http://researchcommons.waikato.ac.

nz/

Research Commons at the University of Waikato

Copyright Statement:
The digital copy of this thesis is protected by the Copyright Act 1994 (New Zealand).

The thesis may be consulted by you, provided you comply with the provisions of the
Act and the following conditions of use:

Any use you make of these documents or images must be for research or private
study purposes only, and you may not make them available to any other person.
Authors control the copyright of their thesis. You will recognise the authors right
to be identified as the author of the thesis, and due acknowledgement will be
made to the author where appropriate.
You will obtain the authors permission before publishing any material from the
thesis.
 BIOFIBRE PRODUCTION
FROM
CHICKEN FEATHER

A thesis
submitted in fulfilment
of the requirements for the degree
of
Master of Engineering in Materials and Process Engineering
at
The University of Waikato
by
FAN-CHEN JEANIE TSENG

The University of Waikato

2011
Abstract
The global poultry industry generates at least 2 million tonnes of chicken feather
every year. Feathers are currently hydrolysed into meal used for animal feed and
fertiliser. Feather consists of 91% keratin, 1% lipid and 8% water. Raw feather
also contains preen oil, offal, faecal matter and poultry processing water. Its
morphology consists of barbs extending at an angle from a central hollow rachis.
Impurities coat the entire feather, and particulates are trapped by layers of
barbules and hooked barbicels holding adjacent barbs together. These
substructures present an extensive and tortuous hydrophobic surface. Feather fibre
is a multipurpose, cost effective reinforcement for polymer composites. Its
incorporation in plastic, wood, concrete and cardboard makes the product lighter,
insulate from heat loss and improve sound attenuation.

The objective of this study was to develop a process to produce clean fibre
recovered from chicken feather. In the treatment process, the heterogeneous
characteristics of feather had to be considered. Raw feather was suspended in 25 L
water in a pulper to be decontaminated using 2 stages of 0.1485% sodium
hypochlorite adjusted to pH 10.0 with 1 M sodium hydroxide and cleaned in 3
stages of 0.15% hydrogen peroxide. The pulper disc impeller agitated the
suspension at 10 Hz for 30 minute at each stage. Bacteriological tests confirmed
pathogens such as Campylobacter, Salmonella and Enterobacteriaceae were
removed during treatment. Off-white clean feathers were more than 10% whiter
than dull yellow raw feather.

Cleaned feather was comminuted in 300 L water using a centrifugal pump at a

flow rate of 30 Hz on full recycle for 4 hours. Rachis and partially cut feather
were removed using a 5 mm aperture filter and fibre was recovered using a 1 mm
filter. Wet fibre was dried to constant mass in an air-forced oven at 70C. Fibre
yield was 27% of feather input, or 54% of theoretical yield. Surface morphology
showed no damage.

i
Acknowledgement
I would like to express my appreciation to the following people for their support
during my studies.

Firstly, I would like to thank my family for many years of care. I am grateful for
the opportunities you have made possible through your sacrifices and values you
have instilled in me. Thank you to friends for your sound advice, patience and
being good company.

I would like to thank my academic supervisors Dr. Johan Verbeek, Assoc. Prof.
Kim Pickering and Dr. Mark Lay. I am grateful for your inspirational guidance in
making this project a valuable learning experience.

Thank you to Wallace Corporation, especially mentor Neville Cross and plant
manager Ashley Boon, for providing industry support.

I would also like to thank Dr. Michael Walmsley, Dr. Martin Atkins, and Timothy
Walmsley from the Energy Research Group for their insight into processing
operations in the pulp and paper industry

Thank you for technical support from the engineering department to Stewart
Finlay, Alan Smith, Yuangi Zhang, Chris Wang, Lisa Li, Indhar Singh and Luke
van de Pas.

Thank you also to Wendy Jackson from the chemistry department for Soxhlet
Extraction facilities, Dr. Barry OBrien from the biology department for
microscopy advice and Helen Turner for Scanning Electron Microscopy facilities.

