Document [original]

Evaluation of Different Pre-treatments of

Chromium Leather Waste and Their Use

in Biogas Production

vorgelegt von

M. Sc.

Carolina Scaraffuni Gomes

ORCID: 0000-0002-6185-8946

an der Fakultät III – Prozesswissenschaften

der Technischen Universität Berlin

zur Erlangung des akademischen Grades

Doktorin der Ingenieurwissenschaften

- Dr.-Ing. -

genehmigte Dissertation

Promotionsausschuss:

Vorsitzender: Prof. Dr.-Ing. Matthias Kraume

Gutachter: Prof. Dr.-Ing. habil. Jens-Uwe Repke

Gutachter: Dr. Michael Meyer

Gutachter: Prof. Dr. Thomas Bley

Tag der wissenschaftlichen Aussprache: 18. Februar 2020

Berlin 2020

Acknowledgements

Firstly, I would like to express my sincere gratitude to my advisors, Prof. Repke

and Dr. Meyer, for the continuous support of my Ph.D. study and related re-

search. Their guidance helped me in all the time of research and writing of this

thesis.

My sincere thanks goes to FILK that provided me an opportunity to join the

team and gave access to the laboratory and research facilities. I also thank my

work colleagues at FILK for the daily support and stimulating discussions.

Without this support it would not be possible to conduct this research.

Lastly, I would like to thank Patricia, Stephan and Hermes who were always

there for me throughout writing this thesis.

Part of the results presented in this dissertation were already published in the

following papers:

Chapters 4.1. and 4.2.1.1.:

Gomes, C. S., Repke, J-U., Meyer, M. (2017). Different pre-treatments of chrome

tanned leather waste and their use in the biogas production. XXXIV IULTCS

Congress, Chennai, India.

Chapter 4.2.1.1.:

Gomes, C.S. (2017). Tanned leather waste for biogas production. International

Leather Maker 25, 80–81.

Chapters 4.1. and 4.2.1.2.:

Gomes, C. S., Repke, J-U., Meyer, M. Investigation of different pre-treatments

of chromium leather shavings to improve biogas production. Submitted to

Waste and Biomass Valorization.

Chapters 4.2.1.3. and 4.2.1.4.:

Gomes, C. S., Repke, J-U., Meyer, M. (2018). Diauxie during biogas production

from collagen-based substrates. 7TH International Symposium on Energy from

Biomass and Waste, Venice, Italy.

Gomes, C. S., Repke, J-U., Meyer, M. (2019). Diauxie during biogas production

from collagen-based substrates. Renewable Energy 141, pp. 20–27. DOI:

https://doi.org/10.1016/j.renene.2019.03.123.

Chapter 4.2.2.:

Gomes, C. S., Repke, J-U., Meyer, M. (2019). Use of different pre-treated chro-

mium leather shavings to produce biogas in continuous scale. XXXV IULTCS

Congress, Dresden, Germany.

Gomes, C. S., Repke, J-U., Meyer, M. (2019). The effect of various pre-treatment

methods of chromium leather shavings in continuous biogas production. Engi-

neering in Life Sciences pp. 1-11. DOI: https://doi.org/ 10.1002/elsc.201900127.

Zusammenfassung/Abstract

Deutsch

Die weltweite wirtschaftliche Bedeutung der Lederindustrie ist unumstritten,

da sie ein Nebenprodukt der Fleischindustrie, die Häute, in ein Produkt mit

hoher Wertschöpfung umwandelt. Allerdings werden durch die Verarbeitung

von Leder jährlich tausende von Tonnen neuer fester Nebenprodukten produ-

ziert, die im Gegensatz zu Häuten eine geringere biologische Abbaubarkeit auf-

weisen. Üblicherweise erfolgt die Gerbung unter Einsatz basischer, dreiwerti-

ger Chromsalze. Aufgrund der enthaltenen Schwermetalle dürfen die Gerbe-

reinebenprodukte, wie die Falzspäne chromgegerbter Häute und Beschneide-

abfälle chromgegerbter Leder nicht deponiert werden. Dennoch erfolgt die Ent-

sorgung überwiegend durch Deponierung oder Verbrennung, ungeachtet mög-

licher Konsequenzen für die Umwelt. Diese Nebenprodukte aus der Lederher-

stellung sind verantwortlich für die größten ökologischen Herausforderungen,

die durch Gerbereien verursacht werden.

Da die chromhaltigen festen Gerbereinebenprodukte hauptsächlich aus Kol-

lagen (organischen Stoffen) bestehen, ist eine Verwertung der Gerbereiabfälle

zur Biogasproduktion durch den anaeroben Abbau möglich. Die größte Her-

ausforderung ist die Stabilität des Nebenprodukts gegenüber Temperaturen

von bis zu 110 ° C und enzymatischen Abbau. Ursache für diese hohe Stabilität

ist die bei Kollagen typische dreidimensionale native Struktur, die durch zu-

sätzliche chemischer Vernetzungen zwischen den Kollagenfasern durch Cr3+

Salze weiter stabilisiert wird. Daher werden chromhaltige feste Gerbereineben-

produkte bisher nicht industriell zur Produktion von Biogas verwendet.

Das Hauptziel dieser Arbeit ist die Beschleunigung und Verbesserung der Bio-

gasgewinnung unter Einsatz von chromhaltigen festen Gerbereinebenproduk-

ten als Substrat, sodass diese industriell genutzt werden können. Dazu muss

die stabile Struktur zuvor denaturiert werden, um einen enzymatischen Abbau

während der Biogasproduktion zu erreichen. Zu diesem Zweck wurde das Sub-

strat durch Extrusion und hydrothermische Behandlung (kontinuierliche und

als Batch) vorbehandelt. Die Charakterisierung der vorbehandelten Gerbe-

reinebenprodukte erfolgte durch Rasterelektronenmikroskopie (SEM), Diffe-

renzkalorimetrie (DSC), Abbau durch Trypsin und Kollagenase sowie durch

Messung ihrer Löslichkeit in Wasser. Anschließend wurde das Biogasproduk-

tionspotential des vorbehandelten Substrates in Batch und in kontinuierlichen

Versuchen bestimmt, um die Verwertung von vorbehandelten schwermetall-

haltigen festen Gerbereinebenprodukten als Substrat zur Produktion von Bio-

gas zu bewerten.

Die Ergebnisse zeigen, dass es möglich ist, die kollagene Struktur chromhaltiger

fester Gerbereinebenprodukte durch Vorbehandlung abzubauen. Die Extru-

sion und hydrothermische Behandlung des Substrates in Batch-Versuchen re-

duzierten die lag-Phase der Biogasproduktion um vier bis fünf Tage und ver-

ringerten die Abfallmenge. Bei Batch-Versuchen mit Substratüberladung durch

vorbehandeltes Material wurde teilweise eine Diauxie beobachtet. Der Zusam-

menhang von Diauxie und übermäßiger Produktion flüchtiger Fettsäuren

wurde untersucht. Im Gegensatz zu den extrudierten Chromfalzspänen (Reste

der Halbfabrikatherstellung) zeigten extrudierte Lederabfälle, welche mittels

Nasszurichtung und Zurichtung in Gerbereien hergestellt wurden, trotz Vor-

behandlung (Extrusion) ein schlechtes Abbauverhalten. Kontinuierliche Versu-

che mit vorbehandelten chromhaltigen festen Gerbereinebenprodukten lassen

den Schluss zu, dass es möglich ist, eine höhere Beladungsrate zu verwenden

und die tägliche Methanproduktion zu steigern. Den Ergebnissen zufolge, kön-

nen vorbehandelte chromhaltige feste Gerbereinebenprodukte als Substrat zur

Produktion von Biogas im industriellen Maßstab eingesetzt werden.

Schlüsselwörter: Biogasproduktion; Anaerober Abbau; Feste Gerbereinebenprodukt

Zusammenfassung/Abstract

English

The worldwide economic importance of the leather industry is undeniable, as

it converts a by-product of the meat industry, the hides, into a value-added

product. However, processing of leather generates thousands of tons of new

solid waste annually and, unlike hides, this waste has a low biodegradability.

Usually, the tanning process is carried out using basified trivalent chromium

salts. As a result, a substantial part of the solid waste is chromium-tanned and

normally disposed of through landfill or incineration processes, despite the pos-

sible ecological consequences. This waste is responsible for the main ecological

challenges caused by tanneries.

As chromium leather waste mainly consists of collagen (organic matter), this

waste can be considered for biogas production through anaerobic digestion.

The main challenge to be overcome is the stability of the waste towards temper-

atures of up to 110 °C and enzymatic degradation. This high stability is caused

by the three-dimensional native structure typical for collagen, and additional

chemical cross-links between the collagen fibres achieved by Cr3+ salts. There-

fore, hitherto chromium leather waste is not utilized industrially to produce bi-

ogas.

The main goal of this study is to accelerate and improve the biogas production

process using chromium leather waste as a substrate in a way that it can be used

industrially. However, to achieve enzymatic degradation this stable structure

has to be denatured, otherwise, the generation of biogas is hindered. For this

purpose, the following pre-treatments were performed on the chromium

leather waste: extrusion, hydrothermal treatment, and autoclave. The

pre-treated waste was evaluated using scanning electron microscopy (SEM),

differential scanning calorimetry (DSC), degradation by trypsin and colla-

genase, and by measuring their solubility in water. Subsequently, the biogas

production potential of the pre-treated waste was investigated using batch and

continuous trials to examine the feasibility of using pre-treated chromium

leather waste as substrate to produce biogas.

Results show that it was possible to degrade the collagenous structure of the

chromium leather waste through pre-treatment. Extrusion and hydrothermal

treatment of the substrate in batch trials reduced the lag-phase of the biogas

production by four to five days and reduced the remaining waste compared to

the untreated waste. In some batches, when using a substrate overload of

pre-treated waste, diauxie was observed. The relation of diauxie and excessive

production of volatile fatty acids was studied. Results for extruded leather off-

cuts show that the wet end and finishing process in tanneries make this material

more inaccessible to degradation, even after pre-treatment. Pre-treated chro-

mium leather waste used as substrate in continuous trials demonstrated that it

was possible to use a higher loading rate and obtain a higher daily methane

production. These results showed that pre-treated chromium leather waste can

be used as substrate to produce biogas on industrial scale.

Keywords: Biogas production; Anaerobic digestion; Leather waste

Content 
Introduction .................................................................................................. 1 
1 Motivation .......................................................................................................... 4 
State of the Art .............................................................................................. 7 
1 Chromium leather waste .................................................................................. 7 
1.1 Collagen structure and stability ............................................................... 8 
1.2 Chromium-based tanning process ........................................................... 9 
1.3 Alternative management of chromium leather waste ........................ 10 
2 Anaerobic digestion ........................................................................................ 13 
2.1 Anaerobic digestion of tannery waste ................................................... 14 
2.2 Microbiology: trophic groups of the anaerobic digestion .................. 17 
2.3 Biogas production studies ....................................................................... 20 
2.4 Inhibitors ................................................................................................... 21 
2.5 Theoretical biogas yield: Buswell equation .......................................... 24 
3 The use of chromium leather waste in the biogas production ................. 24 
Materials and Methods .............................................................................. 27 
1 Materials ........................................................................................................... 27 
2 Pre-treatments of the chromium leather waste ........................................... 29 
2.1 Extrusion ................................................................................................... 29 
2.2 Hydrothermal treatment ......................................................................... 32 
2.3 Autoclave .................................................................................................. 33 
3 Chemical analyses ........................................................................................... 35 
3.1 Analyses of the pre-treated chromium leather waste ......................... 35 

Content 
 
3.2 Analyses of the biomass .......................................................................... 39 
4 Biogas Production ........................................................................................... 40 
4.1 Laboratory scale: batch tests ................................................................... 41 
4.2 Pilot scale: continuous tests .................................................................... 44 
4.3 Possible Sources of Errors ....................................................................... 46 
Results and Discussion .............................................................................. 49 
1 Chemical analyses of the pre-treated samples ............................................ 49 
1.1 Characterization of the pre-treated samples ........................................ 49 
1.2 Scanning electron microscopy ................................................................ 51 
1.3 Differential scanning calorimetry and enzyme assays ....................... 55 
1.4 Solubility in water .................................................................................... 60 
2 Biogas Production ........................................................................................... 64 
2.1 Laboratory scale: batch tests ................................................................... 64 
2.2 Pilot scale: continuous tests .................................................................... 98 
Conclusions ............................................................................................... 123 
 
 

i 
 
List of figures 
FIG. 2.1: DIFFERENT STRUCTURAL LEVELS OF COLLAGEN MATERIALS AND THEIR DENATURATION TEMPERATURE 
(TD). TRIPLE HELIX OF THE COLLAGEN MOLECULE (A), ITS ASSEMBLY IN A TISSUE (B), CHEMICAL CROSS-LINKS 
FORMATION WHILE TANNING WITH CHROMIUM SALTS (C), AND COLLAGEN DENATURATION TO GELATIN 
THROUGH TEMPERATURE EXPOSURE (D). .................................................................................... 10 
FIG. 2.2: PHASES OF ANAEROBIC DIGESTION (1) HYDROLYSIS, (2) FERMENTATION OR ACIDOGENESIS, (3) 
ACETOGENESIS, AND (4) METHANOGENESIS (ADAPTED FROM MURPHY AND THAMSIRIROJ, 2013). ....... 17 
FIG. 2.3: FORMATION OF MONOMERS (ADAPTED FROM DEUBLEIN AND STEINHAUSER, 2008). ..................... 18 
FIG. 2.4: TYPICAL SHAPES OF BIOGAS FORMATION CURVES (ADAPTED FROM VDI 4630, 2006). ................... 23 
FIG. 3.1: CHROMIUM SHAVINGS (A) AND LEATHER OFFCUTS (B). ............................................................. 28 
FIG. 3.2: EXTRUDER WERNER & PFLEIDERER ZSK 25. .......................................................................... 30 
FIG. 3.3: OPERATION OF THE EXTRUDER MACHINE ................................................................................ 30 
FIG. 3.4: CHROMIUM SHAVINGS EXTRUDED (E) DRY (D) OR WET (W). ..................................................... 31 
FIG. 3.5: CONTINUOUS AUTOCLAVE AND REFINER ANDRITZ - FRONT (A) AND BACK VIEW (B). ..................... 32 
FIG. 3.6: CHROMIUM SHAVINGS TREATED HYDROTHERMALLY (H) AFTER COOLING. ..................................... 33 
FIG. 3.7: LABORATORY EQUIPMENT USED AS AUTOCLAVE. ...................................................................... 34 
FIG. 3.8: AUTOCLAVED (A) CHROMIUM SHAVINGS AFTER DRYING. ........................................................... 35 
FIG. 3.9: EXAMPLE OF A DSC CURVE. ................................................................................................ 37 
FIG. 3.10: BIOGAS REACTORS IN BATCH SCALE AND BIOGAS GENERATION MEASUREMENT............................. 41 
FIG. 3.11: TWO-PHASE GOMPERTZ MODEL. ....................................................................................... 43 
FIG. 3.12: CONTINUOUS ANAEROBIC DIGESTION TEST APPARATUS (SOURCE: MYFERM I MANUAL – LANDGRAF 
LABORSYSTEME HLL GMBH). .................................................................................................. 45 
FIG. 4.1: SEM IMAGES OF CHROMIUM SHAVINGS (A), LEATHER OFFCUTS (B), SHAVINGS EXTRUDED DRY AT 100 
°C (C), EXTRUDED LEATHER OFFCUTS (D), SHAVINGS TREATED HYDROTHERMALLY AT 140 °C (E), SHAVINGS 
TREATED HYDROTHERMALLY AT 170 °C (F), AUTOCLAVED OVER A PERIOD OF 3 MINUTES (G), AND 
AUTOCLAVED OVER A PERIOD OF 384 MINUTES (H) AT 60,000 × MAGNIFICATION. ............................ 52 
FIG. 4.2: SEM IMAGES FOR EDX OF CHROMIUM SHAVINGS (A), SHAVINGS EXTRUDED DRY AT 100 °C (B), 
SHAVINGS EXTRUDED WET AT 170 °C (C), SHAVINGS TREATED HYDROTHERMALLY AT 150 °C (D), SHAVINGS 
TREATED HYDROTHERMALLY AT 170 °C (E) AT DIFFERENT MAGNIFICATIONS. ..................................... 54 
FIG. 4.3: ENTHALPY OF THE DENATURATION PROCESS AND DEGREE OF DEGRADATION BY TRYPSIN OF EXTRUDED 
SHAVINGS AND OFFCUTS (A), SHAVINGS TREATED HYDROTHERMALLY AS A FUNCTION OF THE PRE-
TREATMENT TEMPERATURE (B), AND AUTOCLAVED SHAVINGS AS A FUNCTION OF THE PRE-TREATMENT TIME 
(C) AND CHROMIUM SHAVINGS. ................................................................................................ 56 

List of figures 
ii 
 
FIG. 4.4: DEGREE OF DEGRADATION BY COLLAGENASE OF WHITE HIDE POWDER, UNTREATED CHROMIUM 
SHAVINGS, AND SHAVINGS EXTRUDED DRY AT 100 °C AS A FUNCTION OF TIME. ..................................59 
FIG. 4.5: SOLUBILITY IN WATER AT A TEMPERATURE OF 37 °C OF EXTRUDED SHAVINGS AND OFFCUTS (A), 
SHAVINGS TREATED HYDROTHERMALLY AS A FUNCTION OF THE PRE-TREATMENT TEMPERATURE (B), AND 
AUTOCLAVED SAMPLES AS A FUNCTION OF THE PRE-TREATMENT TIME (C) AND CHROMIUM SHAVINGS. ....61 
FIG. 4.6: CUMULATIVE BIOGAS PRODUCTION AND TWO-PHASE GOMPERTZ SIMULATION (RED LINE) FOR 
CHROMIUM SHAVINGS (S/I = 2.6), SHAVINGS EXTRUDED WET AT 170 °C (S/I = 3.2), AND DRY AT 100 °C 
(S/I = 2.9) IN AGITATED BIOREACTORS (A) AND IN NON-AGITATED BIOREACTORS (B). ..........................65 
FIG. 4.7: CUMULATIVE BIOGAS PRODUCTION AND TWO-PHASE GOMPERTZ SIMULATION (RED LINE) FOR THE 
UNTREATED SHAVINGS AND GELATIN (A), SHAVINGS EXTRUDED DRY (B), SHAVINGS TREATED 
HYDROTHERMALLY (C), AND SHAVINGS EXTRUDED WET (D) IN AGITATED BIOREACTORS. ........................69 
FIG. 4.8: CUMULATIVE BIOGAS PRODUCTION AND TWO-PHASE GOMPERTZ SIMULATION (RED LINE) FOR THE 
UNTREATED CHROMIUM SHAVINGS (A), GELATIN (B), SHAVINGS EXTRUDED DRY AT 100 °C (C), AND 
EXTRUDED OFFCUTS (D) WITH DIFFERENT SUBSTRATE QUANTITIES IN AGITATED BIOREACTORS. ...............75 
FIG. 4.9: REPRODUCTION OF THE BIOGAS YIELD, SAMPLING TIME, AND VOLATILE FATTY ACID (VFA) 
CONCENTRATIONS PRESENT IN SAMPLES COLLECTED FOR GELATIN (A), SHAVINGS EXTRUDED DRY AT 100 °C 
(B), AND EXTRUDED OFFCUTS (C) IN AGITATED BIOREACTORS (SOURCE: GOMES ET AL., 2019A). ............85 
FIG. 4.10: CONCENTRATION OF AMMONIA FOUND IN THE BIOMASS COLLECTED AT DIFFERENT REACTION TIMES 
FOR GELATIN (A) AND SHAVINGS EXTRUDED DRY AT 100 °C (B) IN AGITATED BIOREACTORS. ..................88 
FIG. 4.11: COD OF THE BIOMASS SAMPLES COLLECTED AT DIFFERENT REACTION TIMES FOR GELATIN (A), 
SHAVINGS EXTRUDED DRY AT 100 °C (B), AND EXTRUDED OFFCUTS (C) IN AGITATED BIOREACTORS. ........92 
FIG. 4.12: CUMULATIVE BIOGAS PRODUCTION FOR THE GELATIN IN THE BIOREACTORS WITH ADDED ACETIC ACID 
(A), PROPIONIC ACID (B), AND ISOBUTYRIC ACID (C). ......................................................................95 
FIG. 4.13: CUMULATIVE BIOGAS PRODUCTION AND TWO-PHASE GOMPERTZ SIMULATION (RED LINE) FOR THE 
GELATIN IN TWO DIFFERENT TRIALS WITH THE SAME CONDITIONS. ....................................................98 
FIG. 4.14: TIME PLOT OF CONTINUOUS ANAEROBIC DIGESTION OF THE UNTREATED CHROMIUM SHAVINGS (A), 
SHAVINGS EXTRUDED DRY AT 100 °C (B), AND SHAVINGS TREATED HYDROTHERMALLY AT 140 °C AND 150 
°C (C) (SOURCE: GOMES ET AL., 2019B). .................................................................................100 
FIG. 4.15: QUALITY OF THE BIOGAS PRODUCED DURING DIGESTION OF CHROMIUM SHAVINGS (A), SHAVINGS 
EXTRUDED DRY AT 100 °C (B), AND SHAVINGS TREATED HYDROTHERMALLY AT 140 AND 150 °C (C). ...102 
FIG. 4.16: VOLATILE FATTY ACID CONCENTRATIONS DURING ANAEROBIC DIGESTION OF UNTREATED CHROMIUM 
SHAVINGS (A), SHAVINGS EXTRUDED DRY AT 100 °C (B), AND SHAVINGS TREATED HYDROTHERMALLY AT 
140 °C AND 150 °C (C) (SOURCE: GOMES ET AL., 2019B). .........................................................105 
FIG. 4.17: CONCENTRATION OF TOTAL SULFATE IN THE BIOMASS COLLECTED AND HYDROGEN SULFIDE IN THE 
BIOGAS FORMED AT DIFFERENT REACTION TIMES FOR THE REACTOR FED WITH CHROMIUM SHAVINGS (A), 
SHAVINGS EXTRUDED DRY AT 100 °C (B), AND SHAVINGS TREATED HYDROTHERMALLY AT 140 AND 150 °C 
(C). ...................................................................................................................................110 
FIG. 4.18: CONCENTRATION OF IRON IN THE BIOMASS COLLECTED AND MASS OF ADDED IRON CHLORIDE AT 
DIFFERENT REACTION TIMES FOR THE REACTOR FED WITH UNTREATED CHROMIUM SHAVINGS (A), SHAVINGS 
EXTRUDED DRY AT 100 °C (B), AND SHAVINGS TREATED HYDROTHERMALLY AT 140 °C AND 150 °C (C).
 ........................................................................................................................................112 

List of figures 
iii 
 
FIG. 4.19: CONCENTRATION OF AMMONIA FOUND IN THE BIOMASS COLLECTED AT DIFFERENT REACTION TIMES 
FOR THE REACTOR FED WITH UNTREATED CHROMIUM SHAVINGS (A), SHAVINGS EXTRUDED DRY AT 100 °C 
(B), AND SHAVINGS TREATED HYDROTHERMALLY AT 140 °C AND 150 °C (C). .................................. 113 
FIG. 4.20: COLLAGEN CONTENT OF THE BIOMASS SAMPLES COLLECTED FROM THE REACTOR FED WITH SHAVINGS 
TREATED HYDROTHERMALLY AFTER SUBSTRATE FEEDING (LOGARITHMIC TIMELINE). .......................... 118 
FIG. 4.21: CUMULATIVE COD DEGREE OF DEGRADATION DURING DIGESTION AND COD IN THE BIOMASS SAMPLES 
COLLECTED AT DIFFERENT REACTION TIMES FOR THE REACTOR FED WITH CHROMIUM SHAVINGS (A), 
SHAVINGS EXTRUDED DRY AT 100 °C (B) AND SHAVINGS TREATED HYDROTHERMALLY AT 140 AND 150 °C 
(C). ................................................................................................................................... 120 
FIG. 5.1: SEM IMAGES OF SHAVINGS EXTRUDED DRY AT 130 °C (A), SHAVINGS EXTRUDED DRY AT 150 °C (B), 
SHAVINGS EXTRUDED DRY AT 170 °C (C), SHAVINGS EXTRUDED WET AT 100 °C (D), SHAVINGS EXTRUDED 
WET AT 130 °C (E), SHAVINGS EXTRUDED WET AT 150 °C (F), SHAVINGS EXTRUDED WET AT 170 °C (G), 
SHAVINGS TREATED HYDROTHERMALLY AT 150 °C (H), SHAVINGS AUTOCLAVED OVER A PERIOD OF SIX 
MINUTES (I), SHAVINGS AUTOCLAVED OVER A PERIOD OF TWELVE MINUTES (J), AND SHAVINGS AUTOCLAVED 
OVER A PERIOD OF 24 MINUTES (K), AND SHAVINGS AUTOCLAVED OVER A PERIOD OF 48 MINUTES (L), AND 
SHAVINGS AUTOCLAVED OVER A PERIOD OF 96 MINUTES (M), AND SHAVINGS AUTOCLAVED OVER A PERIOD 
OF 192 MINUTES (N) AT 60,000 × MAGNIFICATION. .................................................................. 145 

List of tables

TAB. 1.1: QUANTITY OF SOLID WASTE PRODUCED IN THE LEATHER-MAKING PROCESS. ................................... 2

TAB. 3.1: CHROMIUM SHAVINGS AND LEATHER OFFCUTS CHARACTERIZATION. ........................................... 28

TAB. 3.2: EXTRUDED SAMPLES AND THEIR PRE-TREATMENT CONDITIONS. ................................................. 31

TAB. 3.3: SAMPLES TREATED HYDROTHERMALLY AND THEIR PRE-TREATMENT CONDITIONS. .......................... 33

TAB. 3.4: AUTOCLAVED SAMPLES AND THEIR PRE-TREATMENT TIME. ....................................................... 35

TAB. 4.1: CHARACTERIZATION OF THE PRE-TREATED AND UNTREATED SAMPLES. ........................................ 50

TAB. 4.2: PARAMETERS OF BIOGAS PRODUCTION FROM THE TWO-PHASE GOMPERTZ EQUATION. .................. 65

TAB. 4.3: BIOMASS CHARACTERIZATION AFTER DIGESTION AND DEGRADATION DEGREE OF COLLAGEN FOR

AGITATED AND NON-AGITATED BIOREACTORS. ............................................................................. 67

TAB. 4.4: PARAMETERS OF BIOGAS PRODUCTION FROM THE TWO-PHASE GOMPERTZ EQUATION. .................. 70

TAB. 4.5: BIOMASS CHARACTERIZATION AFTER DIGESTION AND DEGRADATION DEGREE OF COLLAGEN FOR

BIOREACTORS TESTING DIFFERENT PRE-TREATMENTS. .................................................................... 72

TAB. 4.6: PARAMETERS OF BIOGAS PRODUCTION FROM THE TWO-PHASE GOMPERTZ EQUATION. .................. 76

TAB. 4.7: BIOMASS CHARACTERIZATION AFTER DIGESTION AND DEGRADATION DEGREE OF COLLAGEN FOR

BIOREACTORS TESTING DIFFERENT SUBSTRATE TO INOCULUM RATIOS. .............................................. 80

TAB. 4.8: BIOGAS FORMATION POTENTIAL OF DIFFERENT SUBSTRATES. .................................................... 82

TAB. 4.9: BIOMASS CHARACTERIZATION DURING DIGESTION REGARDING ITS ORGANIC AND COLLAGEN OR

HYDROXYPROLINE CONTENT AND BIOGAS YIELD FOR BIOREACTORS INVESTIGATING DIAUXIE. ................. 90

TAB. 4.10: BIOMASS CHARACTERIZATION DURING DIGESTION, DEGRADATION DEGREE OF HYDROXYPROLINE, AND

BIOGAS YIELD FOR BIOREACTORS INVESTIGATING DIAUXIE. .............................................................. 96

TAB. 4.11: PARAMETERS OF BIOGAS PRODUCTION FROM THE TWO-PHASE GOMPERTZ EQUATION. ................ 98

TAB. 4.12: PH VALUES AND CHROMIUM CONTENT OF THE SAMPLES COLLECTED DURING DIGESTION............. 107

TAB. 4.13: INHIBITORS IN ANAEROBIC DIGESTION AND VALUES REACHED IN THE CONTINUOUS TRIALS. .......... 115

TAB. 4.14: BIOMASS CHARACTERIZATION OF THE SAMPLES COLLECTED DURING CONTINUOUS DIGESTION. ..... 116

vii

Symbols

Greek symbols

Symbol

Description

Unit

λ1

Lag-phase

λ2

Switch to the second phase of the anaerobic di-

gestion

µm

Maximum biogas production rate

L kg-1 d-1

Constants

Symbol

Description

Unit

Euler’s number

Latin symbols

Symbol

Description

Unit

Biogas formation potential

L kg-1

COD

Chemical oxygen demand

g L-1

Coll0

Onset collagen content

Collf

Final collagen content

Degradation by trypsin or collagenase

Degradation degree of collagen

Symbols

viii

Organic dry matter of the inoculum in the reac-

tor

Onset mass

Final mass

msubstrate

Substrate addition

g d-1

P-value

Coefficient of determination

Organic dry matter of the substrate

Solubility in water

Time

h; d

VGas

Volume of biogas

mL d-1

xCH4

Amount of methane

Cumulative biogas production

L kg-1

List of abbreviations

Shavings autoclaved during 3 minutes

Shavings autoclaved during 6 minutes

A12

Shavings autoclaved during 12 minutes

A24

Shavings autoclaved during 24 minutes

A48

Shavings autoclaved during 48 minutes

A96

Shavings autoclaved during 96 minutes

A192

Shavings autoclaved during 192 minutes

A384

Shavings autoclaved during 384 minutes

ANOVA

Analysis of variance

CDU

Collagen digestion units

C/N

Carbon to nitrogen ratio

Chromium Shavings

DSC

Differential scanning calorimetry

E100D

Shavings extruded dry at 100 °C

E130D

Shavings extruded dry at 130 °C

E150D

Shavings extruded dry at 150 °C

E170D

Shavings extruded dry at 170 °C

E100W

Shavings extruded wet at 100 °C

E130W

Shavings extruded wet at 130 °C

E150W

Shavings extruded wet at 150 °C

E170W

Shavings extruded wet at 170 °C

EDX

Energy-dispersive X-ray spectroscopy

Extruded offcuts

Gly

Glycin

H140

Shavings treated hydrothermally at 140 °C

H150

Shavings treated hydrothermally at 150 °C

H170

Shavings treated hydrothermally at 170 °C

List of abbreviations

rpm

Revolutions per minute

SEM

Scanning electron microscopy

S/I

Substrate to inoculum ratio

TGM

Two-phase Gompertz model

VFA

Volatile fatty acid

1 Introduction

In 2014, the manufacture of leather and related products in the European Union

generated EUR 54 billion in turnover and employed 447.535 people (GROW,

2016) highlighting the importance of the leather industry in this region. The in-

dustry also plays an important role in terms of environment, since its main raw

material (hides) is a by-product of the meat industry, which otherwise would

be transformed into gelatin or disposed of at landfills as waste. Yet it must be

emphasized that hides were completely biodegradable at the slaughterhouse

but the leather-making process decreases the biodegradability of the material

(Flowers, 2017).

Between 2012 and 2014 an average of 558.4 thousand tons of leather from bovine

hides were produced worldwide (FAO, 2016). Europe alone is responsible for

the largest part of the global leather production, about 25% (Dhayalan et al.,

2007). Buljan et al. (2000) estimated the amount of solid waste produced while

processing one ton of wet salted hides (Table 1.1).

Regarding the different waste shown in Table 1.1, disposal of tanned waste gen-

erated after the tanning step is most complicated. As the chromium-based tan-

ning process is performed worldwide in about 90% of the cases (Covington,

2009), most of this waste will contain Cr3+ and need special handling. Currently,

the chromium-tanned leather waste is mainly disposed of through landfill or

incineration processes, despite the ecological consequences (Pati et al., 2014).

Increased environmental restrictions and escalating landfill costs have encour-

aged the leather industry to develop cleaner technologies by minimizing waste

and maximizing reuse (Mu et al., 2003). Attempts have been made to replace

the chromium salts used in the tanning process but the obtained chromium-free

leathers entail higher costs. Furthermore, chromium-based tanning is the most

1 Introduction

robust and cost-effective way to produce leather. It is possible to use vegetable

agents in the tannery process, but those tanning agents cannot be considered

more environmentally friendly than chrome tanning due to the high

wastewater load and low treatability in conventional systems (IULTCS, 2018b;

Trommer and Kellert, 1999). Glutaraldehyde is used as well, but the process is

more complicated to perform and economically disadvantageous for tanneries

(Trommer and Kellert, 1999). Therefore, the leather industry continues to face

the handling and disposal problems of chromium-containing waste.

