) 26–33 29

waste sludge–fleshings–skin trimmings was always constant and were based on the two values obtained for the same substrate at
matched the quantities in a conventional tannery production 37 °C and 55 °C and are between 8–14%.
(50% waste sludge, 40% waste fleshings and 10% skin trimmings, Although degradation efficiency is similar in both temperature
percentages based on the VSS). regimes, we have selected the higher temperature for subsequent
experiments. There were indications that the fleshings and trim-
mings might be more easily and faster degradable at this temper-
3. Results and discussion
ature and especially more easily handled. At temperatures in
excess of 50 °C waste trimmings and fleshings are liquefied and
3.1. Characteristics of tannery waste
are more readily degraded. At temperatures below 50 °C we have
observed that waste trimmings and fleshings form soap like gran-
The basic characteristics of tannery waste treated in this study
are shown in Table 1. The composition of waste fleshings and skin ules 1–10 mm in size. These contain the substrate and conse-
quently are more resistant to degradation in mixed reactor
trimmings was constant during our collection, while waste sludge
varied substantially. Waste fleshings were predominantly lipid vessels. Other authors (Yu et al., 2002) have reported higher degra-
dation rates of lipid substrates at thermophilic temperatures as
substrate containing 90–92% of fat and up to 7% protein in dry mat-
ter. Total nitrogen content was 1.0–1.3% of dry matter. Skin trim- well and therefore we decided to perform our experiments in ther-
mings were predominantly protein substrate with 70–80% of mophilic temperature regime. However mesophilic digestion of
protein and up to 20% fat in dry matter. Correspondingly total tannery waste remains to be investigated.
nitrogen content was in range of 10–12% of dry matter. Waste If the sample composition added to the reactor was 50% waste
sludge, 40% waste fleshings and 10% skin trimmings (on VSS basis),
sludge consists of excess biomass from the wastewater treatment
plant in addition to sludge settled in the primary sedimentation. and we assume that the contribution of SMP (specific methane
productivity) from each of the substrates is in the same ratio, then
In dry matter waste sludge contained 5–10% of fat, 9–13% of pro-
tein, and approximately 60–70% of anorganic material, predomi- the calculated SMP of the composite sample based on BMP assay at
55 °C would be estimated to be 0.558 m3 kg 1. This is in accor-
nantly lime and 1% salt. Total nitrogen content was 1.4–2.0% of
dance with the realistic values of SMP obtained in the reactors,
dry matter. While the excess biomass quantity in waste sludge is
since the average SMP value in the semi-continuous process was
constant, primary sludge can vary throughout the season. There-
0.443–0.521 m3 kg 1 and in the ASBR process was 0.329–
fore waste sludge COD/VSS ratio can be from 1.1 to 1.9, which
0.596 m3 kg 1.
has consequences in varying TCOD and VSS organic loading rates
We concluded that the BMP assay is a valid method to provide
at a constant waste quantity.
the first estimate as to whether the substrate has a potential to be
anaerobically degraded and if further experiments with reactors
3.2. Biochemical methane potential (BMP) assays
are justified.