Finally, I would like to thank TechNZ for granting me a Capability Fellowship for
financial support.

ii
Contents
Abstract ................................................................................................................ i
Acknowledgement ............................................................................................... ii
Contents ............................................................................................................. iii
Figures ................................................................................................................ vi
Tables ............................................................................................................... viii
1 Introduction............................................................................................... 2
2 Chicken Feather ........................................................................................ 5
2.1 The Poultry Industry.................................................................................. 5
2.1.1 Poultry Processing ................................................................................. 6
2.1.2 Poultry Rendering .................................................................................. 9
2.2 Feather Morphology ................................................................................ 10
2.2.1 Development and Differentiation ......................................................... 11
2.2.2 Classification ....................................................................................... 13
2.2.3 Keratin................................................................................................. 13
2.2.4 Lipids .................................................................................................. 15
2.3 Comparison with Other Fibre .................................................................. 15
2.3.1 Synthetic Fibre .................................................................................... 15
2.3.2 Plant Fibre ........................................................................................... 16
2.3.3 Animal Fibre ....................................................................................... 17
2.4 Feather Fibre Applications ...................................................................... 18
2.4.1 Polymer and Composites ..................................................................... 18
2.4.2 Adsorbent in Bioremediation ............................................................... 20
2.4.3 Enzymatic Applications ....................................................................... 20
2.4.4 Crop Fertiliser and Plant Antimicrobial Control ................................... 20
2.4.5 Animal Products .................................................................................. 21
2.5 Fibre Production from Feather ................................................................. 21
2.5.1 Decontamination.................................................................................. 22
2.5.2 Cleaning .............................................................................................. 22
2.5.3 Effluent ............................................................................................... 24
2.5.4 Comminution ....................................................................................... 25
2.5.5 Filtration .............................................................................................. 25
2.5.6 Drying ................................................................................................. 27
3 Experimental ........................................................................................... 30

iii
3.1 Materials ................................................................................................. 30
3.2 Equipment ............................................................................................... 30
3.2.1 Lab-scale ............................................................................................. 31
3.2.2 Large Scale .......................................................................................... 32
3.3 Methods .................................................................................................. 36
3.3.1 Lab-scale Decontamination and Cleaning ............................................ 36
3.3.2 Lab-scale Comminution ....................................................................... 36
3.3.3 Large Scale Decontamination .............................................................. 36
3.3.4 Large Scale Cleaning ........................................................................... 36
3.3.5 Large Scale Comminution ................................................................... 37
3.3.6 Large Scale Fibre Recovery ................................................................. 38
3.3.7 Analysis ............................................................................................... 40
3.4 Experimental Plan ................................................................................... 42
3.4.1 Lab-Scale Decontamination ................................................................. 42
3.4.2 Lab-Scale Cleaning .............................................................................. 42
3.4.3 Lab-Scale Comminution ...................................................................... 44
3.4.4 Large Scale Decontamination and Cleaning ......................................... 44
3.4.5 Large Scale Comminution ................................................................... 44
4 Results and Discussion ............................................................................ 46
4.1 Lab-scale Decontamination ..................................................................... 46
4.2 Lab-scale Feather Processing................................................................... 47
4.3 Lab-scale Analysis .................................................................................. 48
4.3.1 Impurity Evaluation ............................................................................. 48
4.3.2 Colourimetry ....................................................................................... 51
4.3.3 Morphology ......................................................................................... 55
4.4 Large Scale Treatment............................................................................. 57
4.4.1 Bacteriological Test ............................................................................. 58
4.4.2 Colourimetry ....................................................................................... 58
4.4.3 Macroscopic Morphology .................................................................... 60
4.4.4 Stereoscopic Morphology .................................................................... 63
4.4.5 Histological Morphology ..................................................................... 68
4.5 Large Scale Comminution ....................................................................... 70
4.5.1 Macroscopic Morphology .................................................................... 74
4.5.2 Stereoscopic Morphology .................................................................... 76

iv
4.5.3 Histological Morphology ..................................................................... 81
5 Conclusions............................................................................................. 83
5.1 Recommendations ................................................................................... 84
6 References............................................................................................... 85
Appendices ........................................................................................................ 94
Appendix 1: Chicken Feather ............................................................................. 95
Appendix 2: Analysis Procedures....................................................................... 98
Appendix 3: Lab-scale Treatment .................................................................... 101
Appendix 4: Large-scale Treatment ................................................................. 104