Tab. 1.1: Quantity of solid waste produced in the leather-making process.

Solid waste

Quantity (kg/ton of wet

salted hide processing)

Untanned

Trimmings

100 kg

Fleshings

300 kg

Tanned

Unusable split

Trimmings

107 kg

20 kg

Shavings

99 kg

Dyed/finished waste

Fibres and trimmings

Buffing dust

7 kg

1 kg

Offcuts

5 kg

Total

639 kg

Source: Buljan et al. 2000.

The reutilization of chromium leather waste is an environmentally friendly al-

ternative that helps to reduce the amount of waste disposed of by converting it

into value-added products or creating the possibility of playing a new role in

energy generation or in the development of new raw materials.

The use of this waste as substrate to produce biogas through anaerobic diges-

tion is promising due to the reduction of the final amount of waste and simul-

taneously the generation of renewable energy, low level of process complexity,

and low cost. This process is called anaerobic digestion of organic matter the

1.1 Motivation

biological treatment of which is performed in the absence of oxygen while pro-

ducing biogas, a mixture of methane (55 – 70%) and carbon dioxide (30 – 45%)

with traces of other gases (Deublein and Steinhauser, 2008).

Hitherto, only a few studies on the digestion of chromium leather waste were

published because it is considered as complex waste (Dhayalan et al., 2007; Fer-

reira et al., 2010; Priebe et al., 2016; Agustini et al., 2015 and 2018). These authors

consider chromium leather waste as complex waste because of its collagen con-

tent and tanning with chromium salts. These studies demonstrate that it is pos-

sible to produce biogas from this waste but, due to very long periods of time

needed for digestion and low biogas yields, the method needs further research

to reach industrial feasibility. Furthermore, the final biomass produced in the

biogas reactor would contain chromium, which complicates its use as a ferti-

lizer. Extraction and recycling of chromium from the final biomass produced in

the reactors also needs further research.

The collagen molecule, the main component of leather, has a triple helical struc-

ture stabilized by hydrogen bonds and natural cross-links. The amount of hy-

drogen bonds in the structure defines the energy necessary to denature the tri-

ple helix because more hydrogen bonds need more energy to be broken down

and separate the chains (Miles and Ghelashvili, 1999). This complex structure

results in high resistance to heat and enzymatic degradation, and low solubility

in aqueous buffers. The denaturation temperature is different for each struc-

tural level of collagen, and to transform a collagen molecule into gelatin, a tem-

perature of about 38 °C is needed (Meyer, 2019). The denaturation temperature

to denature several collagen molecules kept together by natural cross-links

forming tissue is between 60 and 65 °C (Meyer, 2019). Tissue can be further sta-

bilized with chromium salts by chemical cross-links during the tanning step in

tanneries forming chromium-tanned leather (Dhayalan et al., 2007). The chro-

mium leather has a denaturation temperature of 105 to 110 °C (Schroepfer and

Meyer, 2017). Furthermore, chromium present in chromium-tanned leather

could be toxic for anaerobic bacteria (Deublein and Steinhauser, 2008).

Currently, there are no biogas plants in the industry using chromium leather

waste as a main substrate due to these difficulties. However, the tannery

1 Introduction

SÜDLEDER (Rehau, Germany) already has a biogas plant in operation using

their own organic waste (hair, protein, fat, and chromium-loaded sludge) to

produce energy (Schuberth-Roth, 2013). This kind of initiative illustrates the in-

terest of the industry in biogas production, nevertheless the use of a substrate

as complex as chromium leather waste needs to be further developed.

Irrespective of all difficulties, leather is composed of organic matter, hence it

can be used as raw material to produce biogas. This approach would become

more feasible with the development of a simple and efficient method to reverse

the effects of tanning (Covington, 2009).

Several different pre-treatments can be used to denature the collagen present in

chromium leather waste and enable microorganisms to degrade it in order to

reduce the digestion time and increase the biogas yield. The parameters of

choice for the most appropriated pre-treatment are time, cost, and complexity

of the process. As the collagen molecule itself has high stability and the chro-

mium tanning process makes it more stable (Usha and Ramasami, 2000), colla-

gen molecules are endowed with mechanical and thermal stability of the fibrous

network and high stability to enzymatic degradation. Consequently, it is neces-

sary to denature the collagen fibres to enable the enzymes to degrade this solid

waste (Kanagaraj et al., 2006).

1.1 Motivation

The production of large amounts of chromium leather waste, such as chromium

leather shavings and leather offcuts, cannot be avoided. Therefore it is neces-

sary to reuse the waste generated. Using the waste for biogas production ap-

pears to be an attractive alternative but pre-treatment is necessary to denature

this stable waste.

The aim of this study is to accelerate and improve the biogas production process

through anaerobic digestion of chromium leather waste enabling its application

in the industry. For this purpose, the chromium leather waste underwent pre-

treatment using different heating and mechanical technologies, which could be

1.1 Motivation

easily adjusted to industrial scale. The modifications caused in the collagen

structure of this waste were evaluated using different fast in vitro methods. Fi-

nally, the biogas formation potential was investigated through biogas produc-

tion trials on laboratory (batch) and pilot scale (continuous) to prove the feasi-

bility of the considered pre-treatments.

2 State of the Art

2.1 Chromium leather waste

Chromium leather waste is collagen-based waste generated after the tanning

step with chromium salts in the leather-making process. In this study, we use

two kinds of chromium leather waste, chromium shavings and automotive

leather offcuts. These materials were selected because both are generated in

large amounts and are difficult to reuse.

Chromium shavings are generated after the tanning step. Hides are soaked, un-

haired, limed, delimed, bated, and pickled. These steps are carried out to purify

the hide and open up its structure in preparation for the tanning step with bas-

ified trivalent chromium salts (Covington, 2009). After tanning, tanned leather

has to be shaved to adjust its thickness generating the chromium shavings

(Heidemann, 1993). These are wool-like waste with 40 to 50% of water content.

This waste mainly consists of the collagen of hide with complex bond Cr3+ salts

and some residues from the tanning process (fat, salts, or mineral compounds).

Leather offcuts are finished leather leftovers that underwent more steps in the

leather-making process than the chromium shavings. In addition to the tanning

step, this waste was retanned, dyed, fatliquored, dried, and finished with poly-

mer coatings (Heidemann, 1993). Those are necessary steps to produce the final

product leather. Leather offcuts are low water content waste resulting from the

trimming or cutting of a large piece of leather for upholstery. Generally, this

waste is generated in the automotive industry, in furniture production, or foot-

wear industry. Besides collagen and chromium, this waste contains numerous

of other chemicals. Retanning agents (basified trivalent chromium, vegetable

tannage, and/or other tannages), dyes, fat, and pigments.

2 State of the Art

2.1.1 Collagen structure and stability

It is important to understand the structure of collagen in order to understand

chromium leather waste stability. Collagen is among the most common fibrous

proteins and it is present in tendons, ligaments, bones, dentin, skin, arteries,

cartilage, and in most of the extracellular matrix in general (Fratzl, 2008). There

are at least 27 known different types of collagen in vertebrates and inverte-

brates, type I collagen being the most common of them in skin, tendon and bone

(Birk und Bruckner, 2005). These molecules are assembled in different fibrous

structures with quite different properties, such as elastic skin, soft cartilage, and

stiff bone and tendon (Fratzl, 2008).

What different types of collagen molecules have in common is their occurrence

in the extracellular matrix, their hydroxyproline content (amino acid specific to

collagen), and that they are composed of three polypeptide chains, which form

a triple helix arrangement in the core of the structure (Reich, 2007). Each of the

chains, called α chain, are characterized by the repeating amino acid motif (Gly-

X-Y), where glycine must always be located in the third position, and X and Y

can be any amino acid (Hulmes, 2008). The combination of α chains defines the

type of collagen. For type I collagen, this motif is repeated approximately 350

times, X is frequently proline, and hydroxyproline is always located in position

Y (Reich, 2007).

This typical sequence leads to a triple helical structure of the collagen molecule.

Hydrogen bonds and natural cross-links stabilize this arrangement resulting in

high resistance to heat and enzymatic degradation, and low solubility in aque-

ous buffers (Meyer, 2019).

Several collagen molecules are kept together by the development of molecular

cross-links between them forming the fibrils and, subsequently, different kinds

of tissues. In particular, the fibrils found in the skin are rich in type I collagen

but also contain a significant amount of collagen type III, typically about 20%

(Wess, 2008). The stability of the material is therefore increased motivating its

use in the leather industry.

2.1 Chromium leather waste

2.1.2 Chromium-based tanning process

The leather-making industry needs to further stabilize collagen fibres by chem-

ical cross-linking despite their natural collagen resistance to heat and enzymatic

degradation. To transform hides into leather, the dermis has to be separated

from the two other layers of the hide, the epidermis (along with the hair) and

the hypodermis. Subsequently, hides are washed to complete the separation of

non-collagenous components and the structure of the material is opened up in

preparation for the tanning step, in which tanning agents need to penetrate the

structure (Reich, 2007). The tanning agent (usually basic chromium sulfate) re-

acts with the collagen matrix, chromium complexes form complex bonds with

the carboxyl groups in the structure. The process stabilizes the collagen matrix

resulting in leather (Dhayalan et al., 2007).

These cross-links formed with the tanning agents first of all have to prevent the

collagen structure from collapsing during drying. They separate the molecules

which would otherwise glue together resulting in stiff parchment (Reich, 2007).

Furthermore, cross-linking also enhances mechanical strength in wet state, in-

creases the denaturation temperature, reduces the swelling capacity, and low-

ers susceptibility to enzymatic degradation (Avery and Bailey, 2008). Thermal

stability is closely related to the intrafibrillar water content of the collagen,

simply because the cross-links, natural and chemical, reduce separation of the

molecules leading to dehydration of the fibres (Miles et al., 2005). The chemical

cross-linking process applied to hides is known as the core of the leather man-

ufacturing process.

The stabilization process has an impact on the denaturation temperature, the

more stable the material the higher the needed temperature to break down the

structure into gelatin or denatured collagen (Figure 2.1). Denaturation occurs

when this material is exposed to a temperature higher than the denaturation

temperature and triple helices are unlocked. This random structure loses the

former high stability and can be easily degraded (Figure 2.1a to 2.1d). For solu-

ble collagen molecules, which consist of single triple helices, the structure dis-

integrates into a gelatin with broad molecular weight distributions, and for tis-

sue the structure shrinks (Meyer, 2019).

2 State of the Art

Fig. 2.1: Different structural levels of collagen materials and their denaturation temperature

(TD). Triple helix of the collagen molecule (a), its assembly in a tissue (b), chemical cross-links

formation while tanning with chromium salts (c), and collagen denaturation to gelatin through

temperature exposure (d).

The denaturation temperature of fully hydrated skin collagen is known to be

60 °C - 65 °C (Figure 2.1b), and is increased through vegetable and aldehyde

tanning by up to 20 °C and through chrome tanning by up to 50 °C (Reich, 2007).

Therefore, the denaturation temperature of chromium leather waste reaches

105 °C to 110 °C (Figure 2.1c), which requires a denaturation process at these

temperatures (Schroepfer and Meyer, 2017).

2.1.3 Alternative management of chromium leather waste

It is possible to separate chromium leather waste into collagen protein and chro-

mium by specific treatment. The majority of the published data on recovery of

this waste concerns the alkaline hydrolysis of leather. In some cases, this pro-

cess can even lead to a zero discharge method. However, the procedure is com-

plex, time-consuming and expensive (Katsifas et al., 2004).

The literature suggests that chromium leather waste can be managed through

hydrolysis of the material with alkalis, namely, CaO, Ca(OH)2, MgO, or

Mg(OH)2, or NaOH and temperature- and/or pressure-assisted. Holloway

(a)

(b)

(c)

(d)

2.1 Chromium leather waste

(1978) and Mu et al. (2003) used chromium shavings or scraps and Ferreira et

al. (2014) used finished leather waste in their studies. Two products are ob-

tained in this kind of process: protein hydrolysate, which can be used as a

leather finishing agent or for animal feeding, and a chromium solution useful

in tannery operations. However, the surface grain of leather treated with those

chromium solutions may show a lower quality limiting its later use.

Microbial hydrolysis of chromium shavings is another option for minimum

generation of waste and recovery of chromium. Pillai and Archana (2012) stud-

ied the use of Bacillus subtilis P13 and Katsifas et al. (2004) an Aspergillus car-

bonarius isolate. Both authors agreed that the previous autoclaving of the waste

was essential for its degradation. Katsifas et al. (2004) were able to recover up

to 12% of chromium, the proteinaceous solution might be used for fertilizer,

feed additive or silage production. Pillai and Archana (2012) achieved a recov-

ery of chromium of up to 50%, which could be used in tannery procedures or

sold to chemical suppliers. Additionally, a valuable by-product was generated,

a dehairing protease useful in the pre-tanning steps. Although those methods

do not have large implementation costs, they are not very feasible since they

can be regarded as very time-consuming.

Chromium leather waste can also be hydrolyzed directly by enzymes in acidic

or alkaline conditions. Crispim and Mota (2003) hydrolyzed chromium leather

shavings with proteases in acid and alkaline media to prepare leather board and

a protein hydrolysate for leather finishing. The acidic protease (pepsin) was

used to hydrolyze the chromium shavings. The resulting solution was filtered

and the filtration cake was cross-linked with glutaraldehyde to prepare the

waste for leather board production with no significant chromium release. The

alkali protease (Rodazym ML) hydrolyzed the chromium shavings into a

smaller material and then separated the chromium from the protein hydroly-

sate. This protein hydrolysate is able to replace all or part of the casein formu-

lations currently applied in leather finishing. Unfortunately, the authors do not

present an economic study to prove the feasibility of these processes as there is

a substantial input of chemicals and energy demand.

2 State of the Art

Ferreira et al. (2010) leached finished chromium leather waste using sulfuric

acid. The chromium from the obtained acid chromium extracts might be pre-

cipitated as chromium hydroxide according to Almeida and Boaventura (1997).

The de-chromed material is three times more biodegradable than the original

waste. However, it still contains chromium and is not acceptable at hazardous

landfills since the total chromium released in the leaching test exceeds the

threshold value of 5 mg L-1. This treatment is nearly static and constitutes a low

cost and feasible alternative but nevertheless it needs three to six days to be

complete, a long time on industrial scale.

Cot et al. (2003) developed an oxidative process to recover chromium from chro-

mium leather waste in different processing stages, whereby Cr3+ is oxidized to

Cr6+ using peroxides in an alkaline medium. In this case, the treated de-chromed

collagen residue can be used as substitute of casein or as pre-tanning or re-

tanning resin and the recovered Cr3+ tanning agent resulting in extremely soft,

clear grain leather. However, this method has the disadvantage that Cr6+ is in-

volved, which requires an additional reductive step.

Regular adsorbents for wastewater treatment are expensive for the industry

causing a growing demand of alternative adsorbents. Chromium leather shav-

ings can be considered an alternative adsorbent and be used with or without

any pre-treatment. Most authors concentrate their efforts on the adsorption of

surfactants or dyes. Regarding the adsorption of surfactants, Zhang et al. (2006)

tested three different kinds – anionic, cationic, and non-ionic. Whereas adsorp-

tion of the cationic and non-ionic surfactants was limited, the chromium shav-

ings showed a high adsorption capacity of anionic surfactants. The cationic sur-

face of chromium leather shavings attracts negatively charged molecules en-

hancing the probability of adsorption to occur. For the same reason, most of the

papers involving dyes study the adsorption of anionic dyes (Zhang and Shi,

2004; Piccin et al., 2012 and 2013; Gomes et al., 2015). Oliveira et al. (2007) ex-

tracted chromium in advance and tested the pre-treated and the untreated

leather for comparison. Results suggest that pre-treatment diminishes the effec-

tiveness of the adsorption process. The disadvantage of this method is that there

is no decrease in the amounts of this waste, much less in its toxicity. Contact

with the waste may also result in the contamination of wastewater with Cr3+.

2.2 Anaerobic digestion

As an alternative, chromium leather shavings can be used as fertilizer additive,

in order to completely eliminate the waste. This is possible with (Lima et al.,

2010) or without (Daudt et al., 2007) a prior treatment to remove chromium,

whereas the former option entails high chemical treatment costs, and the latter

will lead to soil contamination, which is restricted in many countries such as

Germany, Brazil and the USA. Another way to achieve the zero solid waste tar-

get is to utilize chromium leather shavings as a protein source for poultry feed

(Paul et al., 2013). For this purpose, waste underwent treatment, in order to re-

move chromium followed by a thermal and enzymatic treatment to produce

gelatin solution. The product resulting thereof achieved the nutrient require-

ments for poultry feed. Some companies produce leather boards from chro-

mium leather shavings in several countries but not all shavings satisfy the strict

quality requirements for the process and therefore the elimination of waste is

very restricted (IULTCS, 2018a).

All these methods represented a serious disadvantage regarding cost, time, or

reutilization of the obtained products. Consequently, a more feasible method

has to be investigated.

2.2 Anaerobic digestion

The anaerobic digestion is a series of microbiological processes intended to

break down organic matter in the absence of oxygen. While organic matter is

degraded, and therefore reduced, biogas is generated as the main product of

these reactions. Biogas in turn can be used to generate energy or heat.

According to Deublein and Steinhauser (2008), benefits of the installation of a

biogas plant include using biomass, which would normally be left to natural

deterioration, to generate high-energy compounds and valuable fertilizers. The

organic waste is also reduced to 4% after squeezing resulting in a substantial

decrease of landfill area and disposal costs. Finally, biogas plants contribute to

a yield increase with regard to agricultural activities on a large and small scale

as biogas production is relatively simple to put into operation and subsidized

2 State of the Art

in many countries. Furthermore, the release of methane from organic waste

degradation is avoided by controlled anaerobic digestion resulting in a signifi-

cant reduction of greenhouse gas emissions (Marin et al., 2010).

Mata-Alvarez et al. (2014) examined the papers about anaerobic digestion pub-

lished between 2010 and 2013 and concluded that the most frequent main sub-

strates studied are animal manures (54%), sewage sludge (22%) and the organic

fraction of municipal solid waste (11%). At the same time, the most commonly

used co-substrates are industrial waste (41%), agricultural waste (23%) and mu-

nicipal waste (20%).

2.2.1 Anaerobic digestion of tannery waste

Most biogas production papers analyzing the digestion of tannery waste focus

their efforts on the digestion of untanned fleshings, sludge from wastewater

treatment, wastewater, or the co-digestion of fleshings along with sludge or

wastewater.

Shanmugam and Horan (2009) studied the anaerobic digestion of limed flesh-

ings but a co-digestion with municipal solid waste was necessary to enhance

the biogas yield due to the low C/N (Carbon/Nitrogen) ratio and alkaline pH.

Kameswari et al. (2014a) tested different pre-treatments (ozonation, alkaline

thermal treatment, and sonication) to degrade sludge from wastewater treat-

ment and accelerate the rate of hydrolysis during the anaerobic digestion. They

found that ozonation showed a higher increase in soluble chemical oxygen de-

mand (COD) in the sludge. Banu and Kaliappan (2007) carried out a successful

experiment using a hybrid upflow anaerobic sludge blanket reactor to treat veg-

etable tannery wastewater and achieving a COD removal of 88%. Thangamani

et al. (2010) studied the co-digestion of fleshings and primary sludge from a

tannery wastewater treatment plant in batch experiments and concluded that

both contain a significant quantity of organic matter amenable to biodegrada-

tion. Thangamani et al. (2015) verified that the co-digestion of fleshings and ef-

fluent treatment of liquid waste from a tannery was important to minimize am-

monia toxicity in a two-phase digester. Kameswari et al. (2011) studied the co-

2.2 Anaerobic digestion

digestion of fleshings along with pre-treated primary and secondary sludge

(ozonation and ultrasonication). They were able to enhance biogas production

by 53% using sludge pre-treated by ultrasonication. Kameswari et al. (2012)

studied the substrate to inoculum ratio (S/I) in the co-digestion of fleshings

along with mixtures of primary and secondary sludge from treatment of tan-

nery wastewater. They concluded that a decrease of the S/I ratio beyond 1.0 has

no significant influence on biogas generation and leads to an increase of the

digester volume. Kameswari et al. (2014b) optimized proportions of fleshings,

primary sludge, and secondary sludge as substrates to be used for co-digestion,

which played a significant role in this process. A higher proportion of primary

sludge enhanced the biogas production. Kameswari et al. (2015) studied the

co-digestion of fleshings with primary and secondary sludge in semi-continu-

ous mode reactors to evaluate the effect of multiple feeds on the co-digestion

process. The organic load with the highest biogas production also reached the

largest percentage reduction for volatile fatty acids. Only a small concentration

within the range of 1.6 to 2 g L-1 can be found in the primary sludge. Ravin-

dranath et al. (2015) studied the co-digestion of fleshings and tannery effluent

together in an upflow anaerobic sludge blanket reactor and showed that it is

possible to obtain an additional methane yield of 37.5% when using co-diges-

tion instead of digesting the wastewater alone.

2.2.1.1 Anaerobic digestion of chromium leather waste

Dhayalan et al. (2007) and Ferreira et al. (2010) studied the anaerobic digestion

of chromium leather waste. The former concluded that degradation of this

waste is possible using anaerobic sludge, which however is a very slow process

and leads to low biogas amounts. The latter found that the results are depend-

ent on the anaerobic sludge concentration and origin. Agustini et al. (2015) also

studied the degradation of chromium leather shavings and detected 55% of me-

thane in the produced biogas. Priebe et al. (2016) found a low performance of

the chromium shavings to generate biogas and associated this with the satura-

tion of the reactive sites due to chromium bonding and low water solubility.

More recently, Agustini et al. (2018) concluded that a scale-up of the anaerobic

digestion of chromium shavings (300 mL to 2.5 L digesters) can increase biogas

yields and destruction of organic matter. In all cases, the experiments were

2 State of the Art

batch tests which lasted one to four months, and in most of the studies more

than ten days were needed to start the biogas production. For instance, Agustini

et al. (2015), in studies concerning the anaerobic digestion of chromium shav-

ings, needed more than twenty days to start a significant production of biogas.

2.2.1.2 Destabilization of chromium leather waste

As previously mentioned, the tanning process increases the stability of collagen

transforming it into leather which is resistant to enzymatic degradation. It is

well known that enzymatic attack and decay of leather in soil is still possible,

but only very slowly (Reich, 2007).

Anaerobic digestion is based on enzymatic degradation. Denaturation of the

leather structure is a possibility to ease the enzymatic attack and enable the pro-

cess. The prior denaturation of the structure can be accomplished increasing the

temperature to the denaturation temperature or higher (105 °C – 110 °C).

Most of the research concerning the pre-treatment of substrates for biogas pro-

duction addressed mechanical pre-treatments (33%), such as ultrasound and

extrusion, followed by thermal pre-treatments (24%), such as steam explosion

and autoclave and chemical pre-treatments (21%) (Mata-Alvarez et al., 2014).

Mechanical pre-treatments are able to reduce particle size and, consequently,

increase the specific surface available to the medium, which could improve gas

production and lead to more rapid digestion (Mata-Alvarez et al., 2000). Fur-

thermore, pre-treatments help to breakdown complex polymers into smaller

molecules promoting hydrolysis (Penaud et al., 1999), the stage responsible for

limiting the rate of degradation. The main goal is to render a more biodegrada-

ble substrate to the digestion process thus increasing methane production

(Mata-Alvarez et al., 2014).

Some studies were published on the pre-treatment of chromium leather waste.

As mentioned before, Pillai and Archana (2012) and Katsifas et al. (2004) con-

cluded that the previous autoclaving of chromium leather shavings was essen-

tial for its degradation by microorganisms. Therefore, a pre-treatment prior to

2.2 Anaerobic digestion

anaerobic digestion can denature the chromium leather waste and start hydrol-

ysis of the material in order to ease anaerobic digestion and improve the biogas

production.

2.2.2 Microbiology: trophic groups of the anaerobic digestion

The production of biogas is accomplished through anaerobic digestion of or-

ganic matter. This is a quite complex microbial process that occurs in the ab-

sence of oxygen with many types of strict and facultative anaerobic bacteria

(Murphy and Thamsiriroj, 2013; Deublein and Steinhauser, 2008). This process

can be divided into four phases – hydrolysis, fermentation or acidogenesis,

acetogenesis, and methanogenesis, according to the decomposition process of

the substrates and the bacteria acting to degrade them (Figure 2.2).

Fig. 2.2: Phases of anaerobic digestion (1) hydrolysis, (2) fermentation or acidogenesis, (3) aceto-

genesis, and (4) methanogenesis (adapted from Murphy and Thamsiriroj, 2013).

1 - Hydrolysis: Exoenzymes (hydrolase) of hydrolytic bacteria break down pol-

ymers (undissolved compounds) into monomers (water-soluble fragments), the

covalent bonds are split in a chemical reaction with water (Figure 2.3).

Complex organic polymers

(proteins, polysaccharides)

Monomers and oligomers

(sugars, amino acids, peptides)

Propionate, butyrate, VFAs,

long chain fatty acids

Acetate

H2+ CO2

CH4+ CO2

2 State of the Art

Fig. 2.3: Formation of monomers (adapted from Deublein and Steinhauser, 2008).

If the substrate is highly complex and therefore difficult to degrade, the hydro-

lytic stage limits the rate of degradation. The hydrolysis time varies with the

substrate, in the case of proteins, such as collagen, it is a matter of days (Deu-

blein and Steinhauser, 2008).

2 - Fermentation or acidogenesis: The fermentative bacteria ferment the resultant

monomers from the hydrolysis into acetic acid, hydrogen, carbon dioxide and

volatile fatty acids such as propionate, butyrate and alcohols. The concentration

of the hydrogen formed intermediately affects the kind of products developing

in this phase. The higher the partial pressure of hydrogen, the fewer acetate is

formed (Deublein and Steinhauser, 2008).

Amino acids, products of the hydrolysis of proteins, are mainly degraded

through the Stickland Reaction (Schink and Stams, 2013; Ramsay and Pul-

lammanappallil, 2001). This reaction is based on the coupled deamination be-

tween two amino acids acting as donor and acceptor of hydrogen. The products

of digestion are ammonia, carbon dioxide, and volatile fatty acids, which vary

from the coupled amino acids being digested and the present bacteria in the

medium (Nisman, 1954). Equation 2.1 shows the example of the coupled deam-

ination of alanine and glycine.

CH3CHNH2COOH+2NH2CH2COOH+2H20 → 3CH3COOH+3NH3+CO2

(2.1)

Alanine Glycine Acetic Acid

The main proteolytic bacteria in sludge are gram-positive bacteria, principally

Clostridia. The majority of these bacteria are known to degrade amino acids

through the Stickland Reaction (McInerney, 1988). This reaction is the main way

2.2 Anaerobic digestion

these bacteria obtain energy to grow when amino acids are the only source of

carbon and nitrogen (Nisman, 1954).

3 - Acetogenesis: Acetogenic bacteria convert the fermentative intermediates

(volatile fatty acids) into methanogenic substrates, hydrogen, carbon dioxide,

acetic acids and unicarbon compounds. Even though these bacteria are obliga-

tory H2 producers, this product is toxic to them. They must act in symbiosis

with bacteria in a different trophic group (methanogenic organisms) as they can

only survive and grow if the hydrogen partial pressure is kept at a very low

level. If the hydrogen partial pressure is low, the acetogenic bacteria will mainly

produce H2, CO2, and acetate, otherwise the main products will be butyric,

capronic, propionic, and valeric acids and ethanol. From these products, only

the three named first can be processed by methanogenic archaea.

The homoacetogenic bacteria reduce H2 and CO2 to acetic acid (Equation 2.2).

Although these bacteria (hydrogen-consuming acetogens) are not able to com-

pete with methanogens for hydrogen, they help to maintain low hydrogen par-

tial pressures and increase the amount of acetate available for the methanogen-

esis (Deublein and Steinhauser, 2008).

2CO2 + 4H2 ↔ CH3COOH + 2H2O

(2.2)

4 - Methanogenesis: Methane is generated in two different ways under strictly

anaerobic conditions. About 30% of the whole methane is produced by hy-

drogenotrophic methanogenic archaea. In symbiosis with the acetogenic bacte-

ria, they utilize the H2 produced in the previous step and ensure very low hy-

drogen partial pressure through the reduction of CO2 by H2:

4H2 + CO2 → CH4 + 2H2O

(2.3)

Aceticlastic methanogenic archaea are responsible for about 70% of methane

produced using acetic acid as substrate (Equation 2.4). Unlike the hydrogen-

otrophic methanogens these archaea are relatively inefficient in acetate uptake

and have slow reproduction rates (Deublein and Steinhauser, 2008).

CH3COOH ↔ CH4 + CO2

(2.4)

2 State of the Art

2.2.3 Biogas production studies

Biogas production trials can be carried out discontinuously or continuously.

They should be performed in a gastight reactor to assure the absence of oxygen.

A substrate, composed mostly of organic matter, and an inoculum, preferable a

sludge with a diversified biocoenosis, are placed inside the reactor. After suffi-

cient time of contact, biogas is produced as a result of the interaction between

substrate and inoculum.

2.2.3.1 Batch trials

Laboratory batch trials are essential to investigate biogas production. It allows

multiple experiments to be run simultaneously testing numerous variables and

collecting large amounts of data in relatively short periods of time. The bench-

scale tests reduce the amount of materials required for the trials thus minimiz-

ing waste generation and costs. This type of research helps to determine feasi-

bility of the process, and to optimize it in order to improve waste treatment and

biogas production (Gamble et al., 2015). These trials enable evaluation of the

biogas yield and the biogas degradability of a material, the speed of anaerobic

degradation, and its inhibitory effect within the range of tested concentrations

(VDI 4630, 2006).

In batch tests, the reactor is filled all at once. The substrate to inoculum mixture

remains in the closed reactor until the end of the pre-specified digestion time.

At the end of digestion, the resulting biomass is removed for further analyses

(Gamble et al., 2015).

2.2.3.2 Continuous trials

Continuous tests simulate long-term process conditions, which allow to inves-

tigate capabilities and loading limits of the process, mean residence time, and

formation and accumulation of metabolic intermediates and their influence on

process stability. The apparatus is larger and necessary supervision higher for

continuous trials than for batch tests and several measurements of a large num-

ber of parameters in the gas and liquid phases (VDI 4630, 2006) are required.

2.2 Anaerobic digestion

In continuous tests, the substrate to inoculum mixture is added to the digester

at pre-designated times. Most large-scale industrial digesters operate in contin-

uous mode as it allows the digester to continually produce biogas (Gamble et

al., 2015).

2.2.4 Inhibitors

2.2.4.1 Chromium content

Heavy metals such as chromium are essential for bacterial growth in very small

quantities, but higher quantities have a toxic effect (Abbasi et al., 2012). In par-

ticular, lead, cadmium, copper, zinc, nickel, and chromium can lead to disturb-

ances in biogas plants (Deublein and Steinhauser, 2008).

Chromium acts as trace element at low concentrations stimulating the activity

of the bacteria. The minimum amount of trace element required is 0.005 –

50 mg L-1. The toxic effect starts at concentrations higher than 28 - 300 mg L-1 as

free ions or 530 mg L-1 as carbonate (Deublein and Steinhauser, 2008). Gayatri

et al. (2000) were able to demonstrate that the deactivation of collagenase by

unbound chromium complexes contained in chromium-tanned leather is pos-

sible. The direct binding of chromium complexes to the enzyme was proved.

Leather contains up to 1500 mg kg-1 of non-bonded Cr3+ complexes and it must

be assumed that other enzymes are also susceptible to this effect (Reich, 2007).

Consequently, anaerobic digestion can be inhibited.

2.2.4.2 Hydrogen sulfide (H2S)

The sulfate-degrading bacteria act to reduce sulfate forming hydrogen sulfide

following Equation 2.5. This is problematic for the methane formation because

these bacteria compete with the hydrogenotrophic methanogenic archaea for

hydrogen, and hydrogen sulfide is toxic to methanogenics (Polster and Brum-

mack, 2005). As sulfate-degrading bacteria need less energy and do not need a

symbiosis partner, this process disturbs the methane formation and causes

overacidification.

4H2 + 2H+ + SO42- → H2S + 4H2O

(2.5)

2 State of the Art

Hydrogen sulfide escapes with the biogas and is dissolved in undissociated and

dissociated form in the substrate as weak acid, developing hydrogen sulfide

ions (HS-) and sulfide ions (S2-). With decreasing pH, the dissolved undissoci-

ated hydrogen sulfide concentration rises and works as a cellular poison at con-

centrations higher than 50 mg L-1. Hydrogen sulfide can also cause inhibition

due to precipitation of essential trace metals as insoluble sulfides (Deublein and

Steinhauser, 2008).