The methane and biogas production determined in BMP assays

are presented in Table 2. Due to different consistency of the three 3.3. Process start-up and semi-continuous operation
substrates we expected significant differences in their anaerobic
biodegradability at different temperatures. However, waste sludge Throughout the experiment, the OLR varied due to different
and trimmings resulted in slightly better gas production when concentrations of waste sludge gathered in the tannery wastewa-
treated at 37 °C than at 55 °C while waste fleshings and a reference ter treatment plant over the course of the experiment. By adding
substance produced more biogas at a higher temperature. All the tannery waste, OLR was raised from the initial low OLR of
wastes are degradable up to a similar extent at both temperatures; approximately 0.5 kgVSS m 3 d 1 to 1.30 ± 0.20 kgVSS m 3 d 1 (lat-
these differences are within experimental error and are presumed ter corresponding to 3.00 ± 0.30 kgTCOD m 3 d 1). We experienced
to be due to different inoculum sludges employed at different tem- reactor overload and, 12 days after addition of tannery waste, a se-
peratures. According to the OECD standard (OECD 311:2006) quite vere pH drop from the usual 7.40 to 6.98. We resolved this simply
high coefficients of variation (the ratio between the standard devi- by stopping the influent for few days until the pH rose back to de-
ation and the mean) of up to 45% were found when the same sub- sired value (7.2). After two such pauses the process stabilized, with
stance was tested for anaerobic biodegrability in different pH > 7.4, and after an additional 30 days the biomass was ready to
laboratories. The coefficients of variation calculated in this study start with the semi-continuous experiments. MPR (methane pro-
duction rate) and SMP (specific methane productivity) were low
during the start-up phase. Prior to addition of the tannery waste
Table 1
Characteristics of waste from a local tannery. (up to day 45) average MPR was 0.099 m3 m 3 d 1 and the SMP
was 0.198 m3 kg 1. After adding the tannery waste and stabilizing
Average Average VSS/TSS (%) TCOD
the process average MPR was 0.547 m3 m 3 d 1 and SMP was
TSS (%) VSS (%)
0443 m3 kg 1. The start-up process is shown in Fig. 2.
Waste sludge 6.6 3.6 54.5 40–70 g L 1
1
In experiment 1, in which the average ORL was
Waste fleshings 63.5 60.6 95.4 1600–1700 mg g
Waste trimmings 48.7 30.9 63.3 450–550 mg g 1
3.02 ± 0.02 kgVSS m 3 d 1, it can be seen that after a period of
14 days the process fully adapted to the lipid-rich substrate and

Table 2
BMP assay results.

SMP (m3 kg 1
VSS) SBP (m3 kg 1
VSS)
37 °C 55 °C Difference between 55 °C and 37 °C (%) 37 °C 55 °C Difference between 55 °C and 37 °C (%)
Waste sludge 0.617 0.535 13.3 0.973 0.851 12.5
Waste trimmings 0.377 0.308 18.3 0.547 0.499 8.8
Waste fleshings 0.552 0.649 17.6 0.891 1.074 20.5
Standard (glucose) 0.329 0.372 13.0 0.552 0.663 20.1
30 G.D. Zupančič, A. Jemec / Bioresource Technology 101 (2010) 26–33

a substantial rise in biogas production occurred. When the biogas 3.4. ASBR experiments
production stabilized the average MPR was 1.592 m3 m 3 d 1 and
SMP was 0.521 m3 kg 1; the pH was in the expected range at all We have performed five ASBR experiments. Experiment 1 had a
times. The average VSS removal was 62.5% and the average TCOD slightly lower OLR compared to the semi-continuous experiment 1.
degradation was 81.6%. In experiment 2, the average OLR was We gradually increased the OLR in each experiment up to the point
4.37 ± 0.02 kgVSS m 3 d 1 and we increased the loading rate by of organic overload. The results of the experiments are shown in
approximately 40–50%. After approximately two retention times Table 3 and Fig. 3. The nature of the ASBR process is such that ini-
it was evident, that the OLR was too high and the process started tial OLR values start higher and then decrease due to increasing
to collapse. Prior to the collapse, the average MPR was reactor volume. It is usually the case, and an advantage of the ASBR
1.939 m3 m 3 d 1 and the SMP was 0.446 m3 kg 1. The average process, that for short period of time (one day) higher loads do not
VSS removal was 52.8% and the average TCOD degradation was endanger the stability of the ASBR process although in the longer
71.9%. A higher MPR and a lower SMP was observed compared to term an overload is not sustainable. In all our ASBR and semi-
experiment 1, which was expected. In addition, lower treatment continuous experiments, we have determined the maximum yet
efficiencies were observed, which were decreasing as the experi- economically viable average OLR to be approximately
ment was terminated. At the beginning of the experiment the pH 4.0 kgVSS m 3 d 1 (corresponding to 8.5 kgTCOD m 3 d 1). With this
decreased to 7.1 and never recovered to the initial value. From OLR, organic overload can be avoided. Initial OLR values in ASBR
experiment 2 we can observe that with a slightly lower OLR the process were in this case 4.7 kgVSS m 3 d 1 (10.5 kgTCOD m 3 d 1),
process could be sustainable. and higher OLR’s resulted in overload. The process still finished

Fig. 2. Start-up and semi-continuous processes.