v
Figures
Figure 1: Broiler chickens in poultry farm. ........................................................... 6
Figure 2: Typical poultry processing procedure [14]. ............................................ 8
Figure 3: Poultry anatomy [20]. ........................................................................... 9
Figure 4: Feather hierarchical structure. ............................................................. 11
Figure 5: Feather formation [26]. ....................................................................... 12
Figure 6: Feather classification. ......................................................................... 13
Figure 7: Amino acid chemical skeletal structures. ............................................. 14
Figure 8: Beloit pressure screen. ........................................................................ 27
Figure 9: Raw feather receiving at Wallace Corporations Waitoa rendering plant.
.................................................................................................................. 30
Figure 10: Boltac 6-unit mixer (left)................................................................... 31
Figure 11: Lab-scale comminution equipment. ................................................... 32
Figure 12: Lamort pulper. .................................................................................. 33
Figure 13: Comminution system. ....................................................................... 34
Figure 14: Scanpump 310-0133. ........................................................................ 34
Figure 15: Tall, wide and fine filters used during fibre recovery. ........................ 35
Figure 16: Decontamination block flow diagram. ............................................... 37
Figure 17: Cleaning block flow diagram. ........................................................... 37
Figure 18: Comminution block flow diagram. .................................................... 38
Figure 19: Fibre recovery block flow diagram. ................................................... 39
Figure 20: Hitachi S4000 series FE-SEM [3]. .................................................... 40
Figure 21: Soxhlet extractor apparatus [4]. ......................................................... 41
Figure 22: Konica Minolta CR410 Chroma meter sensor [6]. ............................. 42
Figure 23: Raw feather. ...................................................................................... 46
Figure 24: Feather treated by 0.525% Janola for 5 minutes. ............................... 47
Figure 25: Hexane extractable content after ethanol cleaning. ............................ 49
Figure 26: Hexane extractable content after hydrogen peroxide cleaning. ........... 50
Figure 27: Hexane extractable content after SDS cleaning. ................................ 50
Figure 28: Feather whiteness after Janola treatment. .......................................... 51
Figure 29: Feather whiteness after ethanol treatment. ......................................... 52
Figure 30: Feather whiteness comparison after mixing and treatment. ................ 54
Figure 31: Feather morphology after 5 min or 10 min Janola treatment. ............. 55
Figure 32: Feather morphology after ethanol lab-scale treatment........................ 56

vi
Figure 33: Feather morphology after hydrogen peroxide lab-scale treatment. ..... 57
Figure 34: Large scale treatment average feather whiteness. ............................... 59
Figure 35: Feather photographs after each treatment step. .................................. 62
Figure 36: Morphology of untreated feather. ...................................................... 63
Figure 37: morphology after first stage decontamination. ................................... 64
Figure 38: Morphology after second stage decontamination. .............................. 65
Figure 39: Morphology after first stage cleaning. ............................................... 66
Figure 40: Morphology second stage cleaning. ................................................... 67
Figure 41: Morphology after third stage cleaning. .............................................. 68
Figure 42: Untreated, decontaminated and clean feather surface morphology. .... 69
Figure 43: Feather deposition in impeller chamber. ............................................ 72
Figure 44: Comminution centrifugal pump impeller. .......................................... 72
Figure 45: Comminution slurry temperature change through time. ..................... 74
Figure 46: Dried and packaged fibre and feather from comminution. ................. 74
Figure 47: Comminution feather and fibre photographs. ..................................... 75
Figure 48: Feather morphology before comminution. ......................................... 76
Figure 49: Feather and fibre morphology after 1 hour of comminution. .............. 77
Figure 50: Feather and fibre morphology after 2 hours of comminution. ............ 78
Figure 51: Feather and fibre morphology after 3 hours of comminution. ............ 79
Figure 52: Feather and fibre morphology after 4 hours of comminution. ............ 80
Figure 53: Fibre surface morphology in comminution. ....................................... 82