2.2.4.3 Volatile fatty acids

The most commonly formed volatile fatty acids in biogas reactors are acetic

acid, propionic acid, and isobutyric acid. The sum of the concentration of all

volatile fatty acids formed should be lower than 4 g L-1 to prevent inhibition.

The concentration of acetic acid alone should be lower than 3 g L-1, that of pro-

pionic acid lower than 1 g L-1, and that of isobutyric acid lower than 0.5 g L-1

(Kaiser et al., 2008). However, a propionic acid concentration of 0.3 g L-1 only is

sufficient to disturb anaerobic digestion (Deublein and Steinhauser, 2008).

The production of volatile fatty acids leads to a decrease of the pH value, even

though they are intermediate products for generating biogas. The growth of

methanogenic archaea is inhibited if the pH value is below 6.5, but the acido-

genic bacteria continue to work until the pH value drops to 4.5. Consequently,

there is a fast accumulation of volatile fatty acids (Murphy and Thamsiriroj,

2013). The drop in pH value is buffered by formation of alkalinity through CO2

production. For this reason, using the pH value as an indicator for the stability

of anaerobic digesters is not reliable (Kaiser et al., 2008). The volatile fatty acids

concentration should be controlled.

2.2.4.4 Inhibition by diauxie

The biogas formation curves in batch trials are defined as the difference be-

tween the biogas production of the substrate less the biogas production of the

inoculum. They can also indicate if digestion works adequately. This qualitative

information enables the detection of inhibition in the reactor and that of difficult

substrates. Figure 2.4 shows the typical shapes of biogas formation curves.

2.2 Anaerobic digestion

Fig. 2.4: Typical shapes of biogas formation curves (adapted from VDI 4630, 2006).

If the substrates to be digested are easily convertible substances, they are con-

verted rapidly into biogas and the characteristic curve shows a steep increase

in the accumulated biogas quantity. For complex substrates, which are difficult

to degrade, biogas production is delayed and a retarded gas formation curve is

typical for that. A retarded gas formation curve can also be caused by slight

inhibition. If biogas production is strongly or completely inhibited, curves indi-

cate a negative net biogas production – in other words, gas formation is less

than that of the batch from the zero sample. In addition to the curve shapes

shown here, there is a large number of mixed forms (VDI 4630, 2006).

If degradation occurs in two stages (the curve resembles stairway steps), this

indicates a two-phase decomposition, also known as diauxie (VDI 4630, 2006).

The diauxie curve is characterized by a plateau-phase in the middle of the bio-

gas production. After that the system recovers, and substrates continue to be

transformed into biogas. This inhibition increases the time necessary for the

complete digestion of the substrate, which has a negative effect on the biogas

production.

2 State of the Art

2.2.5 Theoretical biogas yield: Buswell equation

The maximum biogas yield expected can be calculated using the Buswell equa-

tion after Boyle (1976) (Equation 2.6) if the basic elementary formula of the sub-

strate is known.

C𝑎H𝑏O𝑐N𝑑S𝑒+ (𝑎−𝑏

4−𝑐

2+3𝑑

4+𝑒

2)H2O →

(𝑎

2+𝑏

8−𝑐

4−3𝑑

8−𝑒

4)CH4+ (𝑎

2−𝑏

8+𝑐

4+3𝑑

8+𝑒

4)CO2

+𝑑NH3+ 𝑒H2S

(2.6)

It is important to bear in mind that this is a theoretical value, which will proba-

bly not be achieved because part of the substrate, 3 to 10%, is converted into

biomass and is not available for biogas formation. For proteins, a conversion

rate of 50 to 70% is expected (VDI 4630, 2006).

2.3 The use of chromium leather waste in the biogas pro-

duction

Hitherto the use of chromium leather waste to produce biogas has not been

studied in depth. Some studies proved that degradation of this waste through

anaerobic digestion is possible but very slow and leads to low biogas amounts.

The substrate is challenging due to the natural structure of collagen and chem-

ical cross-links acquired in the tanning process, which provide high stability to

anaerobic digestion. Using pre-treatments is a possibility to ease the hydrolysis

step in the anaerobic digestion and improve biogas production.

The present work will illustrate the use of different pre-treatments with regard

to chromium leather waste. This approach to produce biogas has not been stud-

ied to date. Different mechanical and thermal pre-treatment methods are con-

sidered and their degradation results are presented.

2.3 The use of chromium leather waste in the biogas production

Investigation of the biogas production using untreated and pre-treated chro-

mium leather waste as substrate was performed in batch and continuous reac-

tors. Whereas some studies exist for the treatment of chromium leather waste

under batch conditions, there is no paper published on continuous reactors for

this material.

Aim of the present investigation was to:

1. Characterize the pre-treated chromium leather waste to predict their per-

formance in the anaerobic digestion.

2. Prove the advantages of using pre-treated chromium leather waste instead

of the untreated substrate to produce biogas in batch and continuous reac-

tors.

3. Determine the causes of inhibition in the batch and continuous reactors.

3 Materials and Methods

Three different steps were carried out in this study. At first, chromium leather

waste was pre-treated to initiate material degradation. Three common thermal

and/or mechanical techniques are considered for this purpose – extrusion, hy-

drothermal treatment, and autoclave. Secondly, the pre-treated material was

assessed regarding its degree of degradation using different fast in vitro meth-

ods, e.g. enzymatic degradability and calorimetric analyses. Finally, the biogas

production of the waste was investigated through biogas production trials to

prove the feasibility of using chromium leather waste to produce biogas.

3.1 Materials

The collagen-based materials used as substrate to produce biogas were chro-

mium leather waste. Chromium shavings (shaved from wet chromium-tanned

leather) and automotive leather offcuts (trimmed from finished leather ready

for use in car manufacture) were tested for this purpose (Figure 3.1). The mate-

rials were obtained from a local tannery (HEWA Leder, Freiberg, Saxony, Ger-

many).

They were characterized regarding their water content (DIN EN ISO 4684,

2005), inorganic matter (DIN EN ISO 4047, 1998), and total chromium as chro-

mic oxide content (DIN EN ISO 5398-1, 2007) in per cent by mass. Experiments

were run in triplicate. The chromium leather waste tested is characterized in

Table 3.1. Usually, processed chromium shavings show water contents of 40 to

50% but those used in this work had already been air-dried to some extent and

have a water content of almost 20%.

3 Materials and Methods

Fig. 3.1: Chromium shavings (a) and leather offcuts (b).

Tab. 3.1: Chromium shavings and leather offcuts characterization.

Chromium shavings (CS)

Offcuts

Water content (%)*

19.7 ± 0.1

11.4 ± 0.0

Inorganic Matter (%)**

11.2 ± 0.1

6.4 ± 0.0

Chromium (%)**

4.6 ± 0.0

4.1 ± 0.1

* Mean ± standard deviation, n = 3

**Dry basis; mean ± standard deviation, n = 3

White hide powder supplied by the research institute FILK (Freiberg, Germany)

was used in chemical analyses. This material is bovine hide (pelt), which was

chemically unhaired and does not contain any trivalent chromium (Schroepfer

and Meyer, 2017). Bovine hide gelatin (260 g Bloom, type B) kindly supplied by

Gelita AG (Eberbach, Germany) was used as a reference in the biogas trials.

Gelatin is a hydrolysate of collagen with short chains. Chemically, gelatin is the

same as collagen only differing in the structure. Gelatin is known to be soluble

and easily degradable by enzymes.

For the biogas trials, two different inocula were used. For batch trials, the mes-

ophilic anaerobic inoculum was anaerobic sludge from the municipal sewage

treatment plant (Freiberg, Germany). A reference substrate (microcrystalline

cellulose) was also fermented in every batch to ensure that the seeding sludge

used has a proper biological activity. For continuous trials, mesophilic anaero-

3.2 Pre-treatments of the chromium leather waste

bic sludge from the tannery SÜDLEDER (Rehau, Germany) was used. This in-

oculum showed some quantities of chromium and collagen content (about 1%

chromium content and 3% collagen content on dry basis) because it was pro-

duced in a tannery.

3.2 Pre-treatments of the chromium leather waste

Different heat and mechanical pre-treatment techniques were tested in order to

denature the chromium leather waste and promote its degradation and trans-

formation into biogas. Extrusion, a classical technique from the polymer indus-

try; a continuous hydrothermal treatment, which is commonly used to plastify

wood for the manufacture of wood composites; and autoclave, a very com-

monly available instrument in laboratories, were used to pre-treat and denature

the chromium shavings and leather offcuts. Whereas during extrusion the ma-

terial is affected by heat, mechanical shear, and pressure, the autoclave and hy-

drothermal treatments are based on heat and steam pressure only. During the

process, temperatures higher than the denaturation temperature were achieved

in order to enable enzymatic degradation to produce biogas.

3.2.1 Extrusion

Extrusion was performed on chromium shavings and leather offcuts using a co-

rotating twin-screw-extruder Werner & Pfleiderer ZSK 25 (Figure 3.2) at differ-

ent temperatures and humidity conditions (dry or wet) in a continuous process.

Prior to that, the wet chromium shavings were moistened with water (60.1%

water content), well homogenized and left overnight. The dry chromium shav-

ings (19.7% water content) are air-dried chromium shavings. The leather offcuts

were extruded as received from the tannery (11.4% water content).

3 Materials and Methods

Fig. 3.2: Extruder Werner & Pfleiderer ZSK 25.

This extrusion process starts by feeding the sample from a hopper into the bar-

rel of the extruder. The material is gradually degraded due to the mechanical

energy generated by turning screws and by heaters arranged along the barrel.

The conversion of mechanical energy into heat makes it possible to use this pro-

cess even below the denaturation temperature of chromium-tanned leather

(105 °C to 110 °C). The process takes approximately three minutes (Figure 3.3).

Fig. 3.3: Operation of the extruder machine

The pre-treated shavings that originated from dry and wet chromium shavings

differ in appearance (Figure 3.4). Extrusion of dry chromium shavings resulted

in a powdered sample. In contrast, the wet chromium shavings resulted in sam-

ples with granular shape. A high water content opens up the structure of the

3.2 Pre-treatments of the chromium leather waste

sample increasing the space between the fibres. This eases extrusion and ena-

bles formation of small grains of leather. If leather having a low water content

is extruded, the fibres are close to one another. Consequently, when the leather

structure was broken using mechanical force, the pieces of leather became very

small and a powder was generated. A powder with some small leather pieces

can be seen for the extruded offcuts.

Fig. 3.4: Chromium shavings extruded (E) dry (D) or wet (W).

Tab. 3.2: Extruded samples and their pre-treatment conditions.

Sample

Pre-treatment Tem-

perature (°C)

Pre-treatment water

content (%)*

E100D

100

19.7 ± 0.13

E130D

130

E150D

150

E170D

170

E100W

100

60.1 ± 0.19

E130W

130

E150W

150

E170W

170

11.4 ± 0.0

*Mean ± standard deviation, n = 3

E100D

E130D

E150D

E170D

E100W

E130W

E150W

E170W

3 Materials and Methods

The letter E in the nomenclature of the extruded samples stands for extrusion,

the number is the extrusion temperature, and the letter D or W at the end stands

for dry or wet, respectively. EO stands for extruded offcuts. Table 3.2 lists the

samples nomenclature and pre-treatment conditions.

3.2.2 Hydrothermal treatment

The chromium shavings were subjected to hydrothermal treatment through a

continuous autoclave system attached to a refiner (Andritz CPH 12-1) at the

Institut für Holztechnologie (Dresden, Germany). Usually this equipment is

used to plastify wood chips but it is also adequate to process a variety of organic

materials. The process was carried out at different temperature and pressure

conditions in saturated steam (Figure 3.5).

Fig. 3.5: Continuous autoclave and refiner Andritz - front (a) and back view (b).

The temperature was adjusted regarding the saturated steam relative pressure,

however due to technical reasons the temperature was not as exact as expected.

The material was dosed into a digester in which it was denatured with steam

under pressure. Pre-treatment time was about 45 seconds.

3.2 Pre-treatments of the chromium leather waste

The shavings treated hydrothermally were also different in appearance. Those

that were pre-treated at 170 °C had the consistency of a liquid that solidified

into a gel-like mass after cooling. Chromium shavings pre-treated at 150 °C and

140 °C were very similar, they appeared more like a dough after cooling. Figure

3.6 shows the shavings treated hydrothermally after cooling. The letter H in the

nomenclature of the pre-treated sample stands for hydrothermal treatment and

the number is the treatment temperature. Table 3.3 lists the samples nomencla-

ture and pre-treatment conditions.

Fig. 3.6: Chromium shavings treated hydrothermally (H) after cooling.

Tab. 3.3: Samples treated hydrothermally and their pre-treatment conditions.

Sample

Pre-treatment

Temperature (°C)

Pre-treatment

Pressure (bar)

H140

141

2.2

H150

152

3.8

H170

169

3.2.3 Autoclave

Chromium shavings were autoclaved on laboratory scale. The trials were car-

ried out in a minimized system in order to reproduce the autoclaving conditions

(high temperature and pressure). Screw cap micro tubes tightly closed through

O-ring sealing and a block heater (Stuart SBH130D) at 120 °C (Figure 3.7) were

used. This apparatus allows for better control of the pre-treatment time and

temperature, which would not be possible using a laboratory autoclave. The

H140

H150

H170

3 Materials and Methods

laboratory autoclave would need to be sealed with the chromium shavings in-

side at room temperature, hence the material would be exposed to high tem-

peratures during preheating of the autoclave.

Fig. 3.7: Laboratory equipment used as autoclave.

Chromium shavings were moistened in advance with distilled water until the

saturation point was reached, and left overnight at room temperature. The sam-

ples were placed in the micro tubes, tightly closed, and subsequently placed in

the block heater at 120 °C. The micro tubes were preheated for 3 minutes and

30 seconds, the estimated time necessary for the samples to reach the autoclav-

ing temperature (120 °C). Each sample was exposed to the autoclaving condi-

tions for a predetermined time. Thereafter the samples were dried in a drying

oven at 30 °C for one day.

After drying, the autoclaved shavings were dry and brittle, and in case of longer

pre-treatment times gel-like areas began to emerge. These areas can be seen

from 96 minutes of pre-treatment onwards, and their frequency of appearance

increases based on the pre-treatment time (Figure 3.8). The letter A in the no-

menclature of the pre-treated sample stands for autoclaved and the number is

the pre-treatment time. The pre-treated samples nomenclature and pre-treat-

ment conditions are also shown in Table 3.4.

3.3 Chemical analyses

Fig. 3.8: Autoclaved (A) chromium shavings after drying.

Tab. 3.4: Autoclaved samples and their pre-treatment time.

Sample

Pre-treatment time (min)

A12

A24

A48

A96

A192

192

A384

384

3.3 Chemical analyses

3.3.1 Analyses of the pre-treated chromium leather waste

The pre-treated and untreated chromium leather waste was characterized and

evaluated by scanning electron microscopy (SEM), differential scanning calo-

rimetry (DSC), digestibility by trypsin and collagenase, and solubility in water.

A12

A24

A48

A96

A192

3 Materials and Methods

3.3.1.1 Samples characterization

The pre-treated and untreated samples were characterized regarding their wa-

ter content (DIN EN ISO 4684, 2005), inorganic matter (DIN EN ISO 4047, 1998),

and chromic oxide content (DIN EN ISO 5398-1, 2007) in per cent by mass.

The collagen content of the pre-treated and untreated chromium leather waste

was calculated through determination of the hydroxyproline content

(Stegemann, 1958) and multiplying this value by the factor 7.46 (100 g of protein

contain 13.45 g of hydroxyproline) (Reich, 1966). Samples were hydrolyzed

with hydrochloric acid, oxidized with chloramine T, and then reacted with p-di-

methylamino-benzaldehyde to develop a red chromophore. For gelatin, no col-

lagen chains can be found in the structure but it is possible to detect amino ac-

ids. For this reason, only the hydroxyproline content was determined.

3.3.1.2 Scanning electron microscopy (SEM)

The surfaces and fibres of the pre-treated and untreated chromium leather

waste were investigated using SEM (FEI QUANTA FEG 250). The images ob-

tained using this method were compared visually to evaluate the effect of the

different pre-treatments on chromium leather waste. Energy-dispersive X-ray

spectroscopy (EDX) was used to identify the type of elements that exist in the

samples in case they affect the anaerobic digestion.

3.3.1.3 Differential scanning calorimetry (DSC)

Differential scanning calorimetry (DSC) is a useful technique to study the ther-

mal stability of collagen-based materials (Miles et al., 2005). Using DSC, the en-

thalpy of the denaturation process was evaluated and it was checked whether

the pre-treated chromium leather waste has already been denatured or not. In

DSC trials, the difference between the amount of heat required to increase the

temperature of the studied material and a reference is measured as a function

of temperature. As seen in Figure 3.9, the onset temperature is the denaturation

temperature of the studied material and the area below the denaturation peak

is the heat of denaturation used to calculate the enthalpy of denaturation.

3.3 Chemical analyses

Fig. 3.9: Example of a DSC curve.

For collagen-based materials, the enthalpy represents the necessary energy to

break down the hydrogen bonds that stabilize the triple helix (Miles and

Ghelashvili, 1999). Thermal profiles in fully hydrated state of the pre-treated

and untreated chromium leather waste were taken at temperatures between 0

and 130 °C using DSC (DSC 1 STARe System Mettler Toledo) to assess thermal

changes as a function of input temperature. As the denaturation temperature of

collagen varies with the pH value (Schröpfer, 2012), the pH value was previ-

ously adjusted to 7 by washing of the samples with a KH2PO4/K2HPO4 buffer

solution.

3.3.1.4 Enzyme assays

Susceptibility of the pre-treated and untreated chromium leather waste to en-

zymatic degradation was evaluated with enzyme assays. Enzymes break down

peptide bonds in the backbone of the structure, a process similar to the actual

hydrolysis in the anaerobic digestion. The enzyme assays are based on the

measurement of degradation of the samples. As the non-denatured collagen is

stable to enzymatic degradation, it is possible to determine the fraction of the

samples which was degraded during pre-treatment.

Temperature (°C)

Heat flow (mW)

T onset

Heat of

denaturation

3 Materials and Methods

Trypsin trials were carried out using safe-lock microcentrifuge tubes (2 mL) and

a block heater (Stuart SBH130D). At first, samples were placed in the microtubes

with a NH4HCO3 buffer solution adjusted to pH 8 and left overnight in a labor-

atory refrigerator. Then the 0.01% trypsin solution (SIGMA, 1382 U mL-1) was

added at a temperature of 37 °C over a period of five hours and, finally, washed

out with distilled water. Experiments were run in triplicate and a t-test was con-

ducted to compare means.

Collagenase trials were carried out using chromium shavings and shavings ex-

truded dry at 100 °C. Unlike other proteases, collagenases are capable of desta-

bilizing the triple helical collagen (Meyer, 2019). Trials were carried out using

safe-lock microcentrifuge tubes (2 mL) and a thermomixer (Thermomixer com-

fort, Eppendorf) with agitation at 750 rpm. At first, samples were placed in the

microtubes with a NH4HCO3 buffer solution adjusted to a pH value of 7.5. A

0.05 M solution of CaCl2 was added to provide Calcium ions required for en-

zyme stability and activity. Then the collagenase solution (SIGMA,

≥ 125 CDU mg-1), in a concentration of at least 0.2 CDU per mg of substrate, was

added. The trials were conducted at a temperature of 37 °C over different peri-

ods of time of up to seven days. Every two days collagenase solution was added

to the microtubes. Finally, the samples were washed out with distilled water.

Experiments were run in duplicate.

The degradation (D) by trypsin or collagenase is the portion of the sample that

solubilizes in water after enzymatic treatment and it is represented by Equation

3.1:

D=100−(mf

m0.100)%

(3.1)

Where m0 (mg) is the onset mass of the sample and mf (mg) is the mass after the

reaction time, both masses were considered on a dry basis.

3.3.1.5 Solubility in water

It is known that for identical amino acid compositions, the water solubility of a

protein is correlated with the chain length and number of cross-links and can

therefore be used to measure the degree of degradation (Klüver and Meyer,

3.3 Chemical analyses

2013). In this experiment the parameters expected inside of an anaerobic reactor

such as temperature and pH value are reproduced in order to evaluate the por-

tion of the sample which is promptly soluble.

The solubility of the pre-treated and untreated chromium leather waste in water

was evaluated at a temperature of 37 °C. Approximately 15 mg of sample on

dry basis was placed in safe-lock microcentrifuge tubes (2 mL) with 1.5 mL of

water and its pH value was adjusted between 7 and 8 (pH meter Mettler Toledo)

with a NH4HCO3 solution. The samples were stirred in a thermomixer (Eppen-

dorf) at 300 rpm at the desired temperature for 2 hours. Afterwards the samples

were centrifuged, the supernatant was discarded, the centrifugate dried, and

the dry mass determined. Experiments were run in triplicate and a t-test was

conducted to compare means. The solubility in water (SW) is represented by

Equation 3.2:

SW=100−(mf

m0.100)%

(3.2)

Where m0 (mg) is the onset mass of the sample and mf (mg) is the mass after the

treatment, both masses were considered on a dry basis.

3.3.2 Analyses of the biomass

3.3.2.1 Degradation degree of collagen

Since the pre-treated and untreated chromium leather waste used as substrate

in the anaerobic digestion is collagen-based material, the collagen content of the

substrates and in the final biomass was measured (after anaerobic digestion) in

order to calculate the degradation of the substrate during digestion.

The degradation of the collagen at the end of the trials, the degradation degree

(DD), was calculated as represented by Equation 3.3:

DD=100−(Collf

Coll0.100)%

(3.3)

Where Coll0 (mg) is the collagen content at the beginning of the biogas trials

(substrate + sludge) and Collf (mg) is the collagen content measured in the final

3 Materials and Methods

biomass, both were considered on a dry basis. The collagen content of the sub-

strates and biomass were calculated through determination of the hydroxypro-

line content (Stegemann, 1958).

Gelatin does not have collagen chains in its structure but it is possible to detect

amino acids. For this reason, the hydroxyproline of the substrate and final bio-

mass were analyzed and the degradation of gelatin was measured as degrada-

tion of hydroxyproline.

3.3.2.2 Spectroquant cell tests

Spectroquant® Cell Test Kits for quantitative investigation were used to analyse

the biomass of the bioreactors photometrically. Ammonium content (Number

1.14739.0001), Chemical Oxygen Demand (Number 1.14541.0001), Iron content

(1.14549.0001), Sulfate content (Number 1.14548.0001), and Volatile fatty acids

(Number 1.01749.0001) were measured from the biomass samples.

3.3.2.3 High-performance liquid chromatography (HPLC)

Volatile fatty acids (acetic acid, propionic acid, isobutyric acid, butyric acid, and

isovaleric acid) were determined by HPLC (Shimadzu prominence Serie 20,

equipped with a refractive index detector RID-10A and a photodiode array de-

tector SPD-M20A). Measurements were performed with a mobile phase of

5 mM H2SO4, a flow rate of 0.6 mL min-1, a column temperature of 60 °C and a

detector temperature of 40 °C. The biomass samples were previously centri-

fuged at 14,000 rpm for 10 minutes and filtered using a syringe filter holder (fil-

ter 0.2 m pore size) before injection.

3.4 Biogas Production

Biogas production trials were carried out discontinuously and continuously in

gastight reactors to assure the absence of oxygen. The substrate, pre-treated or

untreated chromium leather waste, and inoculum, sludge with a diversified bi-

ocoenosis, were placed inside the reactor. After a sufficient time of contact bio-

gas was produced.

3.4 Biogas Production

3.4.1 Laboratory scale: batch tests

In this study, laboratory scale batch tests were performed under mesophilic

conditions (37 °C ± 2 °C) according to the guideline VDI 4630 (2006) in duplicate

or triplicate along with three blanks. The volumes of biogas produced by the

blanks (mean of three samples of inoculum without any added substrate) were

subtracted from the values obtained for the individual test samples. A t-test was

conducted to compare means. As inoculum, mesophilic anaerobic sludge from

the local sewage treatment plant was used.

The tests were conducted using glass flasks tightly sealed with a rubber septum

(65 mL) and the gas production was measured indirectly on a daily basis using

a digital manometer (Leo 3 Keller) (Figure 3.10). The gas volume was calculated

from the gas pressure registered and the gas temperature measured. The biogas

yield (quantity of generated biogas per quantity of substrate fed) and the biogas

formation potential (maximum biogas yield generated from a defined quantity

of substrate) are given in norm litres (273 K and 1013 hPa) per kg of organic dry

matter of the added substrate (L kg-1). The frequency of measurements was re-

duced to once every two or three days after a fall in the daily production. Meas-

uring devices for analysing the composition of low biogas amounts (percentage

methane and carbon dioxide) were not available.

Fig. 3.10: Biogas reactors in batch scale and biogas generation measurement.

3 Materials and Methods

The agitation effect was tested in two different systems, with agitation in a shak-

ing water bath (Julabo SW-20C at 150 rpm), and without agitation in a climatic

chamber (drying unit Fratelli Carlessi ARMADIO 5B). Volatile fatty acids were

injected into the bioreactors in some trials piercing the septum using a syringe.

Three different volatile fatty acids were evaluated, acetic acid (Carl Roth, 100%

p.a.), propionic acid (SIGMA-ALDRICH, 99.5% ACS), and isobutyric acid

(SIGMA-ALDRICH, 99.5% p.a.).

At the end of the process, the resulting biomass was analysed regarding its pH

value, water content (DIN EN ISO 4684, 2005), inorganic matter (DIN EN ISO

4047, 1998), chromic oxide content (DIN EN ISO 5398-1, 2007), and collagen con-

tent.

3.4.1.1 Two-phase Gompertz model

In many cases, the cumulative biogas production curve can be described as a

sigmoidal curve with lag, growth, and asymptotic phases. Therefore, it is pos-

sible to use the sigmoidal function of Gompertz, modified as described by

Zwietering et al. (1990), to fit the experimental biogas production data. If the

cumulative biogas production curve shows a two-phase biogas production, this

model needs to be modified to fit the two-phase curve as represented in Figure

3.11 and Equation 3.4.

3.4 Biogas Production

Fig. 3.11: Two-phase Gompertz model.

For the two-phase Gompertz model, the duration of the lag-phase (λ1, d) at the

beginning of the anaerobic digestion, the switch to the second phase of the di-

gestion (λ2, d), the biogas formation of each phase (Ai, L kg-1), and the maximum

biogas production rate of each phase of the anaerobic digestion (µmi, L kg-1 d-1)

were found with Equation 3.4.

y=∑(Ai exp{−exp(µmie

Ai(λi−t)+1)})

i=1

(3.4)

Where y is the cumulative biogas production (L Kg-1), t is the time (days), and e

is the Euler’s number. The biogas formation potential of the substrate is the sum

of the biogas formation of both phases (A1 + A2). The models were fitted using

the software SigmaPlot version 13.0.

3.4.1.2 Substrate to inoculum ratio

According to the guideline VDI 4630 (2006), the substrate should not be over-

large in proportion to the inoculum to prevent inhibition in a batch test. The

substrate to inoculum ratio to be respected is represented by Equation 3.5:

time

Biogas yield

m 2

A1 + A2

m 1

3 Materials and Methods

⁄≤0.5

(3.5)

Where S is the organic dry matter of the substrate (g) and I is the organic dry

matter of the inoculum in the reactor (g). In order to investigate the possibility

of different loads, batches were tested using a substrate to inoculum ratio

higher, lower, and approximately equal to 0.5.

3.4.2 Pilot scale: continuous tests

Continuous anaerobic digestion tests were performed according to the guide-

line VDI 4630 (2006). The apparatus consisted of a 20 L gastight stirred tank

with infeed and outlet, and a gas offtake connection to collect the formed biogas

(Figure 3.12). The temperature was kept under mesophilic conditions (37 °C ±

2 °C) using a circulation thermostat (Huber CC-202C) and the biomass was con-

stantly mixed at 50 rpm using a paddle stirrer (Heidolph RZR 2041). The biogas

formation in norm litres per kg of sludge (L kg-1 d-1) was measured daily using

a drum-type gas meter (Ritter TG05/5), and the quality of the gas as well as the

hydrogen sulfide concentration were measured using an electronic analyser

(OPTIMA 7 BIOGAS) once every two days. The hydrogen sulfide concentration

was controlled by adding iron (III) chloride hexahydrate (Merck) which works

as an H2S scavenger, if necessary. Mesophilic anaerobic sludge from the

SÜDLEDER tannery (Rehau, Germany) was used as inoculum. Because of its

source the sludge was also analysed regarding its chromium and collagen con-

tent.

At the beginning of the test the reactor was filled with approximately 20 kg of

inoculum (wet basis). Substrate was added once every two days excluding

weekends via a dip tube located at the head end of the reactor, starting at a

loading rate of substrate per kg of sludge of 0.5 g kg-1 d-1 (substrate mass in or-

ganic dry matter). After the daily methane production was constant, the loading

rate was raised by 0.5 units and the process was repeated until the biogas pro-

duction stopped to increase.

3.4 Biogas Production

Fig. 3.12: Continuous anaerobic digestion test apparatus (source: MyFerm I manual – Landgraf

Laborsysteme HLL GmbH).

Once a week a sample of biomass was taken for characterization regarding its

pH value, water content (DIN EN ISO 4684, 2005), inorganic matter (DIN EN

ISO 4047, 1998), chromic oxide content (DIN EN ISO 5398-1, 2007), and collagen

content through the determination of the hydroxyproline content. The biomass

was also analysed photometrically using Spectroquant® Cell Test Kits and

chromatographically to determine its volatile fatty acids content.

3.4.2.1 Chemical oxygen demand (COD) balance

The biogas formation measurements and the methane quantity were used to

calculate the substrate degradation through a COD balance with Equation 3.6

(VDI 4630, 2006). Using this method it is possible to determine the degradation

level of the substrates through biogas production (yield and gas composition)

and the substrate quantity fed data.

3 Materials and Methods

COD degree of degradation = VGas. xCH4

320.msubstrate. CODsubstrate %

(3.6)

Where VGas is the volume of biogas in mL d-1 (norm), xCH4 is the amount of me-

thane, msubstrate is the substrate addition in g d-1 (organic dry matter), and CODsub-

strate is the substrate’s chemical oxygen demand in gO2 g-1 (organic dry matter).

As explained in the guideline VDI 4630 (2006), this method is based on the ox-

ygen needed to oxidize the produced methane. From stoichiometry it is known

that 2 moles of O2 (64 g) are necessary to oxidize 1 mole of methane, meaning

that 1 mole of CH4 corresponds to a COD of 64 g. As 1 mole of gas methane

corresponds to 22.4 L (ideal gas law) the value of 350 mLCH4 gO2-1 was reached

but, as about 10% of the converted COD is for the reformation of biogas, this

means about 320 mLCH4 gO2-1. It is possible to have the COD of the produced

methane using this value, the quantity of methane, and the added substrate.

The portion of CO2 can be ignored because this carbon is already oxidized.

The COD of the substrate was calculated for bovine hide collagen with the stoi-

chiometry of Equation 3.7 (VDI 4630, 2006) and data from the UniProt database

(The UniProt Consortium). This value represents the chemically oxidisable ma-

terial, the maximum of energy that could be recovered by biogas (Drosg et al.,

2013).

CaHbOcNd+(2a+b

2−c−3d

2)O → aCO2+b−3d

2H2O+dNH3

(3.7)

3.4.3 Possible Sources of Errors

There are many different sources of error in an experiment, regardless of the

level of precision in which the experiments were carried out. Methods and

equipment used in this study present errors and have influence in the final re-

sults. The used techniques comprise physical, chemical, biochemical and bio-

logical test.

3.4 Biogas Production

The equipment used to measure the production of biogas present errors in their

measurements as specified by the manufacturer. In batch trials, the digital ma-

nometer Leo 3 has an error lower than 0.1% of full scale with a measuring range

of pressure of 0 to 4 bar. For the continuous trials, the electronic gas analyser

OPTIMA 7 BIOGAS has an absolute error of 0.5% or 5% of the measured value

of CH4 and CO2, whichever is larger.

Chemical analyses have errors even when carefully performed. For example,

each Spectroquant® Cell Test Kit has a different error. The Ammonium content

and Iron content kits have an absolute error of 0.05 mg L-1. The Chemical Oxy-

gen Demand kit has an absolute error of 29 mg L-1. The Sulfate content kit has

an absolute error of 8 mg L-1. Finally, the Volatile fatty acids kit has an absolute

error of 69 mg L-1.