Table 3
ASBR experiments results.
3 1 3 1
Avgerage OLRTCOD (kg m d ) Avgerage OLRVSS (kg m d ) Avgerage MPR (m3 m 3
d 1
) SMPVSS (m3 kg 1
) VSSrem (%) TCODdeg (%)
Experiment 1 5.64 2.60 0.961 0.526 75.8 82.4
Experiment 2 6.44 2.96 1.127 0.543 73.1 81.6
Experiment 3 8.44 3.96 1.506 0.596 71.4 76.4
Experiment 4 8.83 4.10 1.339 0.481 66.7 74.2
Experiment 5 10.02 4.30 1.273 0.329 57.7 70.5
 G.D. Zupančič, A. Jemec / Bioresource Technology 101 (2010) 26–33 31

and 1.506 m3 m 3
d 1
when we used OLR slightly below the
maximum.

3.5. Comparison of semi-continuous operation with ASBR experiments

Comparing the results of ASBR experiments to the semi-contin-

uous experiments (with equal OLR), the following can be observed.
The SMP is slightly higher in ASBR, as a result of the prolonged
reaction time in days 6 and 7 when no substrate is added. This is
the most probable cause for better VSS removal. Tannery waste,
especially skin trimmings contain skin collagen, which takes longer
time to degrade. One could speculate that the prolonged reaction
time in the ASBR process has a beneficial effect on collagen degra-
dation, a similar effect having been observed by others (Lalitha
et al., 1994). When observing MPR the semi-continuous process
performed better suggesting that the ASBR process does require
larger reactor volumes than the conventional semi-continuous
process to achieve the same biogas yield. However the advantage
of the ASBR process is that it can be coupled directly to waste dis-
charge whereas the conventional process requires buffer tanks. In
our case the process is designed in this way to collect the waste
discharge from the tannery industry directly. Tannery production
process and the tannery wastewater treatment plant operates only
on first five weekdays and the waste sludge and tannery waste is
collected as reactor influent during these 5 days and processed
for a whole week. If a conventional process is to be used, a buffer
tank would be required for distribution of some of the waste for
the 7 day reactor influent.
Biogas composition was also determined in each experiment.
The usual CH4 concentration was between 55% and 70%. H2S con-
centration was between 0.6% and 1.0%, and the remainder was
CO2. We used a biogas absorption unit in which biogas was
stripped with NaOH solution. After the biogas was absorbed its
composition was 96% CH4 and 4% CO2. H2S was absorbed to less
than 0.0001% (below detection limit).

3.6. Additional concerns in digestion operation

While operating our digesters there appeared several concerns

that would be present in construction and operation of a biogas
plant utilizing tannery waste in the same manner. We have
decided to address several of these, which we judged may be
essential for the biogas plant constructors. These are: salinity and
chromium content, temperature fluctuations, ASBR size and ther-
mophilic heating demands.

3.6.1. Salinity and chromium content

The two main concerns in anaerobic treatment of tannery waste
are high salinity and chromium (VI) content of the waste. Chro-
mium originates from the conventional tanning process and is
present only in waste sludge; waste fleshings and skin trimmings
are removed prior to tanning do not contain any chromium. In
our case the leather production technology employed conventional
chromium tanning and also chromium free tanning, which were
equally prevalent. The Cr concentration measured in our waste
sludge was 268 mg L 1. Although waste sludge was added to reac-
tors in combination with waste fleshings and skin trimmings (50%
waste sludge, 40% waste fleshings and 10% skin trimmings, relative
to VSS), it made up 95–97% of the influent volume and it is reason-
able to assume that the maximum possible concentration of Cr in
Fig. 3. The course of ASBR experiments. our reactors would not exceed 268 mg L 1. There are large differ-
ences in the inhibitory concentrations of Cr among studies per-
formed in the literature, depending on the carbon source, sludge
characteristics and other factors reviewed in Altas (2009). The
without collapse, but efficiencies and biogas production were poor. inhibitory concentrations of Cr that cause 50% reduction in the
As expected, SMP and MPR reached a maximum of 0.596 m3 kg 1 cumulative methane production range from 27 to 3000 mg L 1
32 G.D. Zupančič, A. Jemec / Bioresource Technology 101 (2010) 26–33