vii
Tables
Table 1: Poultry by-product composition [3] ...................................................... 10
Table 2: Lipids in chicken keratin, tissue and skin [31]. ..................................... 15
Table 3: Chicken feather fibre properties [5]. ..................................................... 15
Table 4: Synthetic fibre properties [11] .............................................................. 16
Table 5: Plant fibre species, annual production and earning [33]. ....................... 16
Table 6: Plant fibre physical properties [4]. ........................................................ 17
Table 7: Plant fibre mechanical properties [4] .................................................... 17
Table 8: Animal fibre species, annual production and earning [33] ..................... 17
Table 9: Animal fibre physical properties [4] ...................................................... 18
Table 10: Animal fibre mechanical properties .................................................... 18
Table 11: Filter dimensions ................................................................................ 35
Table 12: Janola concentrations and residence time ............................................ 42
Table 13: Low concentration ethanol cleaning .................................................... 42
Table 14: Tween-80 cleaning.............................................................................. 43
Table 15: Hydrogen peroxide cleaning ............................................................... 43
Table 16: Sodium dodecyl sulfate cleaning......................................................... 43
Table 17: Higher concentration ethanol cleaning ................................................ 43
Table 18: Large scale treatment analysis ............................................................ 44
Table 19: Feather colour after Janola decontamination. ...................................... 51
Table 20: Feather colour after ethanol cleaning. ................................................. 52
Table 21: Feather colour after mixing and treatment. .......................................... 54
Table 22: Bacterial test results for feather treatment. .......................................... 58
Table 23: Feather colour after decontamination and cleaning steps. .................... 59
Table 24: Feather colour after mixing with Lamort pulper impellers................... 60
Table 25: Comminution mass balance on dry basis............................................. 70
Table 26: Comminution moisture content and fibre yield. .................................. 73

viii
1 Introduction

1
1 Introduction
Polymers are used in almost every area of society such as packaging, transport,
construction and casings. Polymers are easily shaped by extrusion, injection
moulding, vacuum forming or foaming. It is durable, environmentally resistant,
tough and light. Tailoring mechanical polymer properties for specific purposes
often require fibre reinforcement. Common synthetic fibres include carbon,
aramid and glass while natural fibres such as wood, hemp and sisal are have also
been shown to be effective. An alternative to the aforementioned fibres is fibre
recovered from chicken feathers, as they are widely available and has good
mechanical properties [1].

New Zealand exported NZ$6.5 billion worth of meat products, NZ$1.2 billion co-
products and 150,000 tonnes of poultry products in 2010 [2, 3]. The major by-
product in poultry processing is chicken feather, of which Wallace Corporation
(Waitoa, Morrisville) processes around 10,400 tonnes of wet chicken feather
every year. Chicken feather is rendered and ground into feather meal as an
additive for feed stock or as a fertiliser, it currently sells for NZ$569 622 per
tonne [4].

Feather structure is an ordered hierarchy comprising a central rachis from which

barbs extend at an angle, with parallel barbules held together by hooked barbicels.
The barbs are approximately 15 to 110 m wide and 3 to 13 mm long, resulting in
an average aspect ratio of 212. Barbules are 4 to 10 m wide, 60 m to 1 mm
long, resulting in an average aspect ratio of 61. Barbs have an average tensile
strength of 113 MPa, with an average elastic modulus of 2.8 GPa and has a 7.7%
elongation at break [5].

Chicken feather is approximately 91% keratin, which is rich in cysteine bonds and
its hydrophobic side chains make it waterproof. Keratin is tough, strong and
chemically resistant, but as a feed additive it lacks methionine, histidine and
lysine, which are essential nutrients for animals. As a fertiliser, it contains excess
nitrogen [1].

Raw chicken feather contain high levels of Salmonella, Campylobacter and

Enterobacteriaceae [6]. In addition, they are discoloured with process water, offal,
chicken feed, excrement, and coated with lipids from the preen gland and

2
epidermal sebaceous secretions.

Useable fibres are recovered barbs removed from the rachis of decontaminated
and cleaned feathers. The objective of this study was to assess various
decontamination and cleaning strategies, thereby identifying suitable chemicals
and concentrations thereof. Secondly, a suitable comminution process will be
identified and tested for its efficiency to produce fibres suitable for use in polymer
composites.