The error of the determination of the hydroxyproline content is 1.3% of rate.

However, when the determination of the collagen content is necessary the value

has to be multiplied by the factor 7.46 and the error of the determination of the

collagen content is 9.7% of rate. Chemical analyses were also performed in du-

plicate or triplicate, with the exception of the Spectroquant® Cell Test Kit,

which were performed only once because of the limited amount of biomass

available for the analyse.

To ensure the quality of the biogas results, the batch trials were performed in

duplicate or triplicate. Continuous biogas trials could not be repeated because

trials were very time-consuming however the courses of the measured values

were carefully checked regarding plausibility. Nevertheless biological tests

show the highest variability because of a limited definition and characterization

of the sludge as a living system.

4 Results and Discussion

The present section illustrates chemical analyses and biogas production results

for the extruded shavings and offcuts, shavings treated hydrothermally, and

autoclaved shavings. Results for untreated chromium shavings and offcuts, and

for gelatin are shown for comparison purposes.

4.1 Chemical analyses of the pre-treated samples

Untreated chromium shavings and offcuts, and pre-treated shavings and off-

cuts were evaluated using different chemical analyses to determine their sus-

ceptibility to anaerobic degradation and get a hint on the effectiveness of the

pre-treatments.

4.1.1 Characterization of the pre-treated samples

The characterization of the pre-treated samples gives important information for

the biogas trials. Only organics are capable of producing biogas and, therefore,

it is important to quantify the organic and inorganic content of the studied sub-

strates. The characterization of the pre-treated shavings and offcuts, untreated

chromium shavings, untreated leather offcuts, and gelatin is shown in Table 4.1.

With the exception of hydrothermal treatment, the water content was reduced

by pre-treatments. Part of the non-bonded water of the chromium shavings and

leather offcuts is evaporated due to the high pre-treatment temperatures part

resulting in a drop of the pre-treated sample's water content.

4 Results and Discussion

Tab. 4.1: Characterization of the pre-treated and untreated samples.

Water content

(%)

Organic matter

(%)*

Hydroxyproline

content (%)*

Chromium

content (%)*,**

Gelatin

90.5 ± 0.0

99.0 ± 0.0

13.1 ± 0.2

Water content

(%)

Organic matter

(%)*

Collagen content

(%)*

Chromium

content (%)*,**

19.7 ± 0.1

88.8 ± 0.1

77.0 ± 0.6

4.6 ± 0.0

Offcuts

11.4 ± 0.0

93.6 ± 0.0

n.d.

4.1 ± 0.1

E100W

45.2 ± 0.7

88.9 ± 0.1

72.6 ± 2.4

n.d.

E130W

25.8 ± 0.2

89.0 ± 0.1

77.8 ± 1.1

n.d.

E150W

10.6 ± 0.2

89.1 ± 0.0

79.1 ± 0.3

n.d.

E170W

6.3 ± 0.1

88.9 ± 0.1

76.8 ± 0.7

4.4 ± 0.0

E100D

15.1 ± 0.1

88.9 ± 0.0

74.2 ± 0.9

4.6 ± 0.0

E130D

4.4 ± 0.0

88.7 ± 0.1

75.4 ± 0.7

n.d.

E150D

3.7 ± 0.1

88.8 ± 0.1

74.5 ± 1.0

n.d.

E170D

2.5 ± 0.1

88.9 ± 0.1

73.0 ± 1.7

n.d.

3.9 ± 0.1

93.0 ± 0.1

50.0 ± 0.5

3.9 ± 0.0

H140

83.1 ± 0.1

88.7 ± 0.1

73.7 ± 1.3

4.4 ± 0.0

H150

83.6 ± 0.1

88.9 ± 0.2

74.7 ± 0.3

4.4 ± 0.0

H170

85.3 ± 0.3

89.0 ± 0.0

76.6 ± 1.2

4.3 ± 0.0

11.8 ± 0.9

n.d.

12.2 ± 0.2

n.d.

A12

11.3 ± 0.1

n.d.

A24

11.0 ± 0.3

n.d.

A48

10.6 ± 0.0

n.d.

A96

10.3 ± 0.3

n.d.

A192

9.4 ± 0.5

n.d.

A384

10.9 ± 0.2

n.d.

*Dry basis; mean ± standard deviation, n = 3

**Measured as chromium oxide

n.d. – not determined

The organic matter in the samples remains the same after pre-treatments, about

93% for leather offcuts and 89% for chromium shavings. This is important be-

cause the organics must be preserved for producing biogas in the anaerobic di-

gestion. The extruded offcuts and untreated offcuts have a higher content of

4.1 Chemical analyses of the pre-treated samples

organics because of all the chemicals added during the wet end and finishing

process, such as dyes, retanning agents, fatliquors, and pigments (Covington,

2009).

The collagen content is also barely unchanged after pre-treatment. This protein

is the main component of the chromium leather waste. Only about 12% of the

chromium shavings are a different type of organics, for instance fats. Therefore,

collagen is the most important parameter to calculate the substrate degradation

after anaerobic digestion. The collagen content of the extruded offcuts is 50%

only. A large part of the organics is from other sources and could have a very

low degradability.

Almost half of the inorganic part of the samples is chromium oxide. Chromium

also remains the same after pre-treatment. Other inorganics in the samples re-

sult from the chemicals used in tanneries.

The autoclaved shavings were produced on a very small scale and, therefore,

most of the characterization parameters were not measured. The pre-treated

sample available was not sufficient for the experiments.

4.1.2 Scanning electron microscopy

Chromium shavings and leather offcuts were compared with the different pre-

treated samples using SEM images. Figure 4.1a shows the untreated chromium

shavings. A fibrous structure can be seen, the fibrils closely aligned together

forming the fibres. Similarly, the leather offcuts show an intact fibrous structure

(Figure 4.1b).

Not all SEM images were shown for length and clarity. However, the SEM im-

ages for the pre-treated chromium leather waste not illustrated in this section

can be seen in the Appendix (Figure 5.1).

4 Results and Discussion

Fig. 4.1: SEM images of chromium shavings (a), leather offcuts (b), shavings extruded dry at

100 °C (c), extruded leather offcuts (d), shavings treated hydrothermally at 140 °C (e), shavings

treated hydrothermally at 170 °C (f), autoclaved over a period of 3 minutes (g), and autoclaved

over a period of 384 minutes (h) at 60,000 × magnification.

4.1 Chemical analyses of the pre-treated samples

Figure 4.1c shows the shavings extruded dry at 100 °C and Figure 4.1d the

leather offcuts extruded at 170 °C. Severe damage to the fibrous structure can

be seen for both samples, fibrils are randomly positioned and the structure ap-

pears to be shrunken. Since the collagen molecules were organized as tissue, the

fibres shrink when exposed to temperatures higher than the denaturation tem-

perature (Meyer, 2019). An organized fibrous structure, as identified for the un-

treated chromium shavings and leather offcuts, cannot be seen anymore. Simi-

lar behaviour was verified before by Meyer et al. (2005) using SEM images. The

authors observed a fibrillar structure in the initial collagenous material, a parch-

ment-like raw material from limed cattle hides, but after thermomechanical

treatment in an internal mixer (thermo-mechanical treatment) this structure dis-

appeared.

The shavings treated hydrothermally at 140 °C and 170 °C (Figures 4.1e and

4.1f) show a rather deteriorated structure. The high temperatures, which the

chromium shavings had been exposed to, started a melting process of the fibres.

However, the melting appears to be superficial and the randomly positioned

fibres are still present. In Figure 4.1e, the captured surface appears to be molten

but the pattern of the fibres can still be seen, indicating superficial melting. In

Figure 4.1f, molten areas can be seen in the lower part of the capture but fibres

are also clearly visible. The presence of molten areas is a strong indicator for the

fact that the samples are denatured.

The shavings autoclaved over a period of three minutes (Figure 4.1g) are very

similar to the untreated chromium shavings (Figure 4.1a), however fibrils on

the surface are detached from the main structure, probably due to initial shrink-

ing. The shavings autoclaved over a period of 384 minutes (Figure 4.1h) were

pre-treated for a sufficient time to denature the chromium shavings and the

surface of the material appears to be molten.

Additionally, for the shavings treated hydrothermally at 170 °C (Figure 4.1f),

formation of crystals was identified after the hot liquid sample generated by

hydrothermal treatment had been solidified. Crystals were also found for other

extruded shavings (Figure A.1). In order to identify the observed crystals, chro-

mium shavings, shavings extruded dry at 100 °C, shavings extruded wet at

4 Results and Discussion

170 °C, shavings treated hydrothermally at 150 °C, and shavings treated hydro-

thermally at 170 °C were analysed using energy-dispersive X-ray spectroscopy

(EDX). Results are shown in Figure 4.2.

Fig. 4.2: SEM images for EDX of chromium shavings (a), shavings extruded dry at 100 °C (b),

shavings extruded wet at 170 °C (c), shavings treated hydrothermally at 150 °C (d), shavings

treated hydrothermally at 170 °C (e) at different magnifications.

4.1 Chemical analyses of the pre-treated samples

Formation of crystals was observed for all analysed samples and it is presumed

that these crystals can be found on the surface of all samples. The elements pre-

sent in the crystals were analysed using EDX. Results revealed that these crys-

tals are mainly composed of calcium (about 20%), sulfur (about 15%) and oxy-

gen (about 65%). Consequently, these crystals could be CaSO4 crystals, which

possibly originated from the basic chromium sulfate used in the tanning process

to produce leather. The shape of the crystal observed in the SEM images is also

compatible with the CaSO4 observed by Sirota et al. (1992).

4.1.3 Differential scanning calorimetry and enzyme assays

Specific enthalpy found in DSC analyses and degradation by trypsin results for

the extruded shavings and offcuts (Figure 4.3a), the shavings treated hydrother-

mally (Figure 4.3b), and autoclaved shavings (Figure 4.3c) are compared to re-

sults for the untreated chromium shavings. The error bars represent the stand-

ard deviation for the experimental data. Analysis of the results led to the con-

clusion that the triple helical structure in almost all samples could be denatured

due to the pre-treatments. DSC results show that, with exception of the shav-

ings autoclaved over a period of three minutes, there is no enthalpy or hardly

any enthalpy for the pre-treated samples. Therefore, no hydrogen bonds stabi-

lizing the triple helix remain.

Even though collagen in the samples was denatured due to the pre-treatments,

the hydroxyproline present in the triple helix chains appears to be intact. This

means that the pre-treatment opens up the triple helix structure without de-

stroying the chains. This can be seen in the collagen content of the samples (Ta-

ble 4.1), which is very similar for untreated and pre-treated samples. The

method used to measure the collagen content is also used to determine the con-

tent of the amino acid hydroxyproline (Stegemann, 1958), which is almost the

same before and after pre-treatment.

4 Results and Discussion

Fig. 4.3: Enthalpy of the denaturation process and degree of degradation by trypsin of extruded

shavings and offcuts (a), shavings treated hydrothermally as a function of the pre-treatment

temperature (b), and autoclaved shavings as a function of the pre-treatment time (c) and chro-

mium shavings.

Pre-treatment time (min)

0100 200 300 400

Specific enthalpy (J g-1)

D. by trypsin after 5 h (%)

100

Enthalpy

Degradation by trypsin

Autoclaved shavings

Pre-treatment temperature during extrusion (°C)

02080 100 120 140 160

Specific enthalpy (J g-1)

D. by trypsin after 5 h (%)

100

Enthalpy

Degradation by trypsin - Dry shavings

Degradation by trypsin - Wet shavings

Degradation by trypsin - Offcuts

Extruded shavings and offcuts

Pre-treatment temperature during hydrothermal treatment (°C)

02080 100 120 140 160

Specific enthalpy (J g-1)

D. by trypsin after 5 h (%)

100

Enthalpy

Degradation by trypsin

Shavings treated hydrothermally

4.1 Chemical analyses of the pre-treated samples

Trypsin assays show that trypsin can degrade the chromium shavings more

easily if they were pre-treated. Trypsin was able to degrade the chromium shav-

ings and pre-treated samples, which contain more than 4% chromium oxide

(Table 4.1). Consequently, this quantity of chromium is not toxic for trypsin. It

can be assumed that the levels of chromium content in the studied chromium

leather waste are lower than what would be toxic for other enzymes in the an-

aerobic digestion.

Figure 4.3a reveals that using extrusion as a pre-treatment it is possible to in-

crease the susceptibility of the chromium shavings to be degraded by trypsin.

The degradation increases from 6.7 ± 0.4% up to 35.2 ± 0.4% when using shav-

ings extruded wet at 170 °C instead of untreated chromium shavings. Results

showed an increasing tendency of degradation with increasing extrusion tem-

perature. Degradation levels for the previously moistened extruded samples for

the temperatures 150 °C and 170 °C were slightly higher than for the shavings

extruded dry at the same temperatures. The means of digestibility were com-

pared by using the t-test. There was no statistically significant difference be-

tween samples extruded dry and wet at 100 °C (p = 0.057), and extruded dry

and wet at 130 °C (p = 0.257), but there was a difference between samples ex-

truded dry and wet at 150 °C (p = 0.010), and dry and wet at 170 °C (p = 0.004).

The comparison between samples extruded dry at 100 °C and wet at 170 °C

showed a statistically significant difference (p = 5.10-6).

With regard to the extruded offcuts, the sample had a very low degradation by

trypsin value (4.1 ± 0.3%), even lower than the degradation value obtained for

untreated chromium shavings (6.7 ± 0.4%). Leather offcuts were treated in nu-

merous steps after tanning unlike chromium shavings. The wet end and finish-

ing process in the tannery industry made this chromium leather waste even

more inaccessible for anaerobic degradation and insoluble with the addition of

a variety of polymers during the process.

The results of degradation by trypsin for the shavings treated hydrothermally

showed a clear increase when using higher pre-treatment temperatures. The

shavings pre-treated at 170 °C were highly accessible for trypsin reaching the

highest degradation level among the studied samples, 90.3 ± 0.8% (Figure 4.3b).

4 Results and Discussion

The autoclaved shavings (Figure 4.3c) showed high degradation even with

short pre-treatment time. After three minutes of thermal pre-treatment only, the

degradation of the waste went from 6.7 ± 0.4% (untreated chromium shavings)

to 19.8 ± 1.0% and within only 24 minutes it was possible to have 51.8 ± 2.1% of

degradation reaching more than 90% after 192 minutes. After 192 minutes of

heat treatment, degradation reached a plateau and any longer autoclaving pro-

cess would not be reasonable.

Additionally, the untreated chromium shavings and the shavings extruded dry

at 100 °C were tested for degradation by collagenase to evaluate if this enzyme,

capable of destabilizing collagen (Meyer, 2019), would be more effective to de-

grade untreated and pre-treated chromium shavings. The Clostridium histolyti-

cum has the ability to produce collagenase and belongs to the bacteria clostridia,

the main bacteria found in sludge (McInerney, 1988). Clostridium histolyticum is

probably present in the medium producing collagenase. Therefore, this enzyme

could play an important role in the degradation of the studied substrates. Re-

sults of degradation by collagenase of the untreated chromium shavings, shav-

ings extruded dry at 100 °C, and white hide powder, untanned hide used as

reference, are shown in Figure 4.4. The error bars represent the standard devia-

tion for the experimental data.

Even with longer times of contact between substrate and enzyme, collagenase

did not digest the untreated chromium shavings effectively. Figure 4.4 shows

that degradation of the untreated chromium shavings reached a value of about

12% after one day and it remains the same after seven days. The white hide

powder was quickly hydrolyzed by the collagenase. This was to be expected

because untanned hides are prone to degradation by proteases. Degradation of

the shavings extruded dry at 100 °C increased linearly with the time of contact,

reaching up to 86% of degradation after seven days which is almost the degra-

dation degree of the white hide powder. Pre-treatment of the chromium shav-

ings was necessary to enable their degradation by collagenase, which runs

slower than for the white hide powder.

Results published by Covington (2009) also suggest that the untreated chro-

mium shavings cannot be easily degraded by proteases. Protease assays

4.1 Chemical analyses of the pre-treated samples

showed that the enzymes had no visible effect on chromium leather samples

unless the concentration of enzyme is abnormally high, the pH value is close to

9, and the temperature of the reaction is close to 50 °C. Under these conditions

no effects are observed up to the point the chromium-tanned leather is suddenly

and completely degraded.

Fig. 4.4: Degree of degradation by collagenase of white hide powder, untreated chromium shav-

ings, and shavings extruded dry at 100 °C as a function of time.

Additionally, Meyer (2019) agrees that a pre-treatment can ease enzymatic deg-

radation of tanned hides. Under different conditions, degradation of collagen

was tested with different enzymes. Porcine hides with and without synthetic

cross-links (tanning effect achieved by hexamethylendiisocyanate) were de-

graded by enzymes, such as trypsin and collagenase. Synthetic cross-links sup-

pressed the enzymatic action of all enzymes tested. Enzymatic degradation of

the tanned structure was only possible when collagen was denatured

(pre-treated by heat). However, collagenase is much more effective for degra-

dation of the pre-treated collagen than trypsin.

time (days)

0 1 2 3 4 5 6 7 8

Degradation by collagenase (%)

100

White hide powder

Chromium shavings

Extruded shavings

4 Results and Discussion

Results demonstrated that almost all samples were completely denatured by

the pre-treatments tested. Moreover, chromium shavings are more easily acces-

sible to degradation by proteases after pre-treatment than without pre-treat-

ment. Neither trypsin nor collagenase were able to effectively degrade un-

treated chromium shavings. The higher the pre-treatment temperature the eas-

ier the sample degradable by trypsin. However, the wet end and finishing pro-

cess prevents accessibility to trypsin even after the leather offcuts were pre-

treated by extrusion. The quantity of chromium present in the chromium

leather waste was not toxic to trypsin and collagenase.

4.1.4 Solubility in water

Figure 4.5 shows the results of solubility in water at a temperature of 37 °C for

the pre-treated samples and the untreated chromium shavings. The error bars

represent the standard deviation for the experimental data.

Figure 4.5a shows that the extruded shavings and the untreated chromium

shavings had very similar values of solubility in water at a temperature of 37 °C.

Despite the visual similarities, the solubility results for most of the extruded

shavings showed statistically significant differences when compared to un-

treated chromium shavings using the t-test. An exception were the shavings

extruded dry at 130 °C (p = 0.529). In general, solubility was slightly reduced by

pre-treatment, probably due to some additional cross-links generated by extru-

sion (Klüver and Meyer, 2013).

4.1 Chemical analyses of the pre-treated samples

Fig. 4.5: Solubility in water at a temperature of 37 °C of extruded shavings and offcuts (a),

shavings treated hydrothermally as a function of the pre-treatment temperature (b), and auto-

claved samples as a function of the pre-treatment time (c) and chromium shavings.

Pre-treatment time (min)

0100 200 300 400

Solubility in water at 37°C (%)

100

Autoclaved shavings

Pre-treatment temperature (°C)

020 80 100 120 140 160 180

Solubility in water at 37°C (%)

100

Dry shavings

Wet shavings

Offcuts

Pre-treatment temperature (°C)

020 80 100 120 140 160 180

Solubility in water at 37°C (%)

100

Extruded shavings and offcuts

Shavings treated hydrothermally

4 Results and Discussion

Gorham et al. (1992) studied the formation of cross-links on dry collagen treated

thermally (heat cured), and concluded that intermolecular amide bonds were

formed. The cross-linking mechanism involves the reaction of aspartate and

glutamate residues with the amino groups of lysine and hydroxylysine. They

also found a reduction in solubility and an increase in degradation by trypsin

but in their case the solubility reduction was more significant because the sam-

ples were dried and the presence of moisture in the samples accelerated the rate

of denaturation with heat. However, the reduction in solubility verified for the

extruded shavings was very low. The values obtained were also very similar

for chromium shavings treated at the same temperature with or without hu-

midification process. The only extruded shavings that showed significant dif-

ferences were the shavings extruded dry and wet at 100 °C (p = 0.019).

The hydrogen bonds that stabilize the triple helices could be broken down by

extrusion pre-treatment which also enabled degradation of the chromium shav-

ings as can be seen from the DSC results. However, no chromium shavings

could be hydrolyzed and no gelatin generated by extrusion. Consequently, sol-

ubility in water could also not be increased by extrusion pre-treatment. The

transformation of collagen-based materials into gelatin depends on exposing

the material to a temperature higher than the denaturation temperature and

breaking down several bonds to tear apart the triple helix structure. To degrade

the samples, high temperatures and mechanical forces are used for extrusion

pre-treatment. However, it was not possible to generate gelatin. This is proba-

bly due to the fact that the time of contact of the samples with the heated ex-

truder, approximately three minutes, is not sufficiently long to transfer the col-

lagen molecules into gelatin chains, but it was possible of shrink them causing

the small reduction in solubility.

Extruded offcuts (Figure 4.5b) showed the lowest solubility among the tested

samples, 0.5 ± 0.2%, even lower than that of untreated chromium shavings. This

low value is a consequence of the wet end and finishing process in tanneries,

during which chromium-tanned leather is processed to produce finished

leather. It is not possible to revert this effect by extrusion pre-treatment, and the

material remains almost insoluble.

4.1 Chemical analyses of the pre-treated samples

The solubility of the shavings treated hydrothermally (Figure 4.5b) can be re-

lated to the fibrous structure of the samples and the reduction of protein chain

lengths. The solubility of the shavings pre-treated hydrothermally at 170 °C,

which had liquid consistency immediately after the pre-treatment, increased

from 8.2 ± 0.3% (untreated chromium shavings) to 27.1 ± 0.9%. However, com-

pared to soluble gelatin this value is still low. Therefore, the collagen was not

completely transformed into gelatin. The solubility of the shavings treated hy-

drothermally at 140 °C and 150 °C, 9.4 ± 0.9% and 8.4 ± 0.6% respectively,

showed no statistically significant difference when compared to the untreated

chromium shavings for which the t-test was applied (p = 0.197 and p = 0.640,

respectively). As observed in SEM, melting of the structure caused by hydro-

thermal treatment was superficial and the samples still show a partially fibrous

structure (Figure 4.1e and 4.1f).

The solubility of the autoclaved shavings (Figure 4.5c) increases with longer

pre-treatment times. The shavings treated for more than 96 minutes, that is

those which showed melting of the fibrous structure in some areas, had a linear

growth trend of solubility with regard to the pre-treatment time. With three

minutes of autoclaving pre-treatment the results of solubility in water are very

similar to the result of untreated chromium shavings because the pre-treatment

time was insufficient to degrade the chromium shavings. In other words, the

longer the pre-treatment time the higher the solubility in water, and it can also

be concluded that the degree of degradation of the shavings is higher.

In general, solubility in water of the chromium leather waste was not affected

by the pre-treatments. Chromium shavings treated hydrothermally at 170 °C

and shavings autoclaved for more than 96 minutes were the exception. How-

ever, even in these cases the solubility is low compared to that of gelatin. The

extruded offcuts showed the lowest solubility due to the wet end and finishing

process.

The chemical analyses used to evaluate the pre-treated chromium leather waste

show that all extruded shavings and shavings treated hydrothermally, and

most autoclaved shavings are denatured. The chromium shavings are more eas-

ily accessible to degradation by trypsin and collagenase after pre-treatment.

4 Results and Discussion

However, for most of the pre-treated shavings solubility does not change. Ex-

truded leather offcuts show a different behaviour with lower values of degra-

dation by trypsin and lower solubility than untreated chromium shavings.

4.2 Biogas Production

The biogas and methane production of the pre-treated chromium leather waste

was investigated through batch and continuous biogas production trials to

prove feasibility of using pre-treated samples to produce biogas. Autoclaved

shavings were produced on a small scale. Screw cap micro tubes (2 mL) were

used as autoclave apparatus because the use of a laboratory autoclave does not

allow for precise control of the pre-treatment time. Therefore, testing the auto-

claved shavings in biogas trials was unviable.

4.2.1 Laboratory scale: batch tests

The biogas formation potential was measured in batch tests for untreated chro-

mium shavings, pre-treated samples, and gelatin.

4.2.1.1 Comparison of different agitation methods

Initially, two samples pre-treated with extrusion and the untreated chromium

shavings were tested for biogas production. Batches were tested with agitation

in a shaker water bath and without agitation in a climatic chamber. The shav-

ings extruded dry at 100 °C and extruded wet at 170 °C were selected as they

represent the extremes of the extrusion pre-treatment, the former being the sam-

ple with the lowest degree of treatment, and the latter the highest degree of

treatment. The untreated chromium shavings were tested for comparison pur-

poses. Trials with the same substrate had the same substrate to inoculum (S/I)

ratio in the batch with and without agitation. The two-phase Gompertz model

was used to fit the experimental data for a better representation of the hydroly-

sis and lag-phase of the anaerobic digestion of the tested substrates. Results are

4.2 Biogas Production

shown in Figure 4.6 and Table 4.2. The error bars in Figure 4.6 represent the

standard deviation for the experimental data.

Fig. 4.6: Cumulative biogas production and two-phase Gompertz simulation (red line) for chro-

mium shavings (S/I = 2.6), shavings extruded wet at 170 °C (S/I = 3.2), and dry at 100 °C

(S/I = 2.9) in agitated bioreactors (a) and in non-agitated bioreactors (b).

Tab. 4.2: Parameters of biogas production from the two-phase Gompertz equation.

λ1 (d)

(L kg-1 d-1)

(L kg-1)

λ2

(d)

(L kg-1 d-1)

A1+A2

(L kg-1)

Non-agitated

8.8

24.5

235.0

∞

0.9907

E100D

1.8

20.0

157.8

38.3

5.9

281.4

0.9989

E170W

1.5

15.6

127.6

29.7

6.8

277.2

0.9980

Agitated

6.9

21.3

196.7

∞

0.9969

E100D

1.6

31.7

228.2

25.2

3.7

370.9

0.9996

E170W

1.0

22.1

164.4

25.9

4.4

326.6

0.9979

The modified Gompertz model has been widely used in the simulation of bio-

gas and methane production results (Li et al., 2011; Kafle and Kim, 2013b;

Chanakya et al., 2015; Kim and Kim, 2017). This model can predict the lag-phase

of the process, a period associated with the hydrolysis of the substrate in the

digester and presents a good fit to experimental results (Kafle and Kim, 2012;

time (days)

020 40 60 80 100

Biogas yield (L kg-1)

100

200

300

400

E170W

E100D

TGM

time (days)

020 40 60 80 100

Biogas yield (L kg-1)

100

200

300

400

4 Results and Discussion

Kafle and Kim, 2013a; Li et al., 2016; Rajput et al., 2018). However, it was neces-

sary to adapt the equation to fit biogas production curves with two phases cre-

ating a two-phase Grompertz model (Equation 3.4).

The adapted equation was able to fit biogas production curves with one or two

phases. For all tested conditions the coefficient of determination (R2) is higher

than 0.99 indicating that the model fitted well the experimental data. Untreated

chromium shavings had a one-phase biogas production curve. This is indicated

in the two-phase Grompertz model with high values of λ2 or an A2 of almost 0.

A comparison between results of batches with and without agitation indicated

that agitation reduces the lag-phase of the anaerobic digestion for untreated and

pre-treated shavings (Table 4.2) and consequently the hydrolysis phase. Re-

garding the maximum biogas production rate of the first phase, values in-

creased for the extruded shavings but not for the untreated chromium shavings.

The agitation of the reactors eases the mass transfer in the bioreactors and

prompted contact between substrate and bacteria increasing the production

rate. The agitation of the reactors favours the biogas production for the ex-

truded shavings and it is expected that the same would occur for all pre-treated

samples.

The degradation degree of collagen of each trial was estimated after the charac-

terization of the final biomass in the reactor. Considering the content of collagen

of each substrate (Table 4.1) and using a mass balance, the collagen content at

the beginning of the trials (combination of inoculum and substrate) can be cal-

culated. Using this value and the collagen content of the final biomass in the

reactor with Equation 3.3 it was possible to calculate the degradation degree of

collagen achieved by the process. Table 4.3 shows the characterization of the

final biomass and the degradation degree of collagen in the process.

The characterization of the final sludge enabled the estimation of the degrada-

tion degree of collagen (Table 4.3). Results with and without agitation were very

similar. The extruded samples showed a degradation degree of collagen above

99%, a very high destruction of the collagen of the original substrate and much

4.2 Biogas Production

higher than the degree reached by the untreated chromium shavings. This in-

dicates that chromium shavings are a complex substrate and the pre-treatments

are important to assure the reduction of the final waste.

Tab. 4.3: Biomass characterization after digestion and degradation degree of collagen for agitated

and non-agitated bioreactors.

Collagen

(%)*

Degradation

degree (%)

Chromium

(%)**

Non-agitated

29.9 ± 3.4

51.4

2.1 ± 0.1

E100D

0.5 ± 0.1

99.5

2.1 ± 0.0

E170W

0.4 ± 0.1

99.6

3.1 ± 0.3

Agitated

32.1 ± 1.3

44.7

2.2 ± 0.0

E100D

0.4 ± 0.0

99.5

2.2 ± 0.1

E170W

0.2 ± 0.0

99.8

3.9 ± 0.5

*Dry basis; mean ± standard deviation, n = 3

**Dry basis; mean ± standard deviation, n = 2, measured as chromium oxide

Comparison of the biogas formation potential results of batches with and with-

out agitation indicated that agitation favours the biogas production for the pre-

treated samples. The shavings extruded wet at 170 °C had a biogas formation

potential of about 327 and 277 L kg-1 in the trials with and without agitation,

respectively. The shavings extruded dry at 100 °C had a biogas formation po-

tential of about 371 and 281 L kg-1 (Table 4.2). This means an increase of biogas

formation potential of almost 20% for the first substrate when comparing the

agitated with the non-agitated batches. The biogas formation potential of the

untreated chromium shavings in both batches were very similar, about

235 L kg-1 for the non-agitated and 197 L kg-1 for the agitated batch.

The use of agitation could not only reduce the lag-phase for the chromium shav-

ings but also improve every aspect of the biogas production using extruded

chromium shavings. The agitation of the system facilitates the mass transfer in

the bioreactors and the contact between substrate and bacteria. Therefore, all

the following biogas production trials were carried out using agitation in a

shaker water bath.

4 Results and Discussion

4.2.1.2 Comparison of different pre-treatments

The biogas yield was measured in batch tests with untreated chromium shav-

ings, gelatin, shavings extruded at different temperatures and humidity condi-

tions and shavings treated hydrothermally at different temperatures. The dif-

ferent substrates were exposed to anaerobic digestion with a substrate to inoc-

ulum ratio of approximately 0.5 to investigate their biogas formation potential

according to the guideline VDI 4630 (2006). The two-phase Gompertz model

was used to fit the experimental data (Table 4.4). All bioreactors were agitated.

The biogas formation curves of the substrates are represented in Figure 4.7. The

error bars represent the standard deviation for the experimental data. The bio-

gas yield did not significantly differ for the substrates in the three clusters –

shavings extruded dry, shavings extruded wet, and shavings treated hydro-

thermally – (Figures 4.7b, 4.7d, and 4.7c, respectively). A test of analysis of var-

iance (ANOVA) showed no statistically significant difference after the eighth

day of anaerobic digestion for all clusters.

The shape of the biogas formation curve for the untreated chromium shavings

(Figure 4.7a) indicates a retarded degradation. Agustini et al. (2015) also

showed a similar degradation behaviour in batch trials when studying biogas

production with chromium shavings. This is the case if the substrate degrades

with difficulty (VDI 4630, 2006).

For all substrates tested, R2 is higher than 0.99 indicating that the experimental

data fitted well to the two-phase Gompertz model. Substrates had a one-phase

biogas production curve, which can be seen in the two-phase Grompertz model

with high values of λ2 or A2 of almost 0 L kg-1. The reactors with shavings ex-

truded dry and wet (Figures 4.7b and 4.7d) had very similar behaviours. Com-

pared to untreated chromium shavings (Table 4.4), the lag-phase of the biogas

production was reduced to two days due to extrusion pre-treatment. The trials

with shavings treated hydrothermally (Figure 4.7c) showed even better results

starting biogas production between four and five days before that of the trial

with untreated chromium shavings. For the shavings treated hydrothermally at

170 °C, production was observed almost immediately after starting the anaero-

4.2 Biogas Production

bic reactor, a lag-phase similar to the value of gelatin. The degradation behav-

iour of the pre-treated shavings rather resembled that of gelatin than that of

untreated chromium shavings. Therefore, the more complex the structure of

collagen, the later biogas production starts. As the untreated chromium shav-

ings did not have their collagen structure denatured by pre-treatment, the an-

aerobic sludge needs a longer period of time to hydrolyze the substrate before

initiating any biogas production.