(Altas, 2009). It was not the purpose of this study to elucidate, for semi-continuous process and the exchange volume for ASBR,
which concentrations of Cr are inhibitory to biogas production the total volume of reactors was calculated. The local tannery is
using our biomass, but to show that concentration conventionally a medium sized plant with an annual production of approximately
present enables satisfactory methane production. Since methane 5000 tonnes of leather. The annual water consumption is
production was high we assume that measured Cr content did 300,000 m3 and the wastewater treatment plant is 70,000 PE (pop-
not affect the process markedly. However, the production of tan- ulation equivalents) in size. The sizes and characteristics of digest-
nery waste in the absence of Cr remains to be investigated. ers are presented in Table 4.
The salinity of the substrates originates mostly from animal We can observe from these results that although the ASBR pro-
skin conservation during transport, and was observed in all three cess yields better results in biogas production and VSS removal it
of the substrates used in this study. In the process of preparation far exceeds the reactor volume commonly used in the conventional
of skins for tanning, they are soaked in water and washed thor- process. The most favourable ASBR process (experiment 3) exceeds
oughly. Some salinity remains in the waste fleshings and skin trim- the reactor volume in the semi-continuous process (experiment 1)
mings but most of it is washed into the wastewater. The by 24%. For the implementation of such process the characteristics
concentration of sodium in the reactor sludge was measured after such as possible additional buffer tanks, biogas yield, and efficiency
the addition of all three waste substrates. The concentration varied must be carefully considered. Calculation of cost analysis and eco-
between 2200 and 3100 mg L 1 averaging at 2650 mg L 1, which is nomic benefits can then show which process is actually more
lower than inhibition level reported in the literature (Jackson-Moss favourable.
et al., 1989; Feijoo et al., 1995). This concentration of sodium en-
ables satisfactory methane production. 3.6.3. Temperature fluctuations
In the course of our experiments, we established that the ther-
3.6.2. Digester scale up and size of tannery waste digestion mophilic temperature is absolutely crucial to successful tannery
In accordance with our goals we have also calculated the possi- waste digestion. It liquefies the fat in the waste it also increases
ble amount of biogas-methane produced and the sizes of full scale the rate of hydrolysis. To establish how a sudden temperature drop
digesters. The sizes were calculated using average daily waste pro- affects the biogas production, we performed an experiment in
duction in the local tannery, where we collected our substrates. which we decreased the reactor temperature from 55.8 °C to
Implying the hydraulic retention time gained in the experiments 51.4 °C (Fig. 4). The reactor conditions were similar to those in
the semi-continuous experiment 1 (OLR 3.0 kgVSS m 3 d 1;
6.5 kgTCOD m 3 d 1 and HRT of 17 days). The results showed a
Table 4
25% decrease in biogas production; the MPR decreased from
Full scale sizes and characteristics of digesters.
1.500 to 1.100 m3 m 3 d 1 and the SMP decreased from 0.500 to
Methane Digester ASBR VSS output 0.360 m3 kg 1. This is a significant reduction in biogas production;
(m3 per year) size (m3) exchange (tones per
therefore it is advisable to maintain a constant temperature,
volume (%) year)
slightly above 55 °C.
S-continuous 980,000 1700 – 705
When thermophilic digestion is applied energy intensive heat-
experiment 1
S-continuous 840,000 1200 – 887
ing demands must be considered. When such decrease in biogas
experiment 2 production occurs, most likely it will cause the decrease of temper-
ASBR experiment 1 990,000 3050 22 331 ature in the reactor, which will further worsen the digestion effi-
ASBR experiment 2 1020,000 2800 24 365 ciency. Therefore such temperature fluctuations must be avoided
ASBR experiment 3 1120,000 2100 32 444
at all cost, for successful thermophilic digestion. Good regulation
ASBR experiment 4 905,000 1800 38 485
ASBR experiment 5 620,000 1750 39 555 devices and safety systems should be installed when building such
thermophilic digesters.

Fig. 4. Effects of reactor temperature decrease.