3
2 Chicken Feather

4
2 Chicken Feather
2.1 The Poultry Industry
The global poultry industry is enormous, producing approximately 2 million
tonnes of chicken feather every year. There are many organisations involved in
this global industry, with the Worlds Poultry Science Association (WPSA)
representing a research network dealing with issues such as breeding, nutrition,
welfare, husbandry, production, processing, product development, physiology,
product quality and economics [7].

In the United States, the Department of Agriculture (USDA) is responsible for

developing food information, economic opportunities and monitoring sustainable
natural resource use. The US poultry industry produced 22 million tonnes broiler
chickens in 2010, stimulating research into manufacturing processes for
recovering fibre from chicken feather to produce specialty products [8, 9].

In New Zealand, the Institute of Food Science and Technology (NZIFST) links the
food industry to researchers dealing with food processing and distribution and
provides news and information about food manufacturing, packaging, processing
and safety [10]. More specifically, institutions like AgResearch (formerly The
Meat Industry Research Institute of New Zealand or MIRINZ) deals with ways to
add value to bio-based products by increasing market access, ensuring food safety
and prolonging shelf-life [11].

Poultry meat producers are represented by The Poultry Industry Association of

New Zealand (PIANZ) ensure standards compliance, consumer wellbeing and
animal welfare. Its main task is to maintain industry viability and to prevent avian
disease outbreaks [12]. PIANZ works closely with The Ministry of Agriculture
and Forestry (MAF), which manages the primary sector for farmers, food and bio-
based product manufacturers and exporters. It is tasked to prevent biosecurity
threats that could be damaging the environment, ensuring New Zealanders can
benefit from sustainable use of natural resources [13]. Poultry products are also
monitored by AssureQuality New Zealand, a commercial company providing food
inspection, certification and diagnostics for the primary production sector [14].

5
Major poultry producers in New Zealand are Inghams Enterprises (NZ) Pty. Ltd.,
Tegel Foods Ltd., and PH van den Brink Ltd., which comprise 90% of poultry
production in New Zealand. Inghams is a Waikato-based Australasian subsidiary
employing roughly 1,000 people in New Zealand and have eleven feed mills, ten
primary processing plants and nine processing plants that also make pet food
ingredients [15]. Tegel Foods is a fully integrated poultry producer doing poultry
breeding, hatching, processing, marketing and distribution, it is owned by the
multinational Affinity Equity Partners. They have four main processing plants and
employ about 1,700 people [16]. Brinks Chicken is a family business selling retail
fresh and frozen chicken and specialty chicken products [17]. Kakariki Proteins, a
subsidiary of Turks Poultry Farm Ltd. (est. 1966) in Foxton, Kapiti produces
eggs, poultry and does offal rendering [18]. In Waikato, NZ, Wallace Corporation
is one of the largest renderers of chicken off and is the largest producer of chicken
feather.

2.1.1 Poultry Processing

Meat chickens are called broilers (Figure 1), while egg producing hens are layers.
Breeders are selected for rapid growth rate and an efficient feed to weight gain
ratio. Chicks can grow from a hatch weight of 45 g to 2.2 kg after 42 to 45 days at
which point they are slaughtered and processed [19].

Figure 1: Broiler chickens in poultry farm.

Poultry processing is summarised in Figure 2, starting with live birds up to meat

packaging and storage. After killing, feathers are removed by mechanical pluckers
fitted with rubber fingers on rotating discs and the process is finished by operators

6
called pinners, who manually finish plucking. Feathers are pumped over a
separation screen into a container yielding a mixture of dilute blood, grease,
cleaning water and feathers.

7
Figure 2: Typical poultry processing procedure [14].

8
2.1.2 Poultry Rendering
Parts not suitable for human consumption are shown in Figure 3A which includes
the viscera (or internal organs), while Figure 3B depicts the skeletal structure.
Heads, beaks and feet are often mixed up with the feather. Whole chicken and
edible cuts are shown in Figure 3C.

Figure 3: Poultry anatomy [20]. A Chicken internal organs. B Skeletal structure. C Whole
Chicken and edible cuts.