Fig. 4.7: Cumulative biogas production and two-phase Gompertz simulation (red line) for the

untreated shavings and gelatin (a), shavings extruded dry (b), shavings treated hydrothermally

(c), and shavings extruded wet (d) in agitated bioreactors.

time (days)

0 5 10 15 20 25

Biogas yield (L kg-1)

100

200

300

400

500

Gelatin

TGM

time (days)

0 5 10 15 20 25

Biogas yield (L kg-1)

100

200

300

400

500

E100W

E130W

E150W

E170W

TGM

time (days)

0 5 10 15 20 25

Biogas yield (L kg-1)

100

200

300

400

500

E100D

E130D

E150D

E170D

TGM

time (days)

0 5 10 15 20 25

Biogas yield (L kg-1)

100

200

300

400

500

H140

H150

H170

TGM

4 Results and Discussion

The maximum biogas production rate could not be increased by pre-treatments.

The results for the pre-treated shavings are between 69.8 and 87.4 L kg-1 d-1,

while the untreated chromium shavings show a value of 84.5 L kg-1 d-1. All sub-

strate effectively hydrolyzed in the hydrolysis phase or by pre-treatment, can

be processed at a similar rate in the subsequent stages of anaerobic digestion.

Tab. 4.4: Parameters of biogas production from the two-phase Gompertz equation.

λ1

(days)

(L kg-1 d-1)

(L kg-1)

λ2

(days)

(L kg-1 d-1)

(L kg-1)

Gelatin

0.8

89.0

503.4

∞

0.9962

5.9

84.5

430.9

0.0

0.9966

Shavings extruded wet

λ1

(days)

(L kg-1 d-1)

(L kg-1)

λ2

(days)

(L kg-1 d-1)

(L kg-1)

E100W

2.3

72.4

453.2

∞

0.0

0.9973

E130W

2.3

74.6

456.7

0.0

0.9966

E150W

2.0

74.6

446.4

∞

0.0

0.9975

E170W

1.9

69.8

442.0

∞

0.0

0.9964

Shavings extruded dry

λ1

(days)

(L kg-1 d-1)

(L kg-1)

λ2

(days)

(L kg-1 d-1)

(L kg-1)

E100D

1.9

78.8

427.0

0.0

0.9978

E130D

2.0

73.2

412.2

0.0

0.9976

E150D

2.1

83.8

434.6

0.0

0.9977

E170D

2.1

77.4

435.9

0.0

0.9982

Shavings treated hydrothermally

λ1

(days)

(L kg-1 d-1)

(L kg-1)

λ2

(days)

(L kg-1 d-1)

(L kg-1)

H140

2.1

79.8

442.2

0.0

0.9987

H150

1.7

87.4

422.4

0.0

0.9988

H170

0.8

75.4

451.9

∞

0.0

0.9975

The biogas formation potential of all pre-treated shavings was very similar, be-

tween 412 and 457 L kg-1 (Table 4.4). The trial using shavings extruded wet at

130 °C as substrate reached the highest biogas formation potential, 457 L kg-1.

Comparing this with the results of other authors an improvement in the biogas

yield can be found. For instance, Priebe et al. (2016) reached a biogas yield of

4.2 Biogas Production

162.16 L kg-1 testing the degradation of chromium-tanned leather shavings in

different biotas and Agustini (2018) found a biogas cumulative production of

29.91 L kg-1 studying the degradation of chromium-tanned leather shavings us-

ing sludge of wastewater treatment as inoculum.

However, the biogas formation of the pre-treated substrates did not reach the

theoretical biogas yield of 796 L kg-1 calculated with the basic elementary for-

mula of bovine hide collagen and Equation 2.6. Hides are heterotypic fibrils,

which mainly contain collagen type I and a significant amount of type III colla-

gen, typically about 20% by mass (Wess, 2008). Knowing this and with data

from the UniProt database (The UniProt Consortium) it is possible to obtain the

basic elementary formula for bovine hide collagen necessary and calculate the

theoretical biogas yield. The trials using pre-treated chromium shavings as sub-

strate reached 52% to 58% of the maximum conversion capacity for the sub-

strate. A conversion rate between 50 and 70% is expected for proteins (VDI 4630,

2006).

Surprisingly, the biogas formation potential of the untreated chromium shav-

ings was similar to the values found for the pre-treated shavings. Considering

the values found in the enzyme assays with trypsin and collagenase, the chro-

mium shavings were expected to be less degraded and to generate a considera-

ble amount of biogas less than the pre-treated shavings in biogas trials. The

degradation of collagen-based materials in anaerobic digesters appears to be a

complex process. If the substrate is not denatured in advance, hydrolysis prob-

ably involves several enzymes working simultaneously to degrade the material

which requires some days causing a lag-phase. To confirm this and determine

the involved enzymes further research is needed.

Comparing these results with the previous section (Table 4.2), one can see that

the same substrates with the same agitation conditions obtained higher biogas

formation potentials in the present section. The difference is caused by the dif-

ferent substrate to inoculum ratio utilized in the reactor flasks. Better results are

achieved using a smaller amount of sample. A reason can be that this smaller

4 Results and Discussion

quantity favours the mass transfer inside the reactors and hence the contact be-

tween substrate and anaerobic bacteria, while a larger load results in an over-

load of the reactor and inhibits the anaerobic bacteria.

The degradation degree of collagen was estimated by characterization of the

biomass collected at the end of the digestion from each reactor. Table 4.5 shows

the characterization of the final biomass and the degradation degree of collagen

or hydroxyproline for each substrate tested.

Tab. 4.5: Biomass characterization after digestion and degradation degree of collagen for biore-

actors testing different pre-treatments.

Hydroxyproline

(%)*

Degradation

degree (%)

Chromium

(%)**

Gelatin

0.04 ± 0.01

98.8

Collagen (%)*

Degradation

degree (%)

Chromium

(%)**

4.7 ± 0.7

77.2

0.2 ± 0.1

Shavings extruded wet

E100W

0.3 ± 0.0

98.8

0.8 ± 0.0

E130W

0.3 ± 0.1

99.0

0.5 ± 0.0

E150W

0.2 ± 0.1

99.3

0.6 ± 0.0

E170W

0.2 ± 0.0

99.2

0.4 ± 0.0

Shavings extruded dry

E100D

0.3 ± 0.0

99.1

1.3 ± 0.1

E130D

0.3 ± 0.0

99.1

1.1 ± 0.1

E150D

0.3 ± 0.1

99.0

0.9 ± 0.0

E170D

0.4 ± 0.1

98.5

0.8 ± 0.1

Shavings treated hydrothermally

H140

0.4 ± 0.1

98.5

0.6 ± 0.1

H150

0.5 ± 0.1

98.2

0.9 ± 0.1

H170

0.6 ± 0.0

97.8

0.5 ± 0.1

*Dry basis; mean ± standard deviation, n = 3

**Dry basis; mean ± standard deviation, n = 2, measured as chromium oxide

During anaerobic digestion of untreated chromium shavings, the collagen con-

tent only partly decreased while for the pre-treated shavings it was reduced

4.2 Biogas Production

almost to zero. The trials with pre-treated shavings showed a degradation de-

gree above 97% in every case, a value similar to the degradation degree of hy-

droxyproline in the trial using gelatin. There is a clear difference between trials

with pre-treated and untreated shavings which showed a collagen degradation

of 77.2% only. Again chromium shavings have proven to be a complex substrate

that can be pre-treated to assure high decomposition. However, surprisingly

the trials with pre-treated shavings did not reach the same biogas formation

potential as gelatin.

The untreated chromium shavings showed the same biogas formation potential

as the pre-treated shavings, but they had a longer lag-phase and needed a

longer period of time to be degraded. Gelatin was more easily degradable and

digestion of the pre-treated shavings was not able to reach its biogas formation

potential. If the chromium shavings are denatured by extrusion pre-treatment,

the collagen chains are broken down and lose their stability but they are still

mainly arranged in long chains. This stiff arrangement also has to be broken

down and its degradation is difficult (Suzuki et al., 2006). The effort necessary

to degrade the gelatin is much lower because hydrolysation of the main chains

has already been achieved by the manufacturing process of gelatin. Therefore,

microorganisms will need much more energy to degrade pre-treated chromium

shavings than gelatin.

In addition, degradation of gelatin did not reach the theoretical biogas yield of

796 L kg-1 calculated with the basic elementary formula of bovine hide collagen

and Equation 2.6. Gelatin reached 63.2% of the maximum conversion capacity

for the substrate. The value for the biogas formation potential of gelatin is also

lower than that found for casein and blood as protein-rich substrates, 775 and

824 L kg-1, respectively (Kovács et al., 2013), meaning that collagen-based mate-

rials such as chromium leather waste are expected to yield less biogas compared

to other proteinaceous materials. An explanation for this could be that the col-

lagen chain itself is also a substrate which needs much effort to be degraded.

Trials using pre-treated shavings reached low collagen content results within

the range of that achieved by degradation of gelatin. This indicates that the

chromium content in the biomass of up to 1.3% (measured as chromium oxide)

4 Results and Discussion

did not affect degradation of the substrates. Moreover, the pH value of the bio-

mass after digestion was very stable, always between 8 and 8.5, highlighting the

stability of the process.

The pre-treatments were important to partially replace the hydrolysis step of

the biogas production, represented by the lag-phase, and assure a reduction of

the final waste increasing degradation degree of collagen from 77.2% to about

99%. These aspects show that using these pre-treatments would be beneficial to

the production of biogas from chromium shavings. As seen in this section and

in the previous sections, pre-treated shavings in the same cluster did not signif-

icantly differ from each other and showed similar results. Therefore, from the

pre-treated shavings, only the shavings extruded dry at 100 °C were selected

for further investigations in batch trials.

4.2.1.3 Comparison of different substrate to inoculum ratios

Different substrate to inoculum ratios were tested in batch tests to investigate

the effect of the substrate load. Untreated chromium shavings, gelatin (dena-

tured and hydrolyzed collagen), shavings extruded dry at 100 °C, and extruded

offcuts were tested as substrates. The two-phase Gompertz model was used to

fit the experimental data. Figure 4.8 and Table 4.6 compare the results obtained

for these biogas trials. The error bars in Figure 4.8 represent the standard devi-

ation for the experimental data.

A plateau during production, also known as diauxie (two-phase decomposi-

tion) was verified for gelatin, shavings extruded dry at 100 °C, and extruded

offcuts, with the exception of the 0.5 S/I ratio experiment (Figure 4.8). Some au-

thors support the existence of diauxic growth in anaerobic digestion and differ-

ent theories are used to explain this. Misi and Forster (2001), Marin et al. (2010),

and Walter et al. (2016) attributed this curve to the degradation of low complex-

ity substrates at the beginning of digestion followed by a retarded degradation

of more complex substrates afterwards. Ashekuzzaman and Poulsen (2011) in-

dicated adaptation by microorganisms as being the cause of diauxie. Further-

more, this behaviour could be related to the excessive production of volatile

fatty acids during anaerobic digestion, which inhibits methane production. Lee

et al. (2010) verified a diauxic utilization of acetate to propionate, which results

4.2 Biogas Production

in a diauxie biogas production curve. Chanakya et al. (2015) related inhibition

of the anaerobic bacteria to the accumulation of volatile fatty acids. Kim and

Kim (2017) attributed diauxie to the rapid acidification caused by the accu-

mulation of volatile fatty acids.

Fig. 4.8: Cumulative biogas production and two-phase Gompertz simulation (red line) for the

untreated chromium shavings (a), gelatin (b), shavings extruded dry at 100 °C (c), and extruded

offcuts (d) with different substrate quantities in agitated bioreactors.

Trials using untreated chromium shavings as substrate did not show diauxie

due to the complexity of the material. The substrate is not pre-denatured and

time (days)

020 40 60

Biogas yield (L kg-1)

100

200

300

400

500

600

Plateau

time (days)

020 40 60

Biogas yield (L kg-1)

100

200

300

400

500

600

time (days)

020 40 60

Biogas yield (L kg-1)

100

200

300

400

500

600

Plateau

time (days)

020 40 60

Biogas yield (L kg-1)

100

200

300

400

500

600

S/I 0.5

S/I 1.5

S/I 2.5

S/I 0.1

TGM

Plateau

4 Results and Discussion

hydrolysis proceeds slowly. For complex substrates hydrolysis is the rate-lim-

iting step in the anaerobic degradation. Therefore, the production of interme-

diates is slow and they do not accumulate or cause inhibition of microorgan-

isms. A similar case is the anaerobic degradation of cellulose, a complex sub-

strate due to the presence of lignin (Tsavkelova and Netrusov, 2012). The slow

hydrolysis of cellulose even allows using higher S/I ratios, which can be found

within the range from 1.08 to 4.1 (Liew et al., 2011; Liew et al., 2012; Song et al.,

2013).

Tab. 4.6: Parameters of biogas production from the two-phase Gompertz equation.

Untreated chromium shavings

S/I

λ1 (d)

(L kg-1 d-1)

(L kg-1)

λ2 (d)

(L kg-1 d-1)

A1+A2

(L kg-1)

0.5

5.9

86.5

429.3

∞

A2 = 0.0

0.9968

1.5

6.9

35.9

294.9

∞

0.9979

2.5

6.9

21.3

196.9

A2 = 0.0

0.9965

Gelatin

S/I

λ1 (d)

(L kg-1 d-1)

(L kg-1)

λ2 (d)

(L kg-1 d-1)

A1+A2

(L kg-1)

0.5

0.8

91.0

496.4

A2 = 0.0

0.9954

1.5

0.8

32.5

333.6

25.6

11.6

507.4

0.9986

2.5

0.0

16.9

191.4

27.4

22.8

427.2

0.9980

Shavings extruded dry at 100 °C

S/I

λ1 (d)

(L kg-1 d-1)

(L kg-1)

λ2 (d)

(L kg-1 d-1)

A1+A2

(L kg-1)

0.1

0.6

102.8

445.7

24.9

3.2

608.8

0.9980

0.5

1.8

77.7

430.2

A2 = 0.0

0.9980

1.5

1.4

41.3

368.4

31.9

4.4

457.6

0.9994

2.5

1.4

12.8

125.5

23.2

12.1

377.3

0.9990

Extruded offcuts

S/I

λ1 (d)

(L kg-1 d-1)

(L kg-1)

λ2 (d)

(L kg-1 d-1)

A1+A2

(L kg-1)

0.5

0.8

36.3

294.2

A2 = 0.0

0.9988

1.5

3.2

5.5

92.2

34.0

6.3

199.1

0.9984

2.5

22.8

8.6

184.0

A2 = 0.0

0.9767

4.2 Biogas Production

Almost all conditions tested had an R2 higher than 0.99 (Table 4.6) except the

bioreactor using extruded offcuts with an S/I ratio of 2.5. Substrates which did

not show diauxie have a two-phase Grompertz equation with high values of λ2

or A2 of almost 0 L kg-1. As seen in the previous section, the batches using gelatin

as substrate started production earliest, with a lag-phase between 0 and

0.8 days. The trials with extruded shavings showed a small delay in comparison

to gelatin, with a lag-phase between 0.6 and 1.8 days. Using untreated chro-

mium shavings, the lag-phase showed values between 5.9 and 6.9 days. If the

substrate is highly complex, i.e., non-denatured, the hydrolytic stage limits the

rate of degradation and longer lag-phases can be seen. The hydrolysis time var-

ies with the substrate. In case of leather waste it is a matter of days (Deublein

and Steinhauser, 2008).

Additionally, the lag-phase for the extruded offcuts was dependent on the S/I

ratio. With an S/I ratio of 0.5 the lag-phase was very short. However, when us-

ing a higher ratio at least 3.1 days were necessary to start a very weak biogas

production. The S/I ratio of 2.5 for this substrate showed the longest lag-phase

in this work of 22.8 days. As extruded offcuts are denatured, hydrolysis should

occur quickly and the delay is probably not caused by this stage of digestion. A

cause could be the inaccessibility of the substrate due to the wet end and finish-

ing process, as seen in the trypsin assays and solubility in water results (Figures

4.3 and 4.5).

The higher the substrate to inoculum ratio the lower the maximum biogas pro-

duction rate of the first phase. Similarly, the higher the substrate to inoculum

ratio the lower the biogas formation potential (Table 4.6). As recommended in

the guideline VDI 4630 (2006), a substrate to inoculum ratio equal or lower than

0.5 should be aimed at. This S/I ratio showed a high biogas formation potential

and no inhibition. A substrate to inoculum ratio higher than 0.5 caused inhibi-

tion of the anaerobic bacteria responsible for degradation of the substrate and

transformation of it into biogas. A ratio markedly lower than 0.5 (S/I ratio of 0.1,

only tested for shavings extruded dry at 100 °C) showed improvement in the

biogas formation potential but high variance in the results. Nevertheless, a ratio

lower than 0.5 should also be avoided as higher loads make the industrial pro-

cess more economically advantageous.

4 Results and Discussion

Gelatin showed the highest biogas formation potential of all substrates tested,

followed by the extruded shavings and the untreated chromium shavings. The

extruded offcuts showed the lowest biogas formation potential. The shavings

extruded dry at 100 °C were proven to be completely denatured by DSC meas-

urements. The denaturation process breaks down the hydrogen bonds that sta-

bilize the triple helix but not necessarily tears apart the long collagen chains.

Therefore, the remaining structure is still arranged in unorganized long chains

which need to be broken down. In the first step of anaerobic digestion, the hy-

drolysis, microorganisms need energy to degrade these long chains present in

the extruded shavings. In contrast, gelatin is almost ready to start the acidogen-

esis. Even though the extruded offcuts were also completely denatured when

they underwent the same pre-treatment at higher temperatures, they showed

the lowest biogas formation potential.

With an S/I ratio of 2.5 (substrate quantity overload), untreated chromium shav-

ings and gelatin had similar biogas yields during the first phase of digestion

(A1), almost 200 L kg-1. However, in the second phase of digestion gelatin was

able to produce much more biogas generating a very pronounced diauxie. The

trials using extruded shavings as substrate also showed diauxie but reached

lower biogas yield values, about 12% less than gelatin. Trials using untreated

chromium shavings did not restart the biogas production because intermediate

products are produced slowly during hydrolysis and do not overload the next

steps in the anaerobic degradation. The extruded offcuts were degraded with

difficulty and showed a retarded biogas formation curve (VDI 4630, 2006).

Similarly, for trials with an S/I ratio of 1.5, pre-treated substrates (extruded

shavings and offcuts) and gelatin presented diauxie and untreated chromium

shavings did not restart the biogas production. Denatured substrates are

quickly hydrolyzed and overload the next steps of the anaerobic digestion at

this S/I ratio. The diauxie behaviour was more pronounced for gelatin than for

extruded shavings. Trials using extruded offcuts as substrate showed diauxie

reaching a biogas formation potential of 199 L kg-1 and trials using untreated

chromium shavings did not show diauxie.

4.2 Biogas Production

For an S/I ratio of 0.5, no diauxie curve was formed. The biogas formation po-

tential of the substrates could be reached very early in the trials, indicating that

all the potential of the inoculum to produce biogas from the substrate was used.

The same occurred with the untreated chromium shavings after a lag-phase of

six days at the beginning of anaerobic digestion.

Table 4.7 shows the results of characterization of the final biomass and the deg-

radation degree of collagen for each S/I ratio. The reactors loaded with un-

treated chromium shavings were unable to reach a complete degradation of col-

lagen. A clear relation to the S/I ratio can be seen. A higher ratio results in a

higher final collagen and organic matter content and a lower degradation de-

gree of collagen, indicating that part of the substrate remains unprocessed in

the biomass. Degradation of collagen was almost complete for all ratios tested

when extruded shavings were used. These reactors showed a degradation de-

gree of collagen similar to the degradation degree of hydroxyproline of the gel-

atin trials. Furthermore, their final organic matter contents were lower than

those for the untreated chromium shavings. Extruded offcuts showed a degra-

dation degree of collagen between 90% and 96%, values higher than those

found for the untreated chromium shavings. However, a comparison of these

values is difficult because extruded offcuts showed a lower value of collagen

content, about 50% (Table 4.1). Instead it is more reasonable to compare colla-

gen and organic matter content values in the final biomass of the reactors. The

collagen content measured in the final biomass of the reactor with extruded off-

cuts as substrate was low, lower than that for untreated chromium shavings.

However, the organic matter content in this biomass was high, similar to the

final biomass of the reactor with untreated chromium shavings.

Trials with untreated chromium shavings revealed a lower degradation of col-

lagen performance compared to the extruded shavings in all S/I ratios. The trials

with untreated chromium shavings also revealed lower values for the biogas

formation potential than the trials with pre-treated shavings, except for an S/I

ratio of 0.5 when values were very similar. The pre-treatment was important to

increase degradation of collagen. As previously seen, trials using extruded

shavings as substrate did not reach the same biogas formation potential as gel-

atin because extruded shavings are still arranged in long chains. Consequently,

4 Results and Discussion

more energy is necessary for the degradation of extruded shavings than that of

gelatin.

Tab. 4.7: Biomass characterization after digestion and degradation degree of collagen for biore-

actors testing different substrate to inoculum ratios.

Untreated chromium shavings

S/I

Organic

matter (%)*

Collagen (%)**

Degradation

degree (%)

Chromium

(%)*,***

0.5

47.2 ± 0.1

4.7 ± 0.7

77.2

0.2 ± 0.1

1.5

56.0 ± 0.1

15.6 ± 1.3

69.5

1.7 ± 0.0

2.5

64.0 ± 0.1

32.1 ± 1.3

44.8

2.2 ± 0.0

Gelatin

S/I

Organic

matter (%)*

Hydroxypro-

line (%)**

Degradation

degree (%)

Chromium

(%)*,***

0.5

46.0 ± 1.1

0.04 ± 0.01

98.8

1.5

48.8 ± 0.2

0.05 ± 0.01

99.6

2.5

49.1 ± 0.4

0.06 ± 0.00

99.6

Shavings extruded dry at 100 °C

S/I

Organic

matter (%)*

Collagen (%)**

Degradation

degree (%)

Chromium

(%)*,***

0.1

43.6 ± 0.2

0.0 ± 0.0

100

0.1 ± 0.0

0.5

45.8 ± 0.1

0.3 ± 0.0

99.1

1.3 ± 0.1

1.5

44.5 ± 0.2

0.3 ± 0.1

99.5

2.0 ± 0.2

2.5

51.3 ± 0.6

0.0 ± 0.0

100

3.6 ± 0.1

Extruded offcuts

S/I

Organic

matter (%)*

Collagen (%)**

Degradation

degree (%)

Chromium

(%)*,***

0.5

51.9 ± 0.2

1.9 ± 1.0

95.7

1.5

59.1 ± 0.6

3.3 ± 0.9

89.9

2.5

63.5 ± 0.1

2.0 ± 0.2

95.7

*Dry basis; mean ± standard deviation, n = 2

**Dry basis; mean ± standard deviation, n = 3

***Measured as chromium oxide

Source: Gomes et al., 2019a.

Results indicate that extruded offcuts are much more complex than untreated

chromium shavings. As seen before in the degradation by trypsin, the wet end

4.2 Biogas Production

and finishing process in the leather industry made the material even less inac-

cessible for the production of biogas than untreated chromium shavings. The

organic matter content, which comprises not only collagen but also dyes, re-

tanning agents, pigments, and other polymers (Covington, 2009), was very

high. The structure of the leather offcuts can be denatured by extrusion

pre-treatment, and detectable collagen in the final biomass is almost eliminated,

but these high quantities of organic matter remain unprocessed and the biogas

formation potential is the lowest among the substrates tested.

The chromium content in the collected biomass samples reached values of up

to 3.6% and degradation of collagen was not affected indicating that chromium

at this concentration does not affect the bacteria in the anaerobic digestion.

The effective use of chromium shavings to produce biogas depends on many

parameters. One of the most important parameters is the direct comparison of

the biogas formation potential for different substrates. Table 4.8 shows the bio-

gas formation potential (A) of several substrates used in biogas plants. The bio-

gas formation potential values were calculated from literature data (FNR, 2012)

in order to compare the values to the biogas formation potential found for

pre-treated chromium shavings in this work.

On average, trials using pre-treated chromium shavings with an S/I ratio of 0.5

showed a biogas formation potential of 441 L kg-1. This value is comparable with

those found for manure, which represented 44.5% (by weight) of the substrate

input in German biogas plants in 2016 (DBFZ Betreiberbefragung Biogas, 2017).

Compared with energy crops, pre-treated chromium shavings showed a lower

biogas formation, however, tannery waste has other advantages. The substrate

costs can account for up to 50% of the total costs in a biogas plant operated

mainly using energy crops as substrate (FNR, 2012). The use of waste to pro-

duce biogas does not entail production costs and reduces disposal of waste.

Furthermore, the production of energy crops is associated with other problems,

such as competition with the food production, land use, and use of monocul-

tures.

4 Results and Discussion

Tab. 4.8: Biogas formation potential of different substrates.

Manure

Energy crops

Substrates from pro-

cessing industry

Substrate

(L kg-1)*

Substrate

(L kg-1)*

Substrate

(L kg-1)*

Cattle

slurry

313

Maize silage

638

Spent grains

684

Pig slurry

583

Whole-crop

cereal silage

606

Cereal vinasse

691

Cattle

dung

400

Green rye

silage

667

Potato vinasse

667

Poultry

manure

467

Cereal

grains

735

Fruit pomace

632

Horse ma-

nure

300

Grass silage

571

Rapeseed cake

825

Prunings and grass

clippings

Sugar beet

628

Potato pulp

684

Fodder beet

625

Potato juice

1962

Substrate

(L kg-1)*

Sunflower

silage

533

Pressed sugar

beet pulp

298

Prunings

and clip-

pings

1667

Sudan grass

521

Molasses

421

Sweet sor-

ghum

539

Apple pomace

481

Green ryeb

591

Grape pomace

680

*average values, kg of organic dry matter

Source: FNR, 2012.

The substrates from the processing industry show a wide range of biogas for-

mation potential. Their use in biogas plants is not very attractive, they repre-

sented only 2.4% (by weight) of the substrate input in German biogas plants in

2016 (DBFZ Betreiberbefragung Biogas, 2017). The reasons vary with the sub-

strate. For instance, brewer's grains and sugar beet are also used in the food

industry and as animal feed, and vinasse has a low dry matter content and in

some cases is not worth transporting (FNR, 2012). Prunings and grass clippings

from the maintenance of parks and green verges show a biogas formation po-

4.2 Biogas Production

tential considerably higher than that of pre-treated chromium shavings, how-

ever their seasonal and disperse production makes their use in biogas plants

difficult and they are normally composted (FNR, 2012).

The S/I ratio of 0.5 showed the best results amongst the ratios studied for all the

substrates studied (pre-treated or not) and should be aimed at when using these

collagen-based materials as substrate for the production of biogas. The sub-

strates studied showed a high biogas formation potential and no diauxie inhi-

bition in the anaerobic digestion using this S/I ratio. Furthermore, the pre-

treated chromium shavings reveal a biogas formation potential comparable to

that of other substrates commonly used. Their advantages are the availability

in large quantities, no costs, and no significant use as a by-product. Their use as

a substrate for biogas production would even reduce costs with disposal of

waste in tanneries. A disadvantage of their use as substrate to produce biogas

is the content of chromium in this material because it complicates the use of the

final biomass as a fertilizer. Extraction and recycling of chromium from the final

biomass produced in the reactors must be further studied.

4.2.1.4 Diauxie investigation

Hitherto, the diauxie behaviour in anaerobic digestion has not been completely

clarified. Following the theory that inhibition is caused by accumulation of in-

termediate products, supported by Chanakya et al. (2015) and Kim and Kim

(2017), the volatile fatty acids produced along diauxie were measured. This

would mean that there is an imbalance between hydrolysis and methanogene-

sis. As the substrate is denatured, it can be quickly hydrolyzed to produce in-

termediate products. However, methanogenesis is not accelerated and these in-

termediate products accumulate thus inhibiting methanogenic archaea. This

only occurs for an S/I ratio higher than 0.5 and denatured substrates (Gomes et

al. 2019a).

To investigate this, trials of the previous section in which diauxie was verified

were repeated. Bioreactors for gelatin, shavings extruded dry at 100 °C, and

extruded offcuts (S/I ratio of 1.5) were started seven times in triplicate simulta-

neously. Every triplicate was terminated at different times in the anaerobic di-

4 Results and Discussion

gestion to collect a sample of biomass for analytical investigations. The concen-

trations of volatile fatty acids in each batch of the seven samples were deter-

mined by high-performance liquid chromatography. Diauxic growth curve,

sampling time, and volatile fatty acid concentration of the samples are shown

in Figure 4.9. The error bars represent the standard deviation for the experi-

mental data.

The pH values were stable during anaerobic digestion, increasing only slowly

from 7.2 to 8.4, and causing no problems to the process. Even though the volatile

fatty acid concentration increased, the pH value was stable. Ashekuzzaman and

Poulsen (2011), and Misi and Forster (2001) also verified pH values slightly

above neutral in anaerobic digestion batches with diauxic growth. However,

the volatile fatty acid concentration during the batches was not measured.

In their paper, Ramsay and Pullammanappallil (2001) reviewed the products

resulting from fermentation of amino acids using the Stickland reaction. Based

on that, it was possible to define the products mainly expected for fermentation

of the amino acids of bovine hide collagen. The most relevant volatile fatty acids

produced in the three reactors were acetic acid, propionic acid, and isovaleric

acid, which is consistent with the amino acid composition of the substrate. The

most relevant amino acids present in these collagen chains are glycine, which is

fermented to acetic acid, and proline, which is fermented to acetic acid, propi-

onic acid, and valeric acid. Most of the other amino acids present in the collagen

chains are fermented to acetic acid but some are also fermented to propionic

acid and butyric acid (Ramsay and Pullammanappallil, 2001), which is also pre-

sent in the reactors in low concentrations.

Fig. 4.9: Reproduction of the biogas yield, sampling time, and volatile fatty acid (VFA) concen-

trations present in samples collected for gelatin (a), shavings extruded dry at 100 °C (b), and

extruded offcuts (c) in agitated bioreactors (source: Gomes et al., 2019a).

time (days)

020 40 60

VFA concentration (g L-1)

Biogas yield (L kg-1)

100

200

300

400

500

600

Acetic acid

Propionic acid

Isobutyric acid

Butyric acid

Isovaleric acid

Total VFA

Biogas

234 5 6

time (days)

020 40 60

VFA concentration (g L-1)

Biogas yield (L kg-1)

100

200

300

400

500

600

Acetic acid

Propionic acid

Isobutyric acid

Butyric acid

Isovaleric acid

Total VFA

Biogas

23456

time (days)

020 40 60

VFA concentration (g L-1)

Biogas yield (L kg-1)

100

200

300

400

500

600 Acetic acid

Propionic acid

Isobutyric acid

Butyric acid

Isovaleric acid

Total VFA

Biogas

234 5 6

For the extruded shavings (Figure 4.9b), biogas production was suppressed in

case the volatile fatty acid concentration was high. Chanakya et al. (2015) ob-

served the same behaviour. This means that during acidogenesis a large

amount of these acids is produced, which are intermediate products for gener-

ating biogas. However, the bacteria are not able to process all of the produced

acids accumulating. As these acids are consumed and transformed into sub-

strates to produce methane during acetogenesis, their concentration is reduced

and methane production resumed. Lee et al. (2010) also measured the volatile

fatty acid concentration during anaerobic digestion and the same behaviour

was verified. The volatile fatty acids were mainly composed of acetic and pro-

pionic acids. Propionic acid showed a concentration higher than 1.0 g L-1 in the

second and third sample, a concentration known to cause inhibition (Kaiser et

al., 2008).

Gelatin (Figure 4.9a) showed the same behaviour as the extruded shavings.

However, the volatile fatty acid concentration was already very high in the first

biomass sample because gelatin started producing biogas faster than the ex-

truded shavings. The chains of gelatin partly hydrolyzed start production of

volatile fatty acid immediately after substrate and inoculum are in contact. The

propionic acid concentration was not as high as observed for the trial with ex-

truded shavings, but for the first, second, and third biomass samples collected

it was higher than 0.3 g L-1, a concentration discussed to disturb anaerobic di-

gestion (Deublein and Steinhauser, 2008). The isobutyric acid concentration was

higher than the inhibition concentration of 0.5 g L-1 (Kaiser et al., 2008) in the

third biomass sample collected.