 G.D. Zupančič, A. Jemec / Bioresource Technology 101 (2010) 26–33 33

3.6.4. Concerns about sludge heating demands APHA-AWA-WEF, 2005. Standard Methods for the Examination of Waters and
Wastewaters, 21st ed. Washington, DC.
Thermophilic anaerobic digestion is considered energy inten-
BIPM-IEC-IFCC-ISO-IUPAC-IUPAP-OIML, 1995. Guide to the Expression of
sive therefore we feel that in this context the feasibility of thermo- Uncertainty in Measurement. International Organization for Standardization,
philic digestion must be addressed. The heating requirements of Geneva, pp. 1–100.
anaerobic digesters can be divided roughly into two groups: sub- Cenni, F., Dondo, G., Tombetti, F., 1982. Anaerobic digestion of tannery wastes.
Agricultural Wastes 4 (3), 241–243.
strate heating to thermophilic temperature and heat losses of all Cirne, D.G., Paloumet, X., Björnsson, L., Alves, M.M., Mattiasson, B., 2007. Anaerobic
kinds. Some 80–90% of heating in a conventional digester design digestion of lipid-rich waste – effects of lipid concentration. Renewable Energy
are necessary for substrate heating (Zupančič and Roš, 2003). 32, 965–975.
Deublein, D., Steinhauser, A., 2008. Biogas from Waste and Renewable Resources.
Therefore, in the case of thermophilic digestion this substantial de- Wiley-VCH, Verlag, Weinheim, Germany.
mand cannot be satisfied with waste heat from a conventional Dhayalan, K., Fathima, N.N., Gnanamani, A., Rao, J.R., Nair, B.U., Ramasami, T., 2007.
practise – a combined heat and power cogeneration unit. In such Biodegradability of leathers through anaerobic pathway. Waste Management
27 (6), 760–767.
cases, heat regeneration from effluent to influent can be applied European Comission, 2001. European Comission-Directorate-General for
and the shortcoming of needed heat made up. If such regeneration Environment. Survey on Wastes Spread on Land-Final Report. <http://
is applied, simultaneous intake and release of the substrate from ec.europa.eu/environment/waste/studies/compost/landspreading.htm>.
Feijoo, G., Soto, M., Méndez, R., Lema, J.M., 1995. Sodium inhibition in the anaerobic
the digester is necessary. The ASBR process is not suitable for such digestion process: antagonism and adaptation phenomena. Enzyme and
heat regeneration technology because the intake and release are Microbial Technology 17 (2), 180–188.
timely mismatched, they do not occur at the same time. Therefore, Ioannis, S., Bagley, D.M., 2002. Improving anaerobic sequencing batch reactor
performance by modifying operational parameters. Water Research 36 (1),
additional heat sources must be found to address the heating de-
363–367.
mands. In some cases there is enough waste heat in the tanning ISO 11734:1994. Water Quality – Evaluation of the ‘‘Ultimate” Anaerobic
process to be used but if there is no such possibility, a conventional Biodegradability of Organic Compounds in Digested Sludge – Method by
semi-continuous process is more practical than the ASBR process, Measurement of the Biogas Production.
Jackson-Moss, C.A., Duncan, J.R., Cooper, D.R., 1989. The effect of sodium chloride on
because the intake and release do occur simultaneously. Consider- anaerobic digestion. JALCA 84, 266–271.
ing that ASBR process does perform better and produces more bio- Lalitha, K., Swaminathan, K.R., Padma Bai, R., 1994. Kinetics of biomethanation of
gas than semi-continuous process, needs no waste buffer tanks and solid tannery waste and the concept of interactive metabolic control. Applied
Biochemistry and Biotechnology 47 (1), 73–81.
on the other hand it is less suitable for thermophilic temperature a Lee, J.G., Hur, J.M., Chang, D., Chung, T.H., 2001. Performance characterization of
careful analysis of all factors surrounding such a plant (additional anaerobic sequencing batch reactor process for digestion of night soil. Water
heat sources, waste buffer tanks. . .) for each specific case must Science Technology 43, 27–34.
Lefebvre, O., Vasudevan, N., Torrijos, M., Thanasekaran, K., Moletta, R., 2006.