9
Feathers, blood, offal and trims are rendered into meal by an energy-intensive
pressure cooking process at 142oC for 30 minutes. Feather meal is not a complete
animal food because it is missing essential amino acids, and is indigestible unless
hydrolysed. Commercial feather meal has a composition shown in Table 1. Amino
acid supplements are often added to improve feed quality, resulting in animal feed
comparable to soybean meal [21].

Table 1: Poultry by-product composition [3]

By-Product Poultry blood meal Feather meal Poultry by-product meal
Moisture 7.0 7.0 7.0
Protein 88.9 81.0 60.0
Fat 1.0 7.0 13.0
Calcium 0.41 0.33 3.00
Phosphorus 0.30 0.55 1.70

Feather not converted to animal feed is buried or burned. Incineration releases

dioxin to the atmosphere and leachate from agricultural land disposal could lead
to nitrate contamination in ground water.

New Zealand renderers are exploring opportunities to increase revenue by

developing alternative applications for their products. Some applications are as
feedstock for plastic production, adsorbent for bioremediation, enzyme additives
for detergents, antimicrobial fertiliser and a natural alternative for parts of the
leather tanning process [22]. Feather, blood, fat and carcass tissues should be
considered by-products with potential to increase profit rather than waste streams
for disposal [23].

2.2 Feather Morphology

Feather makes up about 8.5% of a chickens mass. Feathers function as insulation,
protection, waterproofing, colouration and flight [24]. An epidermal sheath
develops to form a short basal tube, or calamus and the main shaft or rachis. Barbs
extend at an angle followed by barbules. Barbules have hooks called barbicels to
connect with barbules on an adjacent barb. Muscle fibres connect the shaft bases
to raise and lower feathers (Figure 4) [25].

10
Figure 4: Feather hierarchical structure. Central shaft rachis, Angled extensions barbs, and
barbules and barbicels [25].

2.2.1 Development and Differentiation

Feather, or avian integument, develops from the rachis, barbules then small
hooked cilia, described in Figure 5. The follicle has a dermal papilla and an
epidermal collar. The bone morphogenetic protein (BMP4) promotes rachis
formation and barb fusion while the noggin enhances rachis and barb branching.
The Sonic hedgehog (Shh) gene induces marginal plate epithelia apoptosis to form
spaces between barbs [26].

11
Figure 5: Feather formation [26]. A Epithelial cells above mesenchymal cells. B Noggin and
BMP4 promote barb ridge formation. C Shh promote barbules and rami formation. D BMP4
and BMP2 signal for barb and barbules to separate. E Apoptosis by Shh. F Differentiation by
BMP4 and BMP2 forms barbs and barbules or no barbs. G Differentiated feather top view. H
Another top view showing follicles developing into barbs or rachis with barbs. I Feather follicle
cross section.

12
2.2.2 Classification
Figure 6 illustrate types of feathers. Contour feathers (Figure 6A) cover the body
surface and include flight feather which are constantly rearranged with sebum
from the uropygeal, or preen gland to ensure barbules stay interlocked. Powder
feathers are white particles that waterproof contour feathers. Down feathers
(Figure 6B) provide insulation and do not have barbicels. Semi-plumes (Figure
6C) are similar to, but larger than down feathers. Filo-plumes (Figure 6D), at the
base of every contour feather are rich in nerve endings that relate proprioceptive
input or feather position [25].

Figure 6: Feather classification. A Contour feather maintains shape. B Down feather provides
insulation. C Semi-plumes are a cross between contour and down. D Filo-plumes have nerve
endings at the tip to help with motion. E Bristles around the eyes and beak acts as tactile sensors.

2.2.3 Keratin
Chicken feather is approximately 91% keratin, 8% water and 1% lipids by mass
[22]. Keratin is a protein with 95 amino acids and has a molecular weight of
10.168 kDa. About half the amino acids found in feather keratin are hydrophilic,
but the waterproof nature of the fibre implies exterior hydrophobic side chains.
Figure 7 shows the chemical structures of amino acids found in keratin. Keratin

13
contains about 12% serine, a polar amino acid as well as 7% of hydrophobic
amino acids, valine and leucine. Aspartic acid has a negatively charged side chain,
while only trace amounts of electrically charged histidine and lysine are present
and no methionine at all. In Figure A 1 to Figure A 4 the amino acid composition
of proteins found in different feather components are shown [27].

Figure 7: Amino acid chemical skeletal structures.

Infrared spectroscopy and x-ray crystallography suggest rachis to have more -

sheet keratins aligned in arrayed layers, while the barbs, barbules and barbicels
are in ordered -helix conformation [1]. Keratin filaments are right-handed helices
with four repeating units per turn. Each unit has two left-hand-twisted -sheets
connected by a perpendicular diad. The filament has internal hydrophobic residues

14
and external charged cysteine residues [28].

2.2.4 Lipids
The preen gland secretes lipids to maintain the feathers physical properties such
as waterproofing and colour. Pigment cells called melanocytes are responsible for
the dull yellow colour of feather [29]. Free fatty acids from lipid decomposition
can lead to pH changes and lead to microbial growth after slaughter. Preen oil
composition is affected by age, gender, diet and contributes to feather taste. The
secretion has antimicrobial properties against Bacillus lichenformis, which would
degrade keratin. Other chemicals contributing to aroma, such as alcohols, ketones
and diones attracts mosquitos [30]. Lipids extracted from different components
and analysed using quantitative thin-layer chromatography are shown in Table 2
[31].

Table 2: Lipids in chicken keratin, tissue and skin [31].

Chicken Component Lipids
Keratin Ceramides, acylceramides, cholesteryl sulfate, glucosylsterols
and acylglucosylsterols
Tissue Lots: Wax diesters, triglyceride and free sterols
Some: Phospholipids, steryl ester and free fatty acids
Full thickness Acylglucosylceramides
epidermis

2.3 Comparison with Other Fibre

Very few studies have looked into the properties of chicken feather fibres; Yang
and Shah have studied fibre supplied by Featherfiber Corporation using a patented
process [32]. The properties of the fibre recovered are summarised in Table 3.

Table 3: Chicken feather fibre properties [5].

Property Value
Density (g cm-1) 0.89
Modulus (GPa) 2.8
Strength (MPa) 113
Elongation (%) 7.7
Component Barbs Barbules
Diameter (m) 15 110 4 10
Length (mm) 3 13 0.060 1
Aspect ratio 212 61

2.3.1 Synthetic Fibre

Some physical and mechanical properties of synthetic fibres are listed in Table 4.

15
Synthetic fibres have higher modulus and strength, implying that feather fibre by
itself is not suitable for high performance applications. It is much less dense than
synthetic fibres, making it an excellent candidate for light-weight composites such
as car panels.

Table 4: Synthetic fibre properties [11]

Density Modulus Strength Elongation
Name
(g cm-1) (GPa) (MPa) (%)
Aramid 1.4 63 67 3000 3150 3.3 3.7
Carbon 1.4 230 240 4000 1.4 1.8
E-glass 2.5 70 2000 3500 2.5
Nylon 6 1.09 1.14 200 300 35 70 20 50
Saran 1.67 1.71 100 150 30 40 20 40
S-glass 2.5 86 4570 2.8
Vinal 1.25 1.35 300 1200 50 75 17 26
Vinyon 1.34 1.36 350 450 25 30 20 30

2.3.2 Plant Fibre

More than 29 million tonnes of plant fibre is produced worldwide each year. A
selection of plant fibre and estimated annual production are listed in Table 5.
Some of these plant fibres are the main source of income for farmers in
developing countries. Fibres like abaca, sisal and coir earns $30, $75 and $100
million per year, respectively [33].

Table 5: Plant fibre species, annual production and earning [33].

Name Species Annual Production (MT)
Plant Fibre Approximately 29
Abaca Musa textiles 0.07
Coir Cocos nucifera 0.5
Cotton Gossypium genus 25
Flax Linum usitatissimum 0.147
Hemp Cannabis sativa 0.113
Jute Corchorus capsularis 2.3 to 2.8
Ramie Boehmeriz nivea 0.28
Sisal Agave sisalana 0.3

Table 6 shows feather barbs have similar diameter range to most plant fibres,
while barbules are much thinner. Plant fibres can be several metres long, but
feather barb lengths are comparable to short cotton fibres, which is the most
commonly used fibre for fabric.

16
Table 6: Plant fibre physical properties [4].
Name Diameter (m) [33] Length (m) [33] Density (g cm-1)
Plant Fibre 11 300 1.15 1.55
Abaca [11] 17 21 Up to 3 -
Coir 17 21 Up to 0.35 1.15 1.46
Cotton 11 22 0.01 to 0.065 1.5
Flax [10] 17 21 0.025 1.5
Hemp [34] 18 38 18 40 1.5
Jute 16 20 1 to 4 1.46
Ramie [35] 25 30 0.012 1.55
Sisal 100 300 Up to 1 1.45

The mechanical properties of some plant fibres are listed in Table 7. Feather fibre
has lower modulus, or stiffness, and similar tensile strength as coir. Plant fibres
have at least 2 times higher tensile strength.

Table 7: Plant fibre mechanical properties [4]

Name Modulus (GPa) Strength (MPa) Elongation (%)
Plant Fibre 4 128 130 1500 1.2 40
Abaca [11] 33.6 813 -
Coir 46 130 220 15 40
Cotton 5.5 27.6 300 1500 38
Flax 27.6 345 1500 2.7 3.2
Hemp [34] 70 690 1.6
Jute 10 55 200 800 1.16 8
Ramie [35] 61.4 128 400 938 1.2 3.8
Sisal 9.4 22 468 700 27
2.3.3 Animal Fibre
About 2.3 million tonnes of animal fibre is produced in the world every year, most
of which is wool from sheep. Alpaca wool has a market share of $50 million a
year, while wool earns about $80 billion a year. Annual production of some
animal-based fibres is listed in Table 8.

Table 8: Animal fibre species, annual production and earning [33]

Name Species Annual Production (kT)
Animal Fibre 2,300
Alpaca wool Lama pacos 6.5
Angora wool Oryctolagus cuniculus 2.5 3.0
Camel hair Camelus bactrianus 2
Cashmere Capra hircus laniger 15 20
Mohair Capra hircus 5
Silk Bombyx mori 150
Wool Ovis aries 2,100

The material properties of some animal fibre are listed in Table 9. The moisture
regain of silk and wool are 11% and 13.6%, respectively. Feather fibre is

17
relatively finer than camel hair and wool.

Table 9: Animal fibre physical properties [4]

Name Diameter (m) Length Density (g cm-1)
Camel hair 20 105 8.89 cm 1.32
Silk 10 13 1 1.5 km 0.93
Wool 25.5 28.5 8.255 cm 1.31
Feather barb 15 110 3 13 mm 0.89

The mechanical properties of a few animal fibres are listed in Table 10. Feather
barb fibre has stiffness comparable to upper limits of camel and wool and slightly
lower strength than other animal fibres.

Table 10: Animal fibre mechanical properties

Name Modulus (GPa) Strength (MPa) Elongation (%)
Camel hair 2.02 3.87 121 212 37 50
Silk 0.015 0.017 610 690 4 16
Wool 2.3 3.4 120 174 25 35
Feather 2.8 113 7.7

These properties indicated that feather fibre is a superior lightweight filler for
textile products such as pillows, duvets and upholstery.

2.4 Feather Fibre Applications

Feather fibre finds wide spread application. Traditionally, feathers have been used
as thermal insulation in products such as jackets or as volumetric filler in pillows.
More recently, it has been recognised that more valuable applications are possible,
such as biodegradable plastics and filter media.

From the late 1970s patent literature would suggest that poultry feathers were
dry-cleaned and used for quilts, pillow filling and jacket insulations [36, 37].
Fibre production from chicken feather has not been commercialised in New
Zealand, and current practice is to hydrolyse feather into feather meal for animal
feed and dispose surplus feather by composting. Some recent applications of
feathers are further discussed below.

2.4.1 Polymer and Composites

Insulation made from ground chicken quill and polypropylene (PP) showed good
sound attenuation properties. The noise reduction coefficient (NRC