The reactor containing the extruded offcuts (Figure 4.9c) showed high volatile

fatty acid concentrations (about 2 g L-1) during all digestion stages. The volatile

fatty acid results indicate that there could be inhibition by these intermediate

products. Furthermore, the highest volatile fatty acid concentration (2.5 g L-1)

was verified during the diauxic plateau-phase. Unlike the extruded shavings,

this reactor shows a high concentration of volatile fatty acid until the end of

digestion. A high propionic acid concentration of 0.4 g L-1 was also verified at

the end of the process for this reactor. These results could indicate that the pro-

duction of biogas reaches its end because of inhibition and that, consequently,

4.2 Biogas Production

part of the available organic matter found in the biomass collected was not

transformed into biogas.

The anaerobic digestion of collagen was previously classified as biphasic by

Lalitha et al. (1994). In their paper, a trial with substrate overload showed two

peaks of biogas production. The first peak was at the beginning of digestion

when collagen was being hydrolyzed and acidogenesis started. This peak pro-

duced a biogas mainly composed of CO2. As the system was overloaded, there

was a rapid increase in the volatile fatty acid concentration, which is detri-

mental to the methanogenic population. The second peak corresponded with

the beginning of the methanogenesis, when the volatile fatty acid concentration

dropped. These results correspond with those represented in this paper.

Diauxie only occurs in trials with substrates already denatured (gelatin, ex-

truded shavings, and extruded offcuts) and with an S/I ratio higher than 0.5,

because these substrates are quickly hydrolyzed to form volatile fatty acids dur-

ing acidogenesis. However, the bacteria are not able to process all of the inter-

mediate products that fast, which means that they start to accumulate and bio-

gas production drops. Production continues when the volatile fatty acid con-

centration drops. This indicates that when using denatured collagen as sub-

strate, methanogenesis is the rate-limiting step during anaerobic degradation

instead hydrolysis.

Accumulation of volatile fatty acid produced during acidogenesis could be the

cause of the diauxie behaviour of the biogas production curves. It is still neces-

sary to find out if accumulation of volatile fatty acid is responsible for inhibition

of methanogenesis or a consequence of inhibition.

Bovine hide collagen is a nitrogen-rich substrate with a C/N ratio of 3.1, calcu-

lated with the basic elementary formula (The UniProt Consortium). Therefore,

inhibition by ammonia is possible, which is a product resulting from fermenta-

tion of amino acids (Nisman, 1954). The ammonia content in the reactors was

monitored during digestion. Figure 4.10 shows the results. The sample point 0

represents the ammonia content of the inoculum.

4 Results and Discussion

Fig. 4.10: Concentration of ammonia found in the biomass collected at different reaction times

for gelatin (a) and shavings extruded dry at 100 °C (b) in agitated bioreactors.

time (days)

020 40 60

Ammonia concentration (mg NH4-N L-1)

0,00

0,02

0,04

0,06

0,08

0,10

1234567

time (days)

020 40 60

Ammonia concentration (mg NH4-N L-1)

0,00

0,02

0,04

0,06

0,08

0,10

1234567

4.2 Biogas Production

For gelatin and extruded shavings, ammonia is present in the biomass in very

low concentrations, always below inhibition level, which starts at a concentra-

tion of 1.5 to 3.0 g NH4-N L-1 (Drosg et al., 2013), and causes no problem to the

anaerobic digestion. Ammonia did not accumulate in the bioreactors probably

because of its utilization for generating bacterial biomass. Lalitha et al. (1994)

also verified low levels of ammonia during anaerobic digestion of collagen. For

the extruded offcuts determination of the ammonia concentration was not pos-

sible due to the presence of dyes in the biomass, which interfered the photomet-

ric measurement.

Table 4.9 shows the characterization of the biomass sample collected in every

sampling point for the three substrates studied. Biomass was analyzed regard-

ing its collagen or hydroxyproline content and organic matter content, and the

degradation degree of its collagen or hydroxyproline was estimated.

At the first sampling point the trial using gelatin as substrate already showed a

very high degradation degree of hydroxyproline, 97%, while the trial with ex-

truded shavings showed a degradation degree of collagen of about 24% and

that with extruded offcuts 8%. Even though extrusion of the collagen structure

eases the onset of digestion, a short adaptation period is necessary to start di-

gestion of the previously extruded samples. The short hydrolyzed chains of gel-

atin can start acidogenesis almost immediately. From sampling point 2 on, the

hydroxyproline of gelatin was almost completely degraded showing a degra-

dation degree of about 99% and the degradation degree of the collagen from the

extruded shavings increased slowly from 96% to about 99%. For the extruded

offcuts, degradation of collagen resulted in a high value (90%) only with regard

to the last sample point.

As previously seen for the other batch reactors, degradation of collagen is al-

most complete at the end of the digestion. The organic matter is also reduced

but not completely removed. The organic matter present at the end of digestion,

which is not identified as collagen, has to be further investigated. Lalitha et al.

(1994) verified a reduction of 85% of collagen content but a decrease of only 65%

of organic matter. The volatile fatty acid concentration is very low at the end of

digestion and, therefore, could not be the cause of this difference.

4 Results and Discussion

Tab. 4.9: Biomass characterization during digestion regarding its organic and collagen or hy-

droxyproline content and biogas yield for bioreactors investigating diauxie.

Gelatin

Sample

Time

(days)

Organic

matter (%)*

Hydroxyproline

(%)**

Degradation

degree (%)

Biogas yield

(L kg-1)**

0.0

73.3***

0.9

61.2 ± 0.0

0.23 ± 0.02

97.3

65 ± 2

6.8

55.5 ± 0.1

0.05 ± 0.00

99.5

168 ± 6

14.8

52.3 ± 0.1

0.05 ± 0.01

99.5

317 ± 5

25.7

52.1 ± 0.5

0.00 ± 0.01

100.0

334 ± 15

32.7

51.0 ± 0.2

0.08 ± 0.01

99.2

391 ± 9

39.7

50.9 ± 0.3

0.07 ± 0.01

99.4

489 ± 14

60.7

48.8 ± 0.2

0.05 ± 0.01

99.6

509 ± 15

Shavings extruded dry at 100 °C

Sample

Time

(days)

Organic

matter (%)*

Collagen (%)**

Degradation

degree (%)

Biogas yield

(L kg-1)**

0.0

66.9***

31.6***

2.0

66.8 ± 0.2

23.9 ± 1.3

27.6

42 ± 2

6.9

55.2 ± 0.0

1.6 ± 0.0

96.3

245 ± 11

11.8

52.3 ± 0.1

0.9 ± 0.1

98.0

360 ± 7

23.0

51.2 ± 0.2

0.7 ± 0.0

98.5

398 ± 6

32.9

50.1 ± 0.5

0.7 ± 0.2

98.6

428 ± 15

46.7

47.5 ± 0.3

0.7 ± 0.1

98.7

483 ± 15

67.7

46.6 ± 0.1

0.5 ± 0.0

99.1

494 ± 3

Extruded offcuts

Sample

Time

(days)

Organic

matter (%)*

Collagen (%)**

Degradation

degree (%)

Biogas yield

(L kg-1)**

0.0

71.1***

22.2***

1.9

72.8 ± 1.3

20.1 ± 1.3

8.4

6 ± 2

11.8

68.1 ± 1.1

60 ± 15

18.8

67.5 ± 0.7

12.1 ± 4.9

53.2

113 ± 27

28.9

63.7 ± 0.3

12.0 ± 3.0

57.9

120 ± 11

39.7

63.0 ± 0.5

15.2 ± 3.3

47.1

124 ± 20

48.8

61.5 ± 0.0

9.8 ± 3.9

67.6

192 ± 14

69.8

59.1 ± 0.6

3.3 ± 0.9

89.9

203 ± 4

*Dry basis; mean ± standard deviation, n = 2

**Dry basis; mean ± standard deviation, n = 3

***Mean value for all reactors at the start of digestion

4.2 Biogas Production

As mentioned before, amino acids in anaerobic digesters are mainly fermented

in pairs through the Stickland reaction (Schink and Stams, 2013; Ramsay and

Pullammanappallil, 2001). The reaction requires one hydrogen donor and one

hydrogen acceptor to occur. One third of the amino acid composition in colla-

gen is made up of glycine. The amino acids proline and hydroxyproline are also

present in high quantities. These amino acids are known to be hydrogen accep-

tors, consequently there is a shortage of amino acids in the medium. Specifi-

cally, for bovine hide collagen, the shortage is at least 11.5% considering that all

other amino acids act as hydrogen donors.

In their paper, Lalitha et al. (1994) also measured collagen through determina-

tion of the amino acid hydroxyproline. It is possible that hydroxyproline is fer-

mented and, therefore, cannot be measured while other amino acids remain un-

processed. The presence of unprocessed amino acids could be the cause of the

difference between collagen content and organic matter content. Another pos-

sibility is the dehydroxylation of hydroxyproline to proline, which would not

be detected by the method. A large quantity of other amino acids from the col-

lagen chains could be unprocessed in the biomass but the method is unable to

detect them.

The seven biomass samples collected during anaerobic digestion were also an-

alyzed regarding their chemical oxygen demand (COD) as another indicator of

organics in the biomass. Results for the three substrates tested are shown in

Figure 4.11.

4 Results and Discussion

Fig. 4.11: COD of the biomass samples collected at different reaction times for gelatin (a), shav-

ings extruded dry at 100 °C (b), and extruded offcuts (c) in agitated bioreactors.

time (days)

020 40 60

COD (g L-1)

time (days)

020 40 60

COD (g L-1)

time (days)

020 40 60

COD (g L-1)

1234567

234567

1234567

4.2 Biogas Production

Similarly to the organic matter results, COD values also decreased slowly dur-

ing digestion, indicating effective destruction of the substrates. An exception

were the biomass samples collected from the extruded offcuts reactors. The

COD values at the beginning of anaerobic digestion (sample 1) were lower than

those for the other two substrates. Presumably this is observed because the sub-

strate is very inhomogeneous. The extruded offcuts were not completely disin-

tegrated in the biomass. Small pieces of the substrate could still be seen. These

pieces of substrate were not used for the COD analyses resulting in the lower

COD values. However, the biomass of the fourth reactor (sample 4) showed the

highest COD value among all reactors studied, indicating that a small piece of

undisintegrated substrate was present in the analysed aliquot. COD values

were also constant for most of the anaerobic digestion confirming that the or-

ganic part of the substrate, which remains unprocessed for the extruded offcuts,

was larger.

A high concentration of volatile fatty acid and the excess of propionic acid could

cause inhibition of the methanogenic archaea and the diauxic production of bi-

ogas in bioreactors with overload of denatured substrate. However, it is also

possible that inhibition of the methanogenesis has a different reason and accu-

mulation of volatile fatty acid is a consequence of inhibition rather than its

cause.

To evaluate the inhibitory effect of accumulation of volatile fatty acid during

anaerobic digestion, bioreactors with gelatin as substrate were injected with dif-

ferent volatile fatty acids. The S/I ratio used, approximately 0.5, is known not to

cause diauxie, and the acids were injected when the biogas production was ex-

pected to increase. Three different volatile fatty acids were evaluated, acetic

acid, propionic acid, and isobutyric acid. The acid concentrations tested in-

volved the inhibitory concentration found in the literature for these acids (Deu-

blein and Steinhauser, 2008; Kaiser et al., 2008), a lower concentration, and a

higher concentration. Acids were injected on the second day of digestion. Ad-

ditionally, the same quantity of acid was injected on the fifth day of digestion.

Figure 4.12 shows the biogas production curves obtained from the experiments,

the injection time, the injected acid, and its expected concentration in the biore-

actor. The error bars represent the standard deviation for the experimental data.

4 Results and Discussion

The pH values of the biomass at the end of anaerobic digestion were slightly

above neutral (between 7.8 and 7.9), apparently causing no problems to the bi-

ogas production process. However, it was not possible to measure the pH value

during anaerobic digestion and at the moment of volatile fatty acid injection

because there would be ingress of oxygen and this would interrupt the anaero-

bic digestion. The injection of volatile fatty acid could have a temporary effect

on the pH value.

All reactors with injected volatile fatty acids produced more biogas than the

reference reactor in which no volatile fatty acid was injected. The higher the

volatile fatty acid concentration the higher the biogas formation potential. This

contradicts the theory that accumulation of volatile fatty acids inhibits the bio-

gas production forming diauxie. It is rather that the acids, which are described

to have an inhibitory effect, are quickly metabolized by the anaerobic bacteria

and transformed into biogas without disturbing anaerobic digestion. This leads

to the conclusion that the high concentration of volatile fatty acid and the excess

of propionic acid verified in the diauxie trials (Figure 4.9) are a consequence of

inhibition of the methanogenic archaea rather than its cause. Therefore, accu-

mulation of volatile fatty acid is not the cause of the diauxie curves verified for

the substrates studied.

Volatile fatty acids were not detected in the final biomass of the bioreactors.

Photometric analyses were performed using a Spectroquant cell test which re-

vealed an acid concentration below the measuring range (< 100 mg L-1). These

results indicate that the injected acid was completely transformed into biogas,

and they also support the findings previously obtained by other authors. Breure

et al. (1986) also studied the addition of volatile fatty acids in bioreactors using

gelatin as substrate and concluded that the volatile fatty acids were not inhibi-

tory to the degradation of gelatin. Franke-Whittle et al. (2014) investigated the

effect of different levels of volatile fatty acid in anaerobic digesters on the meth-

anogenic archaea and concluded that a high concentration of these acids had no

significant effect on the community of methanogenic archaea.

4.2 Biogas Production

Fig. 4.12: Cumulative biogas production for the gelatin in the bioreactors with added acetic acid

(a), propionic acid (b), and isobutyric acid (c).

time (days)

0 2 4 6 8 10 12 14 16

Acetic acid concentration (g L-1)

Biogas yield (L kg-1)

200

400

600

800

1000

1200

Biogas yield - Acetic acid 2 g L-1

Biogas yield - Acetic acid 3 g L-1

Biogas yield - Acetic acid 4 g L-1

Biogas yield - No added acid

Concentration - Acetic acid 2 g L-1

Concentration - Acetic acid 3 g L-1

Concentration - Acetic acid 4 g L-1

time (days)

0 2 4 6 8 10 12 14 16

Propionic acid concentration (g L-1)

Biogas yield (L kg-1)

200

400

600

800

1000

1200

Biogas yield - Propionic acid 0.2 g L-1

Biogas yield - Propionic acid 0.3 g L-1

Biogas yield - Propionic acid 1 g L-1

Biogas yield - Propionic acid 2 g L-1

Biogas yield - No added acid

Concentration - Propionic acid 0.2 g L-1

Concentration - Propionic acid 0.3 g L-1

Concentration - Propionic acid 1 g L-1

Concentration - Propionic acid 2 g L-1

time (days)

0 2 4 6 8 10 12 14 16

Isobutyric acid concentration (g L-1)

Biogas yield (L kg-1)

200

400

600

800

1000

1200

Biogas yield - Isobutyric acid 0.3 g L-1

Biogas yield - Isobutyric acid 0.5 g L-1

Biogas yield - Isobutyric acid 1 g L-1

Biogas yield - No added acid

Concentration - Isobutyric acid 0.3 g L-1

Concentration - Isobutyric acid 0.5 g L-1

Concentration - Isobutyric acid 1 g L-1

4 Results and Discussion

The ammonia content in the reactors was analyzed photometrically in the final

biomass using Spectroquant cell tests. Ammonia was below the measuring

range (< 0.01 mg L-1) in the final biomass of all bioreactors indicating its use for

generating bacterial biomass and probably not causing any inhibition during

anaerobic digestion.

Table 4.10 shows the characterization of the final biomass collected at the end

of digestion. Biomass was analyzed regarding its hydroxyproline content and

organic matter content and the hydroxyproline degradation was estimated.

Tab. 4.10: Biomass characterization during digestion, degradation degree of hydroxyproline,

and biogas yield for bioreactors investigating diauxie.

Injection of acetic acid

Acid conc.

(g L-1)

Organic

matter (%)*

Hydroxyproline

(%)*

Degradation

degree (%)

Biogas formation

potential (L kg-1)

50.7 ± 0.7

0.05 ± 0.01

98.2

1156 ± 30

50.1 ± 0.2

0.05 ± 0.00

98.2

1042 ± 8

50.4 ± 0.2

0.06 ± 0.00

97.8

852 ± 25

Injection of propionic acid

Acid conc.

(g L-1)

Organic

matter (%)*

Hydroxyproline

(%)*

Degradation

degree (%)

Biogas formation

potential (L kg-1)

50.6 ± 0.3

0.05 ± 0.00

98.2

858 ± 10

50.4 ± 0.2

0.05 ± 0.01

98.2

683 ± 12

0.3

49.9 ± 0.3

0.06 ± 0.01

98.3

576 ± 8

0.2

49.3 ± 0.3

0.02 ± 0.00

99.6

522 ± 10

Injection of isobutyric acid

Acid conc.

(g L-1)

Organic

matter (%)*

Hydroxyproline

(%)*

Degradation

degree (%)

Biogas formation

potential (L kg-1)

49.5 ± 0.0

0.04 ± 0.00

98.7

729 ± 49

0.5

49.6 ± 0.5

0.02 ± 0.01

99.2

618 ± 20

0.3

49.9 ± 0.2

0.03 ± 0.01

99.1

568 ± 34

No added acid

Acid conc.

(g L-1)

Organic

matter (%)*

Hydroxyproline

(%)*

Degradation

degree (%)

Biogas formation

potential (L kg-1)

49.8 ± 0.3

0.01 ± 0.00

99.6

489 ± 12

*Dry basis; mean ± standard deviation, n = 2

4.2 Biogas Production

Even though the biogas formation potential increases substantially for the bio-

reactors with added acid, results for organic matter, hydroxyproline content,

and degradation degree of hydroxyproline show very similar values with re-

gard to the reference trial, with no added acid. Generally, a slight increase in

organic matter and hydroxyproline content can be seen for the bioreactors with

a high concentration of added acid, which could indicate that anaerobic bacteria

would preferably use the injected volatile fatty acid rather than degrade the

substrate. However, the increase was very small indicating that the injection of

acid did not affect the degradation of gelatin.

Results in this section show that diauxie during anaerobic digestion of collagen-

based materials only occurs if the substrates were already denatured with a pre-

treatment and if there is a substrate overload (S/I ratio higher than 0.5). In this

case, the rate-limiting step in the anaerobic degradation is assumed to be meth-

anogenesis rather than hydrolysis. Moreover, it was proved that accumulation

of volatile fatty acid is a consequence of inhibition of methanogenesis rather

than its cause. The exact inhibitor or inhibitors of the methanogenic archaea re-

main to be determined.

4.2.1.5 Repeatability of biogas batch trials

To control the variation of the experiments, the same batch trial was performed

in two different occasions for comparison. The inoculum used was collected at

different days and the gelatin prepared as substrate for the trials was from dif-

ferent batches. These trials can also be seen in previous sections (Sections 4.2.1.3

and 4.2.1.4) and use almost the same S/I ratio. The two-phase Gompertz model

was used to fit the experimental data. Figure 4.13 and Table 4.11 compare the

results obtained for these biogas trials.

4 Results and Discussion

Fig. 4.13: Cumulative biogas production and two-phase Gompertz simulation (red line) for the

gelatin in two different trials with the same conditions.

Tab. 4.11: Parameters of biogas production from the two-phase Gompertz equation.

Section

λ1 (d)

(L kg-1 d-1)

(L kg-1)

λ2 (d)

(L kg-1 d-1)

(L kg-1)

4.2.1.3

0.8

88.5

505.4

0.0

0.9960

4.2.1.4

1.0

85.0

497.3

0.0

0.9969

Both trials had R2 higher than 0.99 (Table 4.11) and did not show diauxie, which

can be seen on the A2 value of almost 0 L kg-1. Results show that the biogas

production curves of the two different batches present very similar behaviour.

Furthermore, the parameters from the Gompertz model are very similar. This

indicates that the trials can easily be repeated at different occasions and the re-

sults obtain from different batches can be compared and trusted.

4.2.2 Pilot scale: continuous tests

The scale-up of the process was necessary to have a better understanding of the

method in an industry-like environment. Trials were performed in a continuous

time (days)

0 2 4 6 8 10 12 14 16 18

Biogas yield (L kg-1)

100

200

300

400

500

600

Section 4.2.1.3.

Section 4.2.1.4.

TMG

4.2 Biogas Production

reactor using untreated chromium shavings, shavings extruded dry at 100 °C,

and a mixture of equal amounts of shavings treated hydrothermally at 140 °C

and 150 °C. The mixture of shavings treated hydrothermally was used due to

the large amounts of substrate necessary in continuous trials and because of the

similarity between the two substrates. However, samples treated hydrother-

mally have a high quantity of water, which would excessively increase the

working volume of the reactor. Therefore, it was necessary to dry and manually

grind the samples before feeding, a process that was not needed for the batch

trials with the same substrate.

4.2.2.1 Productivity of the reactors

A low loading rate of 0.5 g kg-1 d-1 was used to initiate the experiments. Conse-

quently, the loading rate was raised by approximately 0.5 units every time the

daily methane production reached a plateau or decreased. Figure 4.14 shows

the time plot of digestion for the substrates studied.

Drops in the daily methane production, indicating technical problems, can be

seen on the seventh day of digestion for the reactor fed with extruded shavings

(Figure 4.14b) and on the 21st day for the reactor fed with shavings treated hy-

drothermally (Figure 4.14c). The former was caused by an oxygen infiltration

on the fifth day due to the rupture of the dip tube used to feed the reactor; the

latter was a consequence of an agitation failure, which resulted in uneven heat-

ing of the reactor. Even though there is a drop in the methane production, the

system recovered its former stability and appeared to function normally after a

few days. The reactor fed with shavings treated hydrothermally needed two

days only to recover, and the reactor fed with extruded shavings needed five

days, that is more time to recover probably because the ingress of oxygen oc-

curred rather early in the digestion process. The daily methane production be-

came very unstable for the extruded shavings when using loading rates higher

than 1.4 g kg-1 d-1 and for the shavings treated hydrothermally for loading rates

of 2.0 g kg-1 d-1.

4 Results and Discussion

100

Fig. 4.14: Time plot of continuous anaerobic digestion of the untreated chromium shavings (a),

shavings extruded dry at 100 °C (b), and shavings treated hydrothermally at 140 °C and 150 °C

time (days)

010 20 30 40 50 60

Daily CH4 production (L kg-1 d-1)

0,0

0,1

0,2

0,3

0,4

0,5

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Daily CH4 production

Loading rate

time (days)

010 20 30 40 50 60

Daily CH4 production (L kg-1 d-1)

0,0

0,1

0,2

0,3

0,4

0,5

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

time (days)

010 20 30 40 50 60

Daily CH4 production (L kg-1 d-1)

0,0

0,1

0,2

0,3

0,4

0,5

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

4.2 Biogas Production

101

The production of a second batch of shavings extruded dry at 100 °C was nec-

essary to feed the reactor starting on the 42nd day. This second batch of the sub-

strate was not completely denatured because it was not possible to use the same

extruder. DSC results still showed a denaturation enthalpy of 12.6 J g-1 com-

pared to 61.8 J g-1 for untreated chromium shavings and 0 J g-1 for gelatin.

Pre-treatment allowed the use of a higher loading rate and increased the maxi-

mum daily methane production. The trial with untreated chromium shavings

(Figure 4.14a) had to be stopped at a loading rate of 1.5 g kg-1 d-1 but the ex-

truded shavings (Figure 4.14b) could be tested up to a loading rate of

1.9 g kg-1 d-1 and the shavings treated hydrothermally (Figure 4.14c) up to

2.0 g kg-1 d-1. The untreated chromium shavings reached the maximum daily

methane production, 0.41 L kg-1 d-1, on the 31st day of digestion with a loading

rate of 1.0 g kg-1 d-1. The extruded shavings reached a higher value,

0.45 L kg-1 d-1, on the 52nd day of digestion, but with a loading rate of

1.4 g kg-1 d-1. Similarly, for the shavings treated hydrothermally, the maximum

daily methane production was 0.45 L kg-1 d-1 on the 48th day of digestion with a

loading rate of 1.5 g kg-1 d-1.

The theoretical methane yield calculated using the basic elementary formula of

bovine hide collagen and Equation 2.6 was found to be 393 L kg-1 (in organic

dry matter of the substrate). Considering the total substrate added to each reac-

tor it is possible to calculate how much of the maximum capacity of the sub-

strate to produce methane was used. The trial using untreated chromium shav-

ings as a substrate produced 70% of the substrate capacity added. The trial us-

ing extruded shavings produced 66% of the capacity. And finally, the trial using

shavings treated hydrothermally produced 70% of the substrate capacity. Re-

sults are in accordance with literature were a conversion rate between 50 and

70% for proteins is expected (VDI 4630, 2006). It was found that the continuous

process reached higher conversion rates for all substrates studied when com-

pared with batch results. Agustini et al. (2018) studied the scale-up of anaerobic

digestion of chromium shavings in batch tests and concluded that a larger scale

can increase biogas yields.

4 Results and Discussion

102

Fig. 4.15: Quality of the biogas produced during digestion of chromium shavings (a), shavings

extruded dry at 100 °C (b), and shavings treated hydrothermally at 140 and 150 °C (c).

time (days)

010 20 30 40 50 60

Percentage of gas (%)

100

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

time (days)

010 20 30 40 50 60

Percentage of gas (%)

100

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Methane

Carbon dioxide

Oxygen

Others

Loading rate

time (days)

010 20 30 40 50

Percentage of gas (%)

100

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

4.2 Biogas Production

103

Figure 4.15 shows the quality of the biogas produced in the reactors. The pro-

duction of methane was low at the beginning of digestion. The composition of

the produced biogas showed less than 40% of methane until the third day of

digestion for the reactor fed with untreated chromium shavings, until the fifth

day for the extruded shavings reactor, and until the second day for the reactor

using shavings treated hydrothermally. After that, methanation increases

reaching its highest level of about 68% for the untreated chromium shavings,

the extruded shavings, and the shavings treated hydrothermally. The reactor

fed with extruded shavings showed the longest delay in methanation among

the substrates studied. This was probably caused by the ingress of oxygen at

the beginning of the anaerobic digestion trial.

A high level of oxygen in the produced biogas was registered for all substrates

studied at the beginning of digestion, about 7%. This value drops during the

first days of digestion as it is consumed by facultative anaerobic bacteria in the

hydrolysis process. Due to the ingress of oxygen in the reactor fed with ex-

truded shavings, oxygen only dropped after the ninth day. The presence of a

small quantity of oxygen during digestion was verified for all continuous trials.

This is probably oxygen from the change of gasbags and does not affect the an-

aerobic digestion.

The biogas quality was slightly reduced with every rise of the loading rate by

0.5 units. The only exception was the rise to a loading rate of 1.0 g kg-1 d-1 for

the reactor fed with shavings treated hydrothermally. As previously seen, high

loads of substrate can cause inhibition of methanogenesis and, in this case, this

is reflected in the quality of the biogas produced.

Trials for all substrates showed a reduction of biogas quality for the loading rate

at which they had their maximum daily methane production. Apart from that,

the quality values were stable. It was possible to produce more biogas by in-

creasing the load, however, methanogenesis started to be inhibited. At the end

of anaerobic digestion, the quality dropped indicating a failure of digestion.

No studies covering the anaerobic digestion of chromium leather waste in con-

tinuous trials could be found for comparison. Some authors studied the diges-

tion of tannery waste in continuous (López et al., 2015a and 2015b) or semi-

4 Results and Discussion

104

continuous mode (Berhe and Leta, 2018; Kameswari et al., 2015; Zupančič and

Jemec, 2010) but the material was always collected from tanneries prior to the

tanning step and this is known to be used already on industrial scale (Schu-

berth-Roth, 2013). These studies mainly used fleshings as substrate, which are

also not suitable for comparison due to a high fat content.

4.2.2.2 Inhibition of digestion

In order to monitor the reactor stability and possible inhibitions, biomass sam-

ples were collected weekly and analyzed regarding their concentration of vola-

tile fatty acids. Figure 4.16 shows the results for the three continuous trials. The

error bars represent the standard deviation for the experimental data.

Volatile fatty acids are important intermediate products resulting from the aci-

dogenesis phase during anaerobic digestion. However, they could cause inhi-

bition and failure of the digester. To avoid failure the total concentration of vol-

atile fatty acids should be lower than 4 g L-1, the acetic acid concentration

should be lower than 3 g L-1, the isobutyric acid concentration lower than

0.5 g L-1, and the propionic acid concentration lower than 1 g L-1 (Kaiser et al.,

2008), but a concentration of propionic acid higher than 0.3 g L-1 is sufficient to

disturb anaerobic digestion (Deublein and Steinhauser, 2008).

Once again the concentration of volatile fatty acids is consistent with that ex-

pected for the amino acids of bovine hide collagen fermented through Stickland

reaction (Ramsay and Pullammanappallil, 2001). The key volatile fatty acid pro-

duced during the three trials was acetic acid, followed by isovaleric acid and

propionic acid.

The reactor fed with untreated chromium shavings (Figure 4.16a) showed a sta-

ble total volatile fatty acid concentration for a loading rate of up to 1.0 g kg-1 d-1

but the concentration increased at a loading rate of 1.5 g kg-1 d-1 to more than

4 g L-1, and the daily methane production dropped. Additionally, for the final

sample collected the acetic acid concentration was 3.5 g L-1 and the propionic

acid concentration reached 0.6 g L-1 indicating a complete failure of the reactor

at this loading rate.

4.2 Biogas Production

105

Fig. 4.16: Volatile fatty acid concentrations during anaerobic digestion of untreated chromium

shavings (a), shavings extruded dry at 100 °C (b), and shavings treated hydrothermally at

140 °C and 150 °C (c) (source: Gomes et al., 2019b).

time (days)

010 20 30 40 50 60

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Volatile fatty acid concentration (g L-1)

10 Loading rate

Total VFA

Acetic acid

Propionic acid

Isobutyric acid

Butyric acid

Isovaleric acid

time (days)

010 20 30 40 50 60

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Volatile fatty acid concentration (g L-1)

10 Loading rate

Total VFA

Acetic acid

Propionic acid

Isobutyric acid

Butyric acid

Isovaleric acid

time (days)

010 20 30 40 50 60

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Volatile fatty acid concentration (g L-1)

10 Loading rate

Total VFA

Acetic acid

Propionic acid

Isobutyric acid

Butyric acid

Isovaleric acid

4 Results and Discussion

106

Similarly, using extruded shavings as substrate (Figure 4.16b), the total volatile

fatty acid concentration was low for a loading rate of up to 1.0 g kg-1 d-1. There

was a small increase of acid concentration for the first loading rate correspond-

ing with the ingress of oxygen registered for this reactor, but the system recov-

ers its former stability and the acid concentration drops again. The acid concen-

tration started increasing at a loading rate of 1.4 g kg-1 d-1 without reaching an

inhibitory concentration, which could explain the unstable daily methane pro-

duction values at this loading rate. Finally, for the last loading rate, the total

acid concentration reached 5.4 g L-1 and the propionic acid concentration

reached a value of 1 g L-1 indicating a failure of the reactor. The acetic acid and

isobutyric acid concentrations were also high for this last biomass sample but

without reaching an inhibition concentration.

The reactor fed with shavings treated hydrothermally (Figure 4.16c) also

showed a small increase in the volatile fatty acid concentration due to the agi-

tation failure at a loading rate of 1.0 g kg-1 d-1. Again, the reactor recovers stabil-

ity and the acid concentration drops. Otherwise, the reactor showed stable vol-

atile fatty acid concentrations. The concentration only increased for the last bi-

omass sample collected at a loading rate of 1.5 g kg-1 d-1. Inhibition began at the

last loading rate and daily methane production dropped. The total volatile fatty

acid concentration increased to up to 10.1 g L-1, the acetic acid concentration to

up to 7.8 g L-1, propionic acid to up to 0.6 g L-1, and isobutyric acid to up to

0.4 g L-1.

The chromium content in the reactors could inhibit the methanogenic archaea

without affecting the acidogenic bacteria resulting in the accumulation of vola-

tile fatty acids. Cr3+ ions could also have other negative effects on the anaerobic

digestion such as a decrease in the total gas production rate, a fall in the pH

value, a decrease in the percentage of methane in the produced biogas, or a de-

crease in the COD removal efficiency (Alkan et al., 1996). The pH value and the

chromium content of the biomass samples collected were analysed. The results

are shown in Table 4.12.

4.2 Biogas Production

107

Tab. 4.12: pH values and chromium content of the samples collected during digestion.

Time (d)

Loading rate (g kg-1 d-1)

Chromium (%)*

0.5

8.1

1.1

1.0 ± 0.0

7.9

1.4 ± 0.1

7.9

1.6 ± 0.1

1.0

7.9

1.9 ± 0.0

8.0

2.4 ± 0.1

8.1

2.7 ± 0.0

8.1

3.5 ± 0.0

8.1

3.3 ± 0.1

1.5

8.1

3.9 ± 0.1

8.0

4.3 ± 0.1

E100D

0.5

8.3

1.0 ± 0.0

7.8

1.3 ± 0.1

7.9

1.6 ± 0.0

8.0

2.0 ± 0.1

7.9

2.5 ± 0.0

1.0

8.0

3.1 ± 0.1

8.1

3.5 ± 0.1

8.0

3.8 ± 0.1

1.4

8.0

4.7 ± 0.1

8.1

4.9 ± 0.4

1.9

8.0

4.5 ± 0.0

H140 - H150

0.5

8.1

1.2 ± 0.0

8.0

1.6 ± 0.0

8.0

1.7 ± 0.0

1.0

7.8

2.3 ± 0.0

8.1

2.8 ± 0.1

8.1

3.1 ± 0.0

8.1

3.1 ± 0.0

1.5

8.1

3.5 ± 0.1

8.1

4.7 ± 0.0

2.0

8.0

4.7± 0.0

*Dry basis; mean ± standard deviation, n = 2; measured as chromium oxide

Source: Gomes et al., 2019b.

4 Results and Discussion

108

The pH value of the biomass during digestion was very stable, between 7.8 and

8.1, causing no disturbance in the system. The same stability was previously

seen in the batch trials. It is known that a pH value between 5.2 and 6.3 is perfect

for bacteria from the hydrolysis and acidogenesis phases, however these bacte-

ria also work for a slightly higher pH value with slightly reduced activity. For

the acetogenic bacteria and methanogenic archaea, the pH value needed ranges

from 6.5 to 8 and, therefore, the pH value of the reactor must be within this

range (FNR, 2012).

For Deublein and Steinhauser (2008) the toxic effect of Cr3+ as free ion starts at

concentrations of 28 to 300 mg L-1. Alkan et al. (1996) studied the presence of

chromium during anaerobic digestion in depth, testing the direct injection of

Cr3+ ions in anaerobic digesters. They concluded that a shock injection of only

500 mg L-1 of Cr3+ was necessary to lead to a system failure. However, a total

concentration of up to 1140 mg L-1 of Cr3+ was tolerated in the anaerobic digester

when injected in a stepwise manner. In the latter case, the soluble chromium

concentration in the digester supernatant was measured on the following day

of each injection and only 20 mg L-1 of Cr3+ were found. The remainder of the

chromium was assumed to have precipitated in other forms, for which how-

ever, no measurements were made. This shows that methanogenic archaea can

adapt to the presence of chromium and that not all of the chromium in the di-

gester is soluble. In fact, soluble chromium is more toxic with regard to anaero-

bic digestion than insoluble chromium. Furthermore, the authors concluded

that Cr3+ reduced the production of methane and increased the concentration of

volatile fatty acids indicating that this metal has direct effect on the methano-

genic bacteria.

For the continuous trials studied, chromium was present in the initial inoculum

and it was added in a stepwise manner, as the substrates contain chromium. At

the end of the trials, a total chromium content of almost 5% (measured as chro-

mium oxide) was achieved in the biomass of all continuous trials which, as ex-

pected, is much higher than the values seen previously in the batch trials.

The chromium content of the anaerobic sludge used as inoculum was high be-

cause the sludge was generated in a tannery. The inoculum for the reactor fed

4.2 Biogas Production

109

with untreated chromium shavings had a total chromium content of 324 mg L-1,

the inoculum for the extruded shavings reactor had a total chromium content

of 233 mg L-1, and the inoculum for the reactor fed with shavings treated hydro-

thermally had a content of 369 mg L-1. The bacteria present in the initial sludge

were already adapted to high quantities of chromium before its utilization as

inoculum in the anaerobic digestion. The biomass of all continuous trials

reached the concentration found to be toxic by Alkan et al. (1996). This is true

for the loading rates determined to be more appropriate for each substrate that

is 1.0 g kg-1 d-1 for the reactor fed with untreated chromium shavings,

1.4 g kg-1 d-1 for the reactor fed with extruded shavings, and 1.5 g kg-1 d-1 for the

reactor fed with shavings treated hydrothermally. However, the soluble part of

the total chromium in the reactors was not measured in this paper. Free ions of

chromium are known to be more toxic to the methanogenic archaea than insol-

uble chromium.

This indicates that the total chromium concentration found in the reactors stud-

ied was not inhibitory for the methanogenic archaea. However, further studies

should measure all fractions of chromium in the anaerobic reactor. Determina-

tion of a more suitable limit concentration of chromium would only be possible

with a long-term continuous trial at an appropriate loading rate, without other

inhibitions. This long-term trial should be carried out with the objective of

reaching a maximum chromium content in the final biomass, which could ease

the removal of chromium at the end of the process. Since chromium is normally

obtained from mining at low cost, the average chromium content for which re-

cycling is economically reasonable is between 14 to 40% by weight (Bertau,

2018). If this chromium content is reached and the chromium can be extracted

from the biomass, it will be possible to use the removed chromium in the tan-

ning step of the leather-making process.

A high concentration of hydrogen sulfide in the generated biogas can be ex-

tremely damaging to the process. Not only because of inhibitory effects but also

because of its toxicity even at low concentrations, and corrosive effects which

reduce equipment lifetime. The presence of sulfate could form hydrogen sulfide

(Equation 2.5). Figure 4.17 shows the concentration of H2S in the biogas pro-

duced and the concentration of sulfate in the biomass samples collected.

4 Results and Discussion

110

Fig. 4.17: Concentration of total sulfate in the biomass collected and hydrogen sulfide in the

biogas formed at different reaction times for the reactor fed with chromium shavings (a), shavings

extruded dry at 100 °C (b), and shavings treated hydrothermally at 140 and 150 °C (c).

time (days)

010 20 30 40 50 60

Sulfate concentration (mg L-1)

200

400

600

800

Hydrogen sulfide concentration (mg L-1)

500

1000

1500

2000

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

time (days)

010 20 30 40 50 60

Sulfate concentration (mg L-1)

200

400

600

800

Hydrogen sulfide concentration (mg L-1)

500

1000

1500

2000

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Sulfate

Hydrogen sulfide

Loading rate

time (days)

010 20 30 40 50 60

Sulfate concentration (mg L-1)

200

400

600

800

Hydrogen sulfide concentration (mg L-1)

500

1000

1500

2000

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

4.2 Biogas Production

111

The concentration of hydrogen sulfide in biogas depends on the substrate. Bo-

vine hide collagen presents low quantities of sulfur in its basic elementary for-

mula (The UniProt Consortium), which is not enough for the production of the

registered quantities of H2S. Furthermore, organic sulfur from protein is not

completely transferred to the gas phase, part of it remains in the biomass (Pol-

ster and Brummack, 2005). H2S is produced from the reduction of inorganic sul-

fate when sulfate-degrading bacteria compete with the methanogenic archaea

to produce hydrogen sulfide. The inorganic sulfate present in the reactor origi-

nates from the basic chromium sulfate used in the tanning process to produce

leather.

All reactors started producing hydrogen sulfide when the loading rate was

raised to 1.0 g kg-1 d-1. Afterwards, the increase in concentration of this toxic gas

was controlled by adding iron chloride. The addition of iron chloride and the

concentration of iron in the biomass samples collected can be seen in Figure

4.18. If H2S drops, the concentration of sulfate steeply rises as H2S can no longer

be converted and the loading rate for the substrate is higher. The trial using

untreated chromium shavings as substrate reached a high concentration of H2S

(Figure 4.17a) as it was the first trial executed and the addition of iron chloride

was based on the author’s own experience. However, the two continuous trials

subsequently performed (Figures 4.17b and 4.17c) indicate that it is possible to

easily control the H2S concentration with the addition of iron chloride.

As previously seen, bovine hide collagen, the main component of the substrates

studied, is a nitrogen-rich substrate with a C/N ratio of 3.1 and there is the pos-

sibility of inhibition by ammonia. The ammonia content in the biomass of the

reactors was monitored during anaerobic digestion. Figure 4.19 shows the con-

centration of ammonia results. The sample point 0 represents the ammonia con-

tent of the inoculum before digestion.

4 Results and Discussion

112

Fig. 4.18: Concentration of iron in the biomass collected and mass of added iron chloride at dif-

ferent reaction times for the reactor fed with untreated chromium shavings (a), shavings ex-

truded dry at 100 °C (b), and shavings treated hydrothermally at 140 °C and 150 °C (c).

time (days)

010 20 30 40 50 60

Iron concentration (mg L-1)

Added iron chloride (g)

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Iron

Added iron chloride

Loading rate

time (days)

010 20 30 40 50 60

Iron concentration (mg L-1)

Added iron chloride (g)

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

time (days)

010 20 30 40 50 60

Iron concentration (mg L-1)

Added iron chloride (g)

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

4.2 Biogas Production

113

Fig. 4.19: Concentration of ammonia found in the biomass collected at different reaction times

for the reactor fed with untreated chromium shavings (a), shavings extruded dry at 100 °C (b),

and shavings treated hydrothermally at 140 °C and 150 °C (c).

time (days)

010 20 30 40 50 60

Ammonia concentration (mg L-1)

0,0

0,1

0,2

0,3

0,4

0,5

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

Ammonia

Loading rate

time (days)

010 20 30 40 50 60

Ammonia concentration (mg L-1)

0,0

0,1

0,2

0,3

0,4

0,5

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

time (days)

010 20 30 40 50 60

Ammonia concentration (mg L-1)

0,0

0,1

0,2

0,3

0,4

0,5

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

4 Results and Discussion

114

The inhibition level of ammonia is known to be 1.5 to 3.0 g NH4-N L-1 (Drosg et

al., 2013). With regard to the three continuous trials, ammonia is present in the

biomass in a concentration below 0.5 g NH4-N L-1 and caused no problem for

anaerobic digestion. The ammonia generated by fermentation of the amino ac-

ids from collagen did not accumulate, presumably as it was used for bacterial

growth.

Table 4.13 shows an overview of the inhibitors present in the reactor and their

inhibitory concentration. Literature values are compared to inhibitory concen-

trations found in the reactors at different loading rates.

Different sources of inhibition are present in the reactor for the substrates stud-

ied. The most important inhibitor appears to be the volatile fatty acids, which

showed an inhibitory concentration for the two loading rates tested last in the

three continuous trials. This could be the cause for the failure of the reactors.

However, the accumulation of these acids was not sufficient to cause acidifica-

tion of the biomass. The pH values are slightly above the optimum during di-

gestion and there is a small reduction for the last loading rate.

For all continuous trials, the total chromium content was above inhibitory con-

centration regarding the second or third loading rate tested, and continues to

increase. However, methane is produced effectively. The biomass appears to be

adapted to the presence of chromium. Concentrations of hydrogen sulfide

above the inhibitory concentration did not disturb anaerobic digestion but

needed to be controlled with the addition of iron salts.

4.2 Biogas Production

115

Tab. 4.13: Inhibitors in anaerobic digestion and values reached in the continuous trials.

Inhibitor

Inhibitory value

Reference

Volatile fatty acid

> 4 g L-1

Kaiser et al. (2008)

< 6.5 and >8

FNR (2012)

Total chromium

> 1.14 g L-1

Alkan et al. (1996)

Hydrogen Sulfide

> 0.05 g L-1

Deublein and Steinhauser (2008)

Ammonia

> 1.5 g L-1

Drosg et al. (2013)

Value reached

Loading rate (g kg-1 d-1)

Volatile fatty acid

5.6 g L-1

1.5

8.1

1.0, 1.5

Total chromium

1.4, 2.0 g L-1

1.0, 1.5

Hydrogen Sulfide

2.0 g L-1

1.0

Ammonia

E100D

Value reached

Loading rate (g kg-1 d-1)

Volatile fatty acid

5.4

1.9

8.1

1.0, 1.4

Total chromium

1.2, 1.8, 1.7 g L-1

1.0, 1.4. 1.9

Hydrogen Sulfide

0.2, 0.7 g L-1

1.0, 1.4

Ammonia

H140-H150

Value reached

Loading rate (g kg-1 d-1)

Volatile fatty acid

10.1 g L-1

2.0

8.1

1.0, 1.5

Total chromium

1.8, 2.0 g L-1

1.5, 2.0

Hydrogen Sulfide

0.8, 0.6, 0.09 g L-1

1.0, 1.5, 2.0

Ammonia

4.2.2.3 Degradation of the substrate

Degradation of the substrate was evaluated analyzing the biomass samples col-

lected weekly during anaerobic digestion. Table 4.14 shows the characterization

of the biomass samples collected. The samples were analyzed regarding their

collagen content, organic matter, and chromium content.

4 Results and Discussion

116

Tab. 4.14: Biomass characterization of the samples collected during continuous digestion.

Time (d)

Loading rate (g kg-1 d-1)

Collagen (%)*

Organic matter (%)*

0.5

2.9 ± 0.3

39.5 ± 1.4

4.3 ± 0.9

41.4 ± 1.2

4.4 ± 0.4

41.2 ± 1.8

5.8 ± 0.6

38.7 ± 0.2

1.0

6.3 ± 0.7

41.1 ± 0.3

5.1 ± 0.4

40.8 ± 0.0

3.6 ± 0.4

40.2 ± 0.3

2.5 ± 0.1

40.3 ± 0.2

2.3 ± 0.1

41.3 ± 0.2

1.5

2.6 ± 0.2

44.6 ± 0.3

3.3 ± 0.0

45.4 ± 0.1

E100D

0.5

2.7 ± 0.3

38.3 ± 0.3

3.2 ± 0.1

38.4 ± 1.2

3.3 ± 0.1

38.5 ± 0.3

3.6 ± 0.1

39.6 ± 1.4

3.4 ± 0.1

37.9 ± 0.1

1.0

3.4 ± 0.3

40.2 ± 0.5

4.5 ± 0.2

40.3 ± 0.2

6.5 ± 0.3

42.0 ± 0.1

1.4

3.4 ± 0.1

43.7 ± 0.2

3.1 ± 0.1

44.4 ± 0.1

1.9

2.8 ± 0.0

46.9 ± 0.1

H140 - H150

0.5

3.3 ± 0.0

37.4 ± 0.2

3.7 ± 0.0

39.1 ± 0.3

4.0 ± 0.1

37.9 ± 0.1

1.0

5.6 ± 0.0

40.7 ± 0.3

4.8 ± 0.3

39.8 ± 0.6

5.1 ± 0.1

39.8 ± 0.4

5.2 ± 0.1

40.6 ± 0.5

1.5

4.9 ± 0.1

42.3 ± 0.3

5.0 ± 0.0

45.3 ± 0.4

2.0

4.0 ± 0.1

48.0 ± 0.1

*Dry basis; mean ± standard deviation, n = 3

4.2 Biogas Production

117

An increase in the organic matter content of the biomass collected would indi-

cate accumulation of unprocessed substrate and, therefore, inefficiency of an-

aerobic digestion. As seen in Table 4.14, the reactor fed with untreated chro-

mium shavings showed a constant value of organic matter content (about 40%)

for most of the time. An increase to 45% was detected at the end of the trial at a

loading rate of 1.5 g kg-1 d-1, indicating that the reactor was overloaded. At this

loading rate methane production was very low and the added substrate accu-

mulated.

The reactor fed with the extruded shavings had an organic matter content of

about 39% at a loading rate of 0.5 g kg-1 d-1. At a loading rate of 1.0 g kg-1 d-1 the

organic matter content remained constant at 40% and showed a slight increase

to 42% at the end of this step. In a next step, an increase in organic matter to

about 44% was observed at a loading rate of 1.4 g kg-1 d-1. Finally, in the last

step, the organic matter content increased to 47% at a loading rate of

1.9 g kg-1 d-1.

For the reactor fed with shavings treated hydrothermally, the organic matter

content of the biomass started at about 38% for the first loading rate. Proceeding

to the next step, a loading rate of 1.0 g kg-1 d-1, increased the organic matter

content of the reactor to about 40%. The rise to a loading rate of 1.5 g kg-1 d-1

resulted in an increase to 45% and the last loading rate led to 48% of organic

matter content, indicating accumulation.

At the end of digestion, the collagen content in the biomass was almost as low

as it was at the beginning of the process for all reactors, showing that collagen

was degraded and that there is no accumulation of collagen in the reactor.

Therefore, the collagen of the added substrate was metabolized. However, the

inorganic part of the substrates accumulated and the chromium content in the

reactor increased.

Although a second batch of extruded shavings was used, which were not com-

pletely denatured, the second batch of the substrate could be degraded by an-

aerobic digestion. When the reactor was shut down, it contained 33.7 g of colla-

gen (dry basis), which represents a final degradation of 96.5% of all of the col-

4 Results and Discussion

118

lagen added (953.8 g of collagen added). The reactor fed with untreated chro-

mium shavings contained 45.6 g of collagen (dry basis) at shutdown, corre-

sponding to 95.5% of degradation (1014.4 g of collagen added) and the reactor

fed with shavings treated hydrothermally showed a final amount of collagen of

58.3 g (dry basis) meaning 95.3% of degradation regarding the collagen added

(1233.4 g of collagen added).

Degradation of collagen after one feed was further studied for the reactor fed

with shavings treated hydrothermally. After feeding the reactor with substrate

on the 51st day of digestion at a loading rate of 2.0 g kg-1 d-1 small biomass sam-

ples were collected from the reactor at different times over a period of two days

for collagen content analyses in triplicate. Figure 4.20 shows the collagen con-

tent results for the biomass samples collected. The error bars represent the

standard deviation for the experimental data.

Fig. 4.20: Collagen content of the biomass samples collected from the reactor fed with shavings

treated hydrothermally after substrate feeding (logarithmic timeline).

time (h)

0,01 0,1 1 10 100

Collagen content (%)

4.2 Biogas Production

119

Prior to feeding, the reactor showed a collagen content of 5%. Shortly after feed-

ing, the collagen content rose to about 44%, a very high value probably due to

the inhomogeneity of the substrate, which can be seen in the high variance of

the results. The substrate needed more time to be solubilized in the biomass.

After two hours only, the collagen content in the biomass was almost as low as

prior to substrate feeding, indicating that most of the collagen had been rapidly

degraded. After forty eight hours, the collagen content dropped to 5%, the same

collagen content detected prior to feeding. This leads to the conclusion that col-

lagen was rapidly degraded at this loading rate even though organic matter

accumulated (Table 4.14).

Although the detectable hydroxyproline in the biomass is low and a high con-

tent of collagen is rapidly degraded, degradation of all of the organic matter

was not possible. Part of the substrate remains unprocessed in the biomass

which can be seen in the organic matter content formed from organics of the

inoculum and intermediate products of the hydrolyzed substrate, such as vola-

tile fatty acids, amino acids, and peptides. In a previous paper (Gomes et al.,

2017), it was shown that anaerobic digestion of inoculum without adding sub-

strate leads to a final biomass of about 40% of organic matter content from

which the conclusion can be drawn that part of the initial organic matter will

remain unprocessed. The increase in organic matter content during digestion is

a consequence of accumulation of unprocessed substrate in the form of inter-

mediate products, which are not transformed into biogas. A possible reason for

the accumulation of unprocessed organic matter is the inhibition of the anaero-

bic bacteria.

The substrate degradation was evaluated through a COD balance calculated

with Equation 3.6 and the theoretical COD of bovine hide collagen as substrate

(1.124 gO2 g-1). The balance was made using the cumulative volume of methane

produced in the reactors and the cumulative mass of added substrate starting

on the first day of digestion (Figure 4.21). The cumulative COD degree of deg-

radation shows degradation of substrate in the process as a whole and indicates

that almost all of the substrate was successfully degraded. Furthermore, the bi-

omass samples collected were analyzed regarding COD as another indicative

of organics in the biomass. The results are shown in Figure 4.21.

4 Results and Discussion

120

Fig. 4.21: Cumulative COD degree of degradation during digestion and COD in the biomass

samples collected at different reaction times for the reactor fed with chromium shavings (a),

shavings extruded dry at 100 °C (b) and shavings treated hydrothermally at 140 and 150 °C (c).

time (days)

010 20 30 40 50 60

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

COD degree of degradation (%)

100

COD (g L-1)

time (days)

010 20 30 40 50 60

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

COD degree of degradation (%)

100

COD (g L-1)

80 Loading rate

COD Degradation

COD

time (days)

010 20 30 40 50 60

Loading rate (g kg-1 d-1)

0,0

0,5

1,0

1,5

2,0

COD degree of degradation (%)

100

COD (g L-1)

4.2 Biogas Production

121

The maximum cumulative COD degree of degradation for the trial using un-

treated chromium shavings as substrate (Figure 4.21a) was 94.9% at a loading

rate of 1.0 g kg-1 d-1, the same loading rate that showed the highest daily me-

thane production for this substrate. For this loading rate the cumulative COD

degree of degradation at first reduces and later increases, indicating that the

system needs to adapt to higher loads before starting production at its maxi-

mum capacity. For the same loading rate, COD increases slowly. In the last

loading rate step, COD rises steeply and the cumulative COD degree of degra-

dation declines.

The cumulative COD degree of degradation for the reactor fed with extruded

shavings (Figure 4.21b) was affected by the ingress of oxygen on the fifth day

of digestion. This trial takes longer than the other continuous trials to reach high

COD degrees of degradation. The highest COD degree of degradation of 83.8%

was achieved at a loading rate of 1.0 g kg-1 d-1 and not for the next step as ex-

pected, when the maximum daily methane production was reached. The value

was lower than that for the other reactors. This is probably due to the instability

caused by the increasing volatile fatty acids recorded for a loading rate of

1.4 g kg-1 d-1. COD values started increasing at a loading rate of 1.0 g kg-1 d-1 and

they became steeper for every subsequent loading rate step.

The reactor fed with shavings treated hydrothermally (Figure 4.21c) showed a

maximum cumulative COD degree of degradation of 92.5% at a loading rate of

1.5 g kg-1 d-1, the loading rate for the maximum daily methane production

reached in this trial. There are two different reasons for the drop in cumulative

COD degree of degradation at a loading rate of 1.0 g kg-1 d-1. The system needed

to adapt to a new loading rate after the rise from 0.5 g kg-1 d-1 to 1.0 g kg-1 d-1,

which caused a slight drop in degradation. Moreover, the agitation failure for

this reactor on the 21st day intensified the drop. The COD values in the biomass

samples were stable for a loading rate of up to 1.0 g kg 1 d-1, increased slightly

for the next step, and steeply at the last step.

Considering that the highest daily methane production for the untreated shav-

ings was reached at a loading rate of 1.0 g kg-1 d-1, it is possible to conclude that

this is the most appropriate loading rate for these collagen-based materials. For

4 Results and Discussion

122

the reactor fed with extruded shavings the highest daily methane production

was reached at a loading rate of 1.4 g kg-1 d-1 proving to be the most suitable

loading rate for the extruded shavings among the loading rates studied. Finally,

the reactor fed with shavings treated hydrothermally reached its highest daily

methane production at a loading rate of 1.5 g kg-1 d-1, indicating that this rate

should be aimed at. Nevertheless, long-term trials of one year or more should

prove this observation.

Accumulation of organic matter and the increasing concentrations of volatile

fatty acids verified in the biomass samples collected for the last loading rate

tested in all continuous trials will lead to the same conclusion. Pre-treatment of

the chromium shavings allowed an increase of the loading rate of 40 to 50% and

an increase of the daily methane production of almost 10%. A higher loading

rate leads to a more economical process and it can be expected that the loading

rate could be increased by 40 to 50% by pre-treating the chromium shavings. In

contrast, the reactor volume for a given loading rate could be decreased. This

increases the feasibility of digesting chromium shavings to produce biogas and

energy in tanneries.

Additionally, the untreated chromium shavings can also be digested in contin-

uous systems to produce methane but using a lower loading rate. An explana-

tion for this observation is still missing, since this contradicts the common

knowledge that untreated chromium leather cannot be digested in biogas reac-

tors. The cause could be the degradation of the untreated material by several

enzymes working simultaneously but more research is needed in this subject.

Results presented in this section indicate the loading rate to be aimed at when

working with each of the substrates studied. However, it is important to bear

in mind that these are preliminary results and long-term trials should be carried

out testing the stability of anaerobic digestion at the aimed loading rates.

123

5 Conclusions

Chromium leather waste produced in tanneries is not commonly considered as

substrate for biogas production through anaerobic digestion due to its high sta-

bility and chromium content. Nonetheless, this waste is produced in high quan-

tities worldwide and the use as substrate would be a convenient solution to its

disposal with the advantage of generating energy. To make this process feasi-

ble, pre-treatments can be used. This approach to produce biogas in batch and

continuous scale has not been studied to date.

Three different pre-treatments were carried out to denature the stable collagen

structure of chromium shavings and one pre-treatment was performed for

leather offcuts. The aim was to enhance anaerobic digestion when using chro-

mium leather waste as substrate to produce biogas. The hydrothermal and au-

toclaving pre-treatments increased degradation of chromium shavings by tryp-

sin from 7% to more than 90% and extrusion to 35%. Extruded offcuts had a

lower degradation by trypsin than chromium shavings due to their complexity

acquired in the wet end and finishing process after tanning. Degradation of the

chromium shavings using collagenase could also be increased from 12% to 86%

by extrusion. To measure digestibility, it was possible to reach a high degrada-

tion degree by trypsin and collagenase of chromium shavings when using a

pre-treatment.

Differential scanning calorimetry results showed that the hydrogen bonds that

stabilize the collagen structure of the chromium leather waste against enzy-

matic degradation were completely or almost completely broken down for most

of the pre-treated samples. An exception were shavings autoclaved for only

three minutes due to insufficient pre-treatment time. Therefore, evaluation of

the pre-treated materials showed that almost all of them were denatured. A

5 Conclusions

124

quantity of 4% of chromium measured as chromium oxide in chromium leather

waste appears not to be toxic for anaerobic bacteria.

In biogas production batch tests, the trials with agitation were proven to have a

better performance for the pre-treated substrates tested, facilitating the mass

transfer inside the reactors. The pre-treatments were important to reduce the

hydrolysis step of anaerobic digestion. Using extrusion as pre-treatment caused

a decrease of the lag-phase at the beginning of the biogas production of the

chromium shavings by four days and hydrothermal treatment resulted in a de-

crease of four to five days. Biogas production could be initiated already one day

after starting the trials with shavings treated hydrothermally at 170 °C as sub-

strate, the same lag-phase found for gelatin. Using a pre-treatment, the collagen

of the chromium shavings degraded more intensively than without pre-treat-

ment and it was possible to reach a collagen degree of degradation of above

98%.

Pre-treated chromium shavings showed a behaviour more similar to the partly

hydrolyzed collagen gelatin than to untreated chromium shavings, confirming

that the use of a pre-treatment on the substrate prior to digestion improves the

biogas production. However, pre-treated shavings could not reach the biogas

formation potential of gelatin. When chromium shavings are denatured by

means of a pre-treatment, the hydrogen bonds, which stabilize the triple helix,

are broken down but the structure is still arranged in long chains needing more

time to degrade. Even gelatin with hydrolyzed short chains could not reach the

biogas formation potential of other proteins due to its arrangement. Collagen

yields less biogas than other proteins because its structure, denatured or not, is

hard to degrade.

Even though there is a lag-phase before starting production, trials using un-

treated chromium shavings as substrate reached a similar biogas formation po-

tential compared to pre-treated chromium shavings. Probably a combination of

several enzymes present in the medium needs to work simultaneously to de-

grade the substrate but the enzymes remain to be identified. The use of ex-

truded offcuts as substrate hindered biogas production compared to the use of

4.2 Biogas Production

125

untreated chromium shavings. The wet end and finishing process in the auto-

motive leather industry increases the complexity of the substrate for anaerobic

digestion by adding organic matter to the substrate, which cannot be processed

during anaerobic digestion. The use of different substrate to inoculum ratios

(S/I) showed that in a bioreactor for all pre-treated and untreated substrates the

ratio of 0.5 should be aimed at. Additionally, for trials using gelatin, extruded

shavings, and extruded offcuts a ratio higher than 0.5 generated a two-phase

decomposition. This kind of cumulative biogas production curve is known as

diauxie.

As denatured collagen-based material is quickly hydrolyzed, the use of an S/I

ratio higher than 0.5 will produce intermediate products with a velocity which

the methanogenesis cannot follow. Analysis of the volatile fatty acid concentra-

tion of biomass samples during anaerobic digestion with an S/I ratio of 1.5 re-

vealed that there was an excess of intermediate products during the biogas pro-

duction plateau. When using pre-treated substrates, hydrolysis occurs quickly

and the quantity of volatile fatty acids produced is high during acidogenesis. In

these cases, the rate-limiting step in the anaerobic digestion is methanogenesis.

In contrast, when the untreated chromium shavings were tested, hydrolysis

needed more time for digestion and methanogenesis occurred slowly and with-

out inhibitions. In this case, hydrolysis is the rate-limiting step.

The diauxie behaviour verified for pre-treated substrates with an S/I ratio

higher than 0.5 was not caused by the high volatile fatty acid concentration dur-

ing the biogas production plateau. As a consequence of inhibition of the meth-

anogenic archaea, high concentrations of these unprocessed acids, mainly acetic

acid and propionic acid accumulate. Further research should be carried out to

determine the exact inhibitor or inhibitors of the methanogenic archaea in this

case.

In continuous tests of biogas production, the reactors fed with pre-treated shav-

ings showed better results than those using untreated chromium shavings as

substrate, even with some technical problems during anaerobic digestion. The

continuous biogas trials showed that it was possible to use a substrate loading

rate which was 40 to 50% higher. Moreover, the daily methane production

5 Conclusions

126

could be increased by almost 10% by using pre-treated shavings rather than

untreated chromium shavings as substrate.

Volatile fatty acids and organics in the biomass started to accumulate when the

loading rate tested was too high for the system, leading to a lower daily me-

thane production and failure of the reactor. Even at high loading rates, the

added collagen was rapidly degraded. Most of it was degraded within two

hours avoiding accumulation of unprocessed collagen in the reactor. The re-

maining organic matter at the end of the process is formed from intermediate

products which could not be identified and were not transformed into biogas.

When using chromium leather waste as substrate, there is a potential of pro-

ducing biogas with high concentrations of hydrogen sulfide mainly due to the

basic chromium sulfate used in the tanning process. This could cause inhibition

of methanogenesis, problems with equipment, and hazard to personnel. How-

ever, the concentrations of H2S were easily controlled with the addition of iron

chloride, a common H2S scavenger.

The use of pre-treated chromium leather waste to produce biogas is a promising

method to be performed in an industrial environment. An economical evalua-

tion should be performed to verify if the pre-treatment costs are compensated

by the energy gains. However, the reduction of the final waste, the disposal of

which otherwise would generate costs, makes this method very attractive for

industry purposes.

The reuse of chromium leather waste (collagen-based material) as a substrate in

the biogas production is an interesting alternative to deal with the large

amounts that are produced every day in tanneries. However, its degradation

was particularly difficult compared to other proteins, such as casein or blood.

An explanation for this could be that the collagen arrangement of triple helical

chains needs considerable effort to be degraded. This could be investigated in

future by analyzing intermediates, not only volatile fatty acids but also amino

acids or oligo-peptides, which can result from degradation processes.

4.2 Biogas Production

127

There are several theories about the reasons for diauxie. Results presented in

this study lead to the conclusion that diauxie is not a direct consequence of vol-

atile fatty acid accumulation as discussed by other authors. In the future studies

should be carried out to determine the inhibitor of methanogenesis and the

cause of diauxie behaviour.

Using leather offcuts to produce biogas appears to be difficult. Even though

they were completely denatured by extrusion, the biogas formation potential

was low and the remaining organic matter in the final biomass was higher than

for the other substrates studied. For the substrate meant in the present paper, it

is important to identify the unprocessed organic matter in the final biomass,

which mainly results from the wet end and finishing process in the leather in-

dustry. Subsequently, a more appropriated pre-treatment should be studied

and tested to break down these organics and increase their susceptibility to an-

aerobic digestion.

The unprocessed organic matter remaining in the final biomass in batch and

continuous reactors is another challenge. These organics have been seen in final

biomass coming from all of the substrates studied, including gelatin. All organic

matter has a potential to be transformed into energy in anaerobic digestion.

Consequently, this can be seen as a waste of resources. Furthermore, the final

biomass is not suitable to be used as a fertilizer because of its high chromium

content and its disposal would generate costs, which make the process less eco-

nomically attractive. These organics should be further analyzed and identified

to find a possible solution. If possible, a different pre-treatment, which allows

these organics to be transformed into biogas, should be identified.

Long-term trials of one year or more should be carried out to confirm results

obtained with continuous trials. The loading rate for each substrate aimed at

should be tested to confirm if the conditions are stable over a long period of

time and if there is accumulation of organic matter. Another important aspect

is the accumulation of inorganic matter, mainly chromium. A chromium con-

tent of almost 5% was not toxic for the continuous reactors but a long-term trial

could determine the maximum chromium content tolerated by the system. If a

5 Conclusions

128

higher chromium content was reached, extraction and recycling of the chro-

mium could be reasonable and there could be further research in this regard.

Results hitherto obtained show that the pre-treatments can increase the effi-

ciency of the biogas production from chromium leather waste (collagen-based)

materials. The process could be accelerated, the waste destruction enhanced

and the load of substrate increased. For batch reactors, the lag-phase is shorter

and reduction of waste is more efficient. For continuous systems, daily methane

productivity is higher, and the load of substrate can be increased or the reactor

volume reduced. All these improvements increase the suitability of producing

biogas using chromium leather waste as substrate in the industry.

129

References

Abbasi, T., Tauseef, S. M., Abbasi, S. A. (2012). Biogas Energy. Springer, New

York. ISBN: 978-1-4614-1039-3.

Agustini, C., Costa, M., Gutterres, M. (2018). Biogas production from tannery

solid wastes - Scale-up and cost saving analysis. Journal of Cleaner Production

187, pp. 158–164. DOI: https://doi.org/10.1016/j.jclepro.2018.03.185.

Agustini, C., Neto, W., Costa, M., Gutterres, M. (2015). Biodegradation and bi-

ogas production from solid waste of tanneries. XXXIII IULTCS Congress, Novo

Hamburgo, Brazil.

Alkan, U., Anderson, G. K., Ince, O. (1996). Toxicity of trivalent chromium in

the anaerobic digestion process. Water Research 30, 3, pp. 731–741. DOI:

https://doi.org/10.1016/0043-1354(95)00181-6.

Almeida, M. A. F., Boaventura, R. A. R. (1997). Chromium precipitation from

tanning spent liquors using industrial alkaline residues: a comparative study.

Waste Management 17 (4), pp. 201–209. DOI: https://doi.org/10.1016/S0956-

053X(97)10006-X.

Ashekuzzaman, S. M., Poulsen, T. G. (2011). Optimizing feed composition for

improved methane yield during anaerobic digestion of cow manure based

waste mixtures. Bioresource Technology 102, pp. 2213–2218. DOI:

https://doi.org/10.1016/j.biortech.2010.09.118.

Avery, N. C., Bailey, A. J. (2008). Restraining cross-links responsible for the me-

chanical properties of collagen fibers: natural and artificial. In: Fratzl P., Ed.

Collagen: structure and mechanics. Springer, New York, pp. 81–110. ISBN: 978-

0-387-73905-2.

References

130

Banu, J. R., Kaliappan, S. (2007). Treatment of tannery wastewater using hybrid

upflow anaerobic sludge blanket reactor. Journal of Environmental Engineering

and Science 6, pp. 415–421. DOI: https://doi.org/10.1139/S06-063.

Berhe, S., Leta, S. (2018). Anaerobic co-digestion of tannery waste water and

tannery solid waste using two-stage anaerobic sequencing batch reactor: focus

on performances of methanogenic step. Journal of Material Cycles and Waste

Management 20, pp. 1468–1482. DOI: https://doi.org/10.1007/s10163-018-0706-

Bertau, M. (2018). Die zukünftige Sicherung der Rohstoffbasis. Chemie Ingeni-

eur Technik 90 (11), pp. 1647–1657. DOI: https://doi.org/10.1002/cite.201800052.

Birk, D. E., Bruckner, P. (2005). Collagen Suprastructures. In: Brinckmann J.,

Notbohm H., Müller P.K. Eds. Collagen Primer in Structure, Processing and As-

sembly. Springer, Berlin, pp. 185–205. ISBN: 978-3-540-23272-8.

Boyle, W. C. (1976). Energy recovery from sanitary landfills – a review. In:

Schlegel, H.G. and Barnea, J. Eds. Microbial energy conversion. Pergamon

Press, Oxford, UK. ISBN: 9781483139128.

Breure, A. M., Mooijman, K. A., van Andel, J. G. (1986). Protein degradation in

anaerobic digestion: influence of volatile fatty acids and carbohydrates on hy-

drolysis and acidogenic fermentation of gelatin. Applied Microbiology and Bi-

otechnology 24, pp. 426–431. DOI: https://doi.org/10.1007/BF00294602.

Buljan, J., Reich, G., Ludvik, J. (2000). Mass balance in leather processing.

UNIDO, Vienna.

Chanakya, H. N., Khuntia, H. K., Mukherjee, N., Aniruddha, R., Mudakavi, J.

R., Thimmaraju, P. (2015). The physicochemical characteristics and anaerobic

degradability of desiccated coconut industry waste water. Environmental Mon-

itoring and Assessment 187, 772. DOI: https://doi.org/10.1007/BF00294602.

Cot, J., Marsal, A., Manich, A., Celma, P., Choque, R., Cabeza, L., Labastida, L.,

López, J., Salmerón, J. (2003). Minimization of industrial wastes: adding value

to collagenic materials. Journal of the Society of Leather Technologists and

Chemists 87, pp. 91–97.

References

131

Covington, A. D. (2009). Tanning Chemistry: The Science of Leather. The Royal

Society of Chemistry, Cambridge. ISBN: 978-0-85404-170-1.

Crispim, A., Mota, M. (2003). Leather shavings treatment - an enzymatic ap-

proach. Journal of the society of leather technologists and chemists 87 (5), pp.

203–207.

Daudt, R. H. S., Gruszynski, C., Kämpf, A. N. (2007). The use of residues of wet

blue leather as a growing media component. Ciência Rural 37 (1), pp. 91–97.

DBFZ Betreiberbefragung Biogas (2017). Anlagenbestand biogas und biome-

than – biogaserzeugung und -nutzung in Deutschland. DBFZ Deutsches Bio-

masseforschungszentrum gemeinnützige GmbH, Leipzig, Germany. ISBN: 978-

3-946629-24-5.

Deublein, D., Steinhauser, A. (2008). Biogas from Waste and renewable Re-

sources An Introduction. Wiley-VCH, Weinheim. ISBN: 978-3527327980.

Dhayalan, K., Fathima, N. N., Gnanamani, A., Rao, J. R., Nair, B. U., Ramasami,

T. (2007). Biodegradability of leathers through anaerobic pathway. Waste Ma-

nagement 27, pp. 760–767. DOI: https://doi.org/10.1016/j.wasman.2006.03.019.

DIN EN ISO 4047 (1998). Leather - Determination of sulphated total ash and

sulphated water-insoluble ash. German Institute for Standardization, Berlin,

Germany. DOI: https://dx.doi.org/10.31030/7560063.

DIN EN ISO 4684 (2005). Leather – Chemical tests – Determination of volatile

matter. German Institute for Standardization, Berlin, Germany. DOI:

https://dx.doi.org/10.31030/9651471.

DIN EN ISO 5398-1 (2007). Leather - Chemical determination of chromic oxide

content. German Institute for Standardization, Berlin, Germany. DOI:

https://dx.doi.org/10.31030/2838221.

Drosg, B., Braun, R., Bochmann, G. (2013). Analysis and characterisation of bi-

ogas feedstocks, chapter 3. In: Wellinger, A., Murphy, J., Baxter, D., Eds. The

biogas handbook. Woodhead Publishing Limited, Cambridge. ISBN:

9780857094988.

References

132

FAO (2016). World statistical compendium for raw hides and skins, leather and

leather footwear 1999-2015. FAO, Rome. ISBN: 978-92-5-109213-2.

Ferreira, M. J., Almeida, M. F., Pinho, S. C., Gomes, J. R., Rodrigues, J. L. (2014).

Alkaline hydrolysis of chromium tanned leather scrap fibers and anaerobic bi-

odegradation of the products. Waste and Biomass Valorization 5, pp. 551–562.

DOI: https://doi.org/10.1007/s12649-013-9252-9.

Ferreira, M. J., Almeida, M. F., Pinho, S. C., Santos, I. C. (2010). Finished leather

waste chromium acid extraction and anaerobic biodegradation of the products.

Waste Management 30, pp. 1091–1100. DOI: https://doi.org/10.1016/j.was-

man.2009.12.006.

Flowers, K. (2017). Degradation of leather. International Leather Maker 21, pp.

84–86.

FNR (2012). Guide to Biogas From production to use. Fachagentur Nachwach-

sende Rohstoffe e. V. (FNR), Gülzow-Prüzen, Germany.

Franke-Whittle, I. H., Walter, A., Ebner, C., Insam, H. (2014). Investigation into

the effect of high concentrations of volatile fatty acids in anaerobic digestion on

methanogenic communities. Waste Management 34, pp. 2080–2089. DOI:

https://doi.org/10.1016/j.wasman.2014.07.020.

Fratzl, P. (2008). Collagen: structure and mechanics, an introduction, chapter 1.

In: Fratzl, P., Ed. Collagen: structure and mechanics. Springer, New York. ISBN:

978-0-387-73905-2.

Gamble, K. J., Houser, J. B., Hambourger, M. S., Hoepfl, M. C. (2015). Anaerobic

Digestion from the Laboratory to the Field: An Experimental Study into the

Scalability of Anaerobic Digestion. ASEE Southeast Section Conference, Gaines-

ville, USA.

Gayatri, R., Rajaram, R., Ramasami, T. (2000). Inhibition of collagenase by

Cr(III): its relevance to stabilization of collagen. Biochimica et Biophysica Acta

1524, pp. 228–237. DOI: https://doi.org/10.1016/s0304-4165(00)00164-1.

References

133

Gomes, C. S., Piccin, J. S., Gutterres, M. (2015). Optimizing adsorption parame-

ters in tannery-dye-containing effluent treatment with leather shaving waste.

Process Safety and Environmental Protection 99, pp. 98–106. DOI:

http://dx.doi.org/doi:10.1016/j.psep.2015.10.013.

Gomes, C. S., Repke, J-U., Meyer, M. (2017). Different pre-treatments of chrome

tanned leather waste and their use in the biogas production. XXXIV IULTCS

Congress, Chennai, India.

Gomes, C. S., Repke, J-U., Meyer, M. (2019). Diauxie during biogas production

from collagen-based substrates. Renewable Energy 141, pp. 20–27. DOI:

https://doi.org/10.1016/j.renene.2019.03.123. (a)

Gomes, C. S., Repke, J-U., Meyer, M. (2019). Use of different pre-treated chro-

mium leather shavings to produce biogas in continuous scale. XXXV IULTCS

Congress, Dresden, Germany. (b)

Gorham, S. D., Light, N. D., Diamond, A. M., Willins, M. J., Bailey, A. J., Wess,

T. J., Leslie, N. J. (1992). Effect of chemical modifications on the susceptibility of

collagen to proteolysis. II. Dehydrothermal crosslinking. International Journal

of Biological Macromolecules 14 (3), pp. 129–138. DOI:

https://doi.org/10.1016/S0141-8130(05)80002-9.

GROW - Informatics unit - EDW team (2016). Estimations of structural data –

Leather. URL: https://circabc.europa.eu/sd/a/be99fe9c-5856-4fd4-bd9e-

51084b6a1a64/EDW%20-%20ITI%20TEX%20-%20Estimations%20of%20struc-

tural%20data%20-%20Leather.pdf (accessed 06.09.2019).

Heidemann, E. (1993). Fundamentals of Leather Manufacturing, Eduard

Roether KG, Darmstadt. ISBN: 3-7929-0206-0.

Holloway, D. F. (1978). Process for recovery and separation of nutritious protein

hydrolysate and chromium from chrome leather scrap. US Patent 4,100,154.

Hulmes, D. J. S. (2008). Collagen diversity, synthesis and assembly, chapter 2.

In: Fratzl, P., Ed. Collagen: structure and mechanics. Springer, New York. ISBN:

978-0-387-73905-2.

References

134

IULTCS (2018). IUE2: Recommendations for tannery solid by-product manage-

ment. URL: http://www.iultcs.org/pdf/IUE_2.pdf (accessed 06.09.2019). (a)

IULTCS (2018). IUE4: Assessment for chromium containing waste from the

leather industry. URL: http://www.iultcs.org/pdf/IUE_4.pdf (accessed

06.09.2019). (b)

Kafle, G. K., Kim, S. H. (2012). Kinetic study of the anaerobic digestion of swine

manure at mesophilic temperature: a lab scale batch operation. Journal of Bio-

systems Engineering 37 (4), pp. 233–244. DOI:

http://dx.doi.org/10.5307/JBE.2012.37.4.233.

Kafle, G. K., Kim, S. H. (2013). Anaerobic treatment of apple waste with swine

manure for biogas production: Batch and continuous operation. Applied En-

ergy 103, pp. 61–72. DOI: http://dx.doi.org/10.1016/j.apenergy.2012.10.018. (a)

Kafle, G. K., Kim, S. H. (2013). Effects of chemical compositions and ensiling on

the biogas productivity and degradation rates of agricultural and food pro-

cessing by-products. Bioresource Technology 142, pp. 553–561. DOI:

http://dx.doi.org/10.1016/j.biortech.2013.05.018. (b)

Kaiser, F., Metzner, T., Effenberger, M., Gronauer, A. (2008). Sicherung der Pro-

zessstabilität in landwirtschaftlichen Biogasanlagen. Bayerische Landesanstalt

für Landwirtschaft (LfL), Freising-Weihenstephan.

Kameswari, K. S. B., Kalyanaraman, C., Porselvam, S., Thanasekaran, K. (2012).

Optimization of inoculum to substrate ratio for bio-energy generation in co-di-

gestion of tannery solid wastes. Clean Technologies and Environmental Policy

14, pp. 241–250. DOI: https://doi.org/10.1007/s10098-011-0391-z.

Kameswari, K. S. B., Kalyanaraman, C., Thanasekaran, K. (2011). Effect of ozo-

nation and ultrasonication pretreatment processes on co-digestion of tannery

solid wastes. Clean Technologies and Environmental Policy 13, pp. 517–525.

DOI: https://doi.org/10.1007/s10098-010-0334-0.

Kameswari, K. S. B., Kalyanaraman, C., Thanasekaran, K. (2014). Evaluation of

various pre-treatment processes on tannery sludge for enhancement of soluble

References

135

chemical oxygen demand. Clean Technologies and Environmental Policy 16,

pp. 369–376. DOI: https://doi.org/10.1007/s10098-013-0632-4. (a)

Kameswari, K. S. B., Kalyanaraman, C., Thanasekaran, K. (2015). Optimization

of organic load for co-digestion of tannery solid waste in semi-continuous mode

of operation. Clean Technologies and Environmental Policy 17, pp. 693–706.

DOI: https://doi.org/10.1007/s10098-014-0826-4

Kameswari, K. S. B., Kalyanaraman, C., Umamaheswari, B., Thanasekaran, K.

(2014). Enhancement of biogas generation during co-digestion of tannery solid

wastes through optimization of mix proportions of substrates. Clean Technolo-

gies and Environmental Policy 16, pp. 1067–1080. DOI:

https://doi.org/10.1007/s10098-013-0706-3. (b)

Kanagaraj, J., Velappan, K. C., Babu, N. K. C., Sadulla, S. (2006). Solid wastes

generation in the leather industry and its utilization for cleaner environment -

A review. Journal of Scientific & Industrial Research 65, pp. 541–548. DOI:

https://doi.org/10.1002/chin.200649273.

Katsifas, E. A., Giannoutsou, E., Lambraki, M., Barla, M., Karagouni, A. D.

(2004). Chromium recycling of tannery waste through microbial fermentation.

Journal of industrial microbiology biotechnology 31, pp. 57–62. DOI:

https://doi.org/10.1007/s10295-004-0115-z.

Kim, M. J., Kim, S. H. (2017). Minimization of diauxic growth lag-phase for

high-efficiency biogas production. Journal of Environmental Management 187,

pp. 456–463. DOI: https://doi.org/10.1016/j.jenvman.2016.11.002.

Klüver, E., Meyer, M. (2013). Preparation, processing, and rheology of thermo-

plastic collagen. Journal of Applied Polymer Science 128 (6), pp. 4201–4211.

DOI: https://doi.org/10.1002/app.38644.

Kovács, E., Wirth, R., Maróti, G., Bagi, Z., Rákhely, G., Kovács, K. L. (2013). Bi-

ogas production from protein-rich biomass: fed-batch anaerobic fermentation

of casein and of pig blood and associated changes in microbial community com-

position. PLoS ONE 8 (10), e77265. DOI: https://doi.org/10.1371/jour-

nal.pone.0077265.

References

136

Lalitha, K., Swaminathan, K. R., Padma Bai, R. (1994). Kinetics of biomethana-

tion of solid tannery waste and the concept of interactive metabolic control. Ap-

plied Biochemistry and Biotechnology 47, pp. 73–87. DOI:

https://doi.org/10.1007/BF02788677.

Lee, C., Kim, J., Shin, S. G., O’Flaherty, V., Hwang, S. (2010). Quantitative and

qualitative transitions of methanogen community structure during the batch

anaerobic digestion of cheese-processing wastewater. Applied Microbiology

and Biotechnology 87, pp. 1963–1973. DOI: https://doi.org/10.1007/s00253-010-

2685-1.

Li, C., Champagne, P., Anderson, B. C. (2011). Evaluating and modeling biogas

production from municipal fat, oil, and grease and synthetic kitchen waste in

anaerobic co-digestions. Bioresource Technology 102, pp. 9471–9480. DOI:

https://doi.org/10.1016/j.biortech.2011.07.103.

Li, Y., Jin, Y., Li, J., Li, H., Yu, Z. (2016). Effects of thermal pretreatment on the

biomethane yield and hydrolysis rate of kitchen waste. Applied Energy 172, pp.

47–58. DOI: http://dx.doi.org/10.1016/j.apenergy.2016.03.080.

Liew, L. N., Shi, J., Li, Y. (2011). Enhancing the solid-state anaerobic digestion

of fallen leaves through simultaneous alkaline treatment. Bioresource Technol-

ogy 102, pp. 8828–8834. DOI: http://dx.doi.org/10.1016/j.biortech.2011.07.005.

Liew, L. N., Shi, J., Li, Y. (2012). Methane production from solid-state anaerobic

digestion of lignocellulosic biomass. Biomass & Bioenergy 46, pp. 125–132. DOI:

http://dx.doi.org/10.1016/j.biombioe.2012.09.014.

Lima, D. Q., Oliveira, L. C. A., Bastos, A. R. R., Carvalho, G. S., Marques, J. G. S.

M., Carvalho, J. G., de Souza, G. A. (2010). Leather industry solid waste as ni-

trogen source for growth of common bean plants. Applied and Environmental

Soil Science 7. DOI: https://doi.org/10.1155/2010/703842.

López, I., Passeggi, M., Borzacconi, L. (2015). Validation of a simple kinetic

modelling approach for agro-industrial waste anaerobic digesters. Chemical

Engineering Journal Volume 262, pp. 509–516. DOI:

http://dx.doi.org/10.1016/j.cej.2014.10.003 (a)

References

137

López, I., Passeggi, M., Borzacconi, L. (2015). Variable kinetic approach to mod-

elling an industrial waste anaerobic digester. Biochemical Engineering Journal

96, pp. 7–13. DOI: http://dx.doi.org/10.1016/j.bej.2014.12.014. (b)

Marin, J., Kennedy, K. J., Eskicioglu, C. (2010). Effect of microwave irradiation

on anaerobic degradability of model kitchen waste. Waste Management 30, pp.

1772–1779. DOI: https://doi.org/10.1016/j.wasman.2010.01.033.

Mata-Alvarez, J., Dosta, J., Romero-Güiza, M. S., Fonoll, X., Peces, M., Astals, S.

(2014). A critical review on anaerobic co-digestion achievements between 2010

and 2013. Renewable and Sustainable Energy Reviews 36, pp. 412–427. DOI:

http://dx.doi.org/10.1016/j.rser.2014.04.039.

Mata-Alvarez, J., Macé, S., Llabrés, P. (2000). Anaerobic digestion of organic

solid wastes. An overview of research achievements and perspectives. Biore-

source Technology 74, pp. 3–16. DOI: https://doi.org/10.1016/S0960-

8524(00)00023-7.

McInerney, M. J. (1988). Anaerobic hydrolysis and fermentation of fats and pro-

teins, chapter 8. In: Zehnder, A. J. B., Ed. Biology of Anaerobic Microorganisms.

John Wiley and Sons, New York. ISBN: 0-471-88226-7.

Meyer, M. (2019). Processing of collagen based biomaterials and the resulting

materials properties. BioMedical Engineering OnLine 18:24. DOI:

https://doi.org/10.1186/s12938-019-0647-0.

Meyer M., Mühlbach, R., Harzer, D. (2005). Solubilisation of cattle hide collagen

by thermo-mechanical treatment. Polymer Degradation and Stability 87, pp.

137–142. DOI: https://doi.org/10.1016/j.polymdegradstab.2004.07.015.

Miles, C. A., Avery, N. C., Rodin, V. V., Bailey, A. J. (2005). The increase in de-

naturation temperature following cross-linking of collagen is caused by dehy-

dration of the fibres. Journal of Molecular Biology 346, pp. 551–556. DOI:

https://doi.org/10.1016/j.jmb.2004.12.001.

Miles, C. A., Ghelashvili, M. (1999). Polymer-in-a-box mechanism for the ther-

mal stabilization of collagen molecules in fibers. Biophysical Journal 76, pp.

3243–3252. DOI: https://doi.org/10.1016/S0006-3495(99)77476-X.

References

138

Misi, S. N., Forster, C. F. (2001). Batch co-digestion of multi-component agro-

wastes. Bioresource Technology 80, pp. 19–28. DOI:

https://doi.org/10.1016/S0960-8524(01)00078-5.

Mu, C., Lin, W., Zhang, M., Zhu, Q. (2003). Towards zero discharge of chro-

mium-containing leather waste through improved alkali hydrolysis. Waste

Management 23, pp. 835–843. DOI: https://doi.org/10.1016/S0956-

053X(03)00040-0.

Murphy, J. D. and Thamsiriroj, T. (2013). Fundamental science and engineering

of the anaerobic digestion process for biogas production, chapter 5. In:

Wellinger, A., Murphy, J., Baxter, D., Eds. The biogas handbook. Woodhead

Publishing Limited, Cambridge. ISBN: 9780857094988.

Nisman, B. (1954). The Stickland reaction. Bacteriological Reviews. 18(1), pp.

16–42.

Oliveira, L., Gonçalves, M., Oliveira, D., Guerreiro, M. C., Guilherme, L. R., Dal-

lago, R. M. (2007). Solid waste from leather industry as adsorbent of organic

dyes in aqueous-medium. Journal of Hazardous Materials 141(1), pp. 344–347.

DOI: https://doi.org/10.1016/j.jhazmat.2006.06.111.

Pati, A., Chaudhary, R., Subramani, S. (2014). A review on management of

chrome-tanned leather shavings: a holistic paradigm to combat the environ-

mental issues. Environmental Science and Pollution Research 21, pp. 11266–

11282. DOI: https://doi.org/10.1007/s11356-014-3055-9.

Paul, H., Antunes, A. P. M., Covington, A. D., Evans, P., Phillips, P. S. (2013).

Towards zero solid waste: utilising tannery waste as a protein source for poul-

try feed. 28th International Conference on Solid Waste Technology and Manage-

ment, Philadelphia, USA.

Penaud, V., Delgenes, J. P., Moletta, R. (1999). Thermo-chemical pretreatment

of a microbial biomass: influence of sodium hydroxide addition on solubiliza-

tion and anaerobic biodegradability. Enzyme and Microbial Technology 25, pp.

258–263. DOI: https://doi.org/10.1016/S0141-0229(99)00037-X.

References

139

Piccin, J., Feris, L. A., Cooper, M., Gutterres, M. (2013). Dye adsorption by

leather waste: mechanism diffusion, nature studies, and thermodynamic data.

Journal of Chemical and Engineering Data 58(4), pp. 873–882. DOI:

https://dx.doi.org/10.1021/je301076n.

Piccin, J., Gomes, C. S., Feris, L. A., Gutterres, M. (2012). Kinetics and isotherms

of leather dye adsorption by tannery solid waste. Chemical Engineering Journal

183, pp. 30–38. DOI: https://doi.org/10.1016/j.cej.2011.12.013.

Pillai, P., Archana, G. (2012). A novel process for biodegradation and effective

utilization of chrome shavings, a solid waste generated in tanneries, using chro-

mium resistant Bacillus subtilis P13. Process Biochemistry 47, pp. 2116–2122.

DOI: http://dx.doi.org/10.1016/j.procbio.2012.07.030.

Polster, A., Brummack, J. (2005). »Verbesserung von Entschwefelungsverfahren

in landwirtschaftlichen Biogasanlagen«. Dissertation. Technische Universität

Dresden Institut für Verfahrenstechnik und Umwelttechnik.

Priebe, G. P. S., Kipper, E., Gusmão, A. L., Marcilio, N. R., Gutterres, M. (2016).

Anaerobic digestion of chrome-tanned leather waste for biogas production.

Journal of Cleaner Production 129, pp. 410–416. DOI:

http://dx.doi.org/10.1016/j.jclepro.2016.04.038.

Rajput, A. A., Zeshan, Visvanathan, C. (2018). Effect of thermal pretreatment on

chemical composition, physical structure and biogas production kinetics of

wheat straw. Journal of Environmental Management 221, pp. 45–52. DOI:

https://doi.org/10.1016/j.jenvman.2018.05.011.

Ramsay, I. R., Pullammanappallil, P. C. (2001). Protein degradation during an-

aerobic wastewater treatment: derivation of stoichiometry. Biodegradation 12,

pp. 247–257. DOI: https://doi.org/10.1023/A:1013116728817.

Ravindranath, E., Chitra, K., Porselvam, S., Srinivasan, S. V., Suthanthararajan,

R. (2015). Green energy from the combined treatment of liquid and solid waste

from the tanning industry using an upflow anaerobic sludge blanket reactor.

Energy Fuels 29, pp. 1892−1898. DOI: https://doi.org/10.1021/ef5022686.

References

140

Reich, G. (1966). Kollagen eine Einführung in Methoden, Ergebnisse und Prob-

leme der Kollagenforschung. Verlag Theodor Steinkopff, Dresden. DOI:

https://doi.org/10.1002/zfch.19680080328.

Reich, G. (2007). From collagen to leather – the theoretical background. BASF

Service center Media and Communications, Ludwigshafen.

Schink, B., Stams, A. J. M. (2013). Syntrophism among prokaryotes. In: Rosen-

berg, E., DeLong, E. F., Lory, S., Stackebrandt, E., Thompson, F., Eds. The Pro-

karyotes: Prokaryotic communities and ecophysiology. Springer, Berlin. ISBN:

978-3-642-30122-3.

Schroepfer, M., Meyer, M. (2017). DSC investigation of bovine hide collagen at

varying degrees of crosslinking and humidities. International Journal of Biolog-

ical Macromolecules 103, pp. 120–128. DOI: http://dx.doi.org/10.1016/j.ijbi-

omac.2017.04.124.

Schröpfer, M. (2012). Influences on Thermal Stability of Fibrous Collagen - Cal-

orimetric Investigations. 5th Freiberg Collagen Symposium, Freiberg, Germany.

Schuberth-Roth, V. T. (2013). Erster Schritt in neue Energie-Ära. Frankenpost,

Rehauer Tagblatt. URL: https://www.frankenpost.de/lokal/hof-

rehau/rehau/Erster-Schritt-in-neue-Energie-Aera;art2452,2627816 (accessed

09.09.2019).

Shanmugam, P., Horan, N. J. (2009). Optimising the biogas production from

leather fleshing waste by co-digestion with MSW. Bioresource Technology 100,

pp. 4117–4120. DOI: https://doi.org/10.1016/j.biortech.2009.03.052.

Sirota, S., Dorozhkin, S. V., Kruchinina, M. V., Melikhov, L. V. (1992). Phase

transformation and dehydration of calcium sulphate dihydrate in solution stud-

ied by SEM. Scanning 14, pp. 269–275. DOI:

https://doi.org/10.1002/sca.4950140505.

Song, Z., Yag, G., Feng, Y., Ren, G., Han, X. (2013). Pretreatment of rice straw

by hydrogen peroxide for enhanced methane yield. Journal of Integrative Agri-

culture 12, 7, pp. 1258–1266. DOI: https://doi.org/10.1016/S2095-3119(13)60355-

References

141

Stegemann, H. (1958). Microdetermination of hydroxyproline with chloramine-

T and p-dimethylaminobenzaldehyde. Hoppe-Seyler's Zeitschrift fur physiolo-

gische Chemie 311, 41.

Suzuki, Y., Tsujimoto, Y., Matsui, H., Watanabe, K. (2006). Decomposition of

extremely hard-to-degrade animal proteins by thermophilic bacteria. Journal of

Bioscience and Bioengineering 102 (2), pp. 73–81. DOI:

https://doi.org/10.1263/jbb.102.73.

Thangamani, A., Parthiban, L., Rangasamy, P. (2015). Two-phase anaerobic di-

gestion model of a tannery solid waste: experimental investigation and model-

ing with ANFIS. Arabian Journal for Science and Engineering 40, pp. 279–288.

DOI: https://doi.org/10.1007/s13369-014-1408-9.

Thangamani, A., Rajakumar, S., Ramanujam, R. A. (2010). Anaerobic co-diges-

tion of hazardous tannery solid waste and primary sludge: biodegradation ki-

netics and metabolite analysis. Clean Technologies and Environmental Policy

12, pp. 517–524. DOI: https://doi.org/10.1007/s10098-009-0256-x.

The UniProt Consortium. URL: https://www.uniprot.org (accessed 01.02.2018).

Trommer, B., Kellert, H. -J. (1999). Comparison of tanning methods from an eco-

logical viewpoint (Part 1). Leather 201 (12), pp. 37–44.

Tsavkelova, E. A., Netrusov, A. I. (2012). Biogas production from cellulose-con-

taining substrates: a review. Applied Biochemistry and Biotechnology 48, 5, pp.

421–433. DOI: https://doi.org/10.1134/S0003683812050134.

Usha, R., Ramasami, T. (2000). Effect of crosslinking agents (basic chromium

sulfate and formaldehyde) on the thermal and thermomechanical stability of rat

tail tendon collagen fibre. Thermochimica Acta 356, pp. 59–66. DOI:

https://doi.org/10.1016/s0040-6031(00)00518-9.

VDI 4630 (2006). Fermentation of organic materials, Characterization of the sub-

strate, sampling, collection of material data, fermentation tests. The Association

of German Engineers, Düsseldorf, Germany.

References

142

Walter, A., Silberberger, S., Juárez, M. F-D., Insam, H., Franke-Whittle, I. H.

(2016). Biomethane potential of industrial paper wastes and investigation of the

methanogenic communities involved. Biotechnology for Biofuels 9, 21. DOI:

https://doi.org/10.1186/s13068-016-0435-z.

Wess, T. J. (2008). Collagen fibrillar structure and hierarchies, chapter 3. In:

Fratzl P., Ed. Collagen: structure and mechanics. Springer, New York. ISBN:

978-0-387-73905-2.

Zhang, M., Liao, X., Shi, B. (2006). Adsorption of surfactants on chromium

leather waste. Journal of the Society of Leather Technologists and Chemists

90(1), pp. 1–6.

Zhang, M., Shi, B. (2004). Adsorption of dyes from aqueous solution by chro-

mium-containing leather waste. Journal of the Society of Leather Technologists

and Chemists 88(6), pp. 236–241.

Zupančič, G. D., Jemec, A. (2010). Anaerobic digestion of tannery waste: Semi-

continuous and anaerobic sequencing batch reactor processes. Bioresource

Technology 101, pp. 26–33. DOI: https://doi.org/10.1016/j.biortech.2009.07.028.

Zwietering, M. H., Jongenburger, I., Rombouts, F. M., Riet, K. V. (1990). Model-

ing of the bacterial growth curve. Applied and Environmental Microbiology,

56(6), pp. 1875–1881.

Appendix

143

Appendix

144

Appendix

145

Fig. 5.1: SEM images of shavings extruded dry at 130 °C (a), shavings extruded dry at 150 °C

(b), shavings extruded dry at 170 °C (c), shavings extruded wet at 100 °C (d), shavings extruded

wet at 130 °C (e), shavings extruded wet at 150 °C (f), shavings extruded wet at 170 °C (g),

shavings treated hydrothermally at 150 °C (h), shavings autoclaved over a period of six minutes

(i), shavings autoclaved over a period of twelve minutes (j), and shavings autoclaved over a pe-

riod of 24 minutes (k), and shavings autoclaved over a period of 48 minutes (l), and shavings

autoclaved over a period of 96 minutes (m), and shavings autoclaved over a period of 192 minutes

(n) at 60,000 × magnification.