be made, before it can be clearly determined which technology Anaerobic digestion of tannery soak liquor with an aerobic post-treatment.
would actually perform better. It is also possible to devise a hybrid Water Research 40 (7), 1492–1500.
system, retaining the coupling the digester to the waste genera- Montoneri, E., Rizzi, G., Rizzi, A., Mordenti, A., Bauli, A., Riolfatti, M., Pellegrini, L.,
1994. Hydrolysis of tannery wastes to protein meal for animal feedstuffs – a
tion. An example of this would be a 5 day conventional semi-con- process and product evaluation. Journal of Chemical Technology and
tinuous process followed by 2 days of resting. The biogas yield, Biotechnology 59 (1), 91–99.
digester sizes and process efficiencies of such a system would most OECD 311:2006. OECD Guidelines for the Testing of Chemicals: Anaerobic
Biodegradability of Organic Compounds in Digested Sludge: by Measurement
likely fall between those of the conventional and the ASBR
of Gas Production.
processes. Özgünay, H., Colak, S., Zengin, G., Sari, O., Sarikahya, H., Yüceer, L., 2007.
Performance and emission study of biodiesel from leather industry
prefleshings. Waste Management 27 (12), 1897–1901.
4. Conclusions Raju, A., Rose, C., Muralidhara Rao, N., 1997. Enzymatic hydrolysis of tannery
fleshings using chicken intestine proteases. Animal Feed Science and
The anaerobic digestion potential of different tannery wastes, Technology 66 (1–4), 139–147.
Song, Z., Williams, C.J., Edyvean, R.G.J., 2003. Tannery wastewater treatment using
including fleshings, skin trimmings and tannery wastewater sludge an upflow anaerobic fixed biofilm reactor (UAFBR). Environmental Engineering
was evaluated. Conventional semi-continuous and ASBR experi- Science 20 (6), 587–599.
ments have shown that OLR of 4.0 kg m 3 d 1 of VSS (correspond- Song, Z., Williams, C.J., Edyvean, R.G.J., 2004. Treatment of tannery wastewater by
chemical coagulation. Desalination 164 (3), 249–259.
ing to 8.5 kg m 3 d 1 of TCOD) represents the limits of successful
Thanikaivelan, P., Rao, J.R., Nair, B.U., Ramasami, T., 2005. Recent trends in leather
and economic operation. In the ASBR process with OLR of making: processes, problems, and pathways. Critical Reviews in Environmental
3.96 kg m 3 d 1 we determined a SMP of 0.596 m3 kg 1, and a Science and Technology 35 (1), 37–79.
Tunick, M.H., Bailey, D.G., Moore, J.H., Cooper, J., 1985. Anaerobic–aerobic treatment
VSS removal of 71.4%. The anaerobic digestion of selected sub-
of vegetable tanning waste. JALCA 80, 101–105.
strates was possible at 268 mg L 1 of chromium (VI) and Urbaniak, M., 2006. Gaseous efficiency of methane fermentation of chromium free
2650 mg L 1 of sodium. We can conclude that tannery wastes are tannery wastes. Journal of the Society of Leather Technologists and Chemists 91,
appropriate substrates for biogas production. 59–62.
Wang, Z., Wang, W., Zhang, X., Zhang, G., 2002. Digestion of thermally hydrolyzed
sewage sludge by anaerobic sequencing batch reactor. Journal of Hazardous
Acknowledgements Materials 162 (2–3), 799–803.
Yu, H.-Q., Fang Herbert, H.P., Gu, G.-W., 2002. Comparative performance of
mesophilic and thermophilic acidogenic upflow reactors. Process
Authors wish to thank Mr. Gregor Grom, Mr. Matjaž Omerzel, Biochemistry 38 (3), 447–454.
Mr. Gasan Osojnik and Mrs Liljana Piščanec for help in the pre- Zhang, R.H., Dugba, P.N., 2000. Evaluation of two-stage anaerobic sequencing batch
sented study. reactor systems for animal wastewater treatment. Transactions of the ASAE 43
(6), 1795–1801.
Zupančič, G.D., Stražiščar, M., Roš, M., 2007. Treatment of brewery slurry in
References thermophilic anaerobic sequencing batch reactor. Bioresource Technology 98
(14), 2714–2722.
Altas, L., 2009. Inhibitory effect of heavy metals on methane-producing anaerobic Zupančič, G.D., Roš, M., 2003. Heat and energy requirements in thermophilic
granular sludge. Journal of Hazardous Materials 162, 1551–1556. anaerobic sludge digestion. Renewable Energy 28 (14), 2255–2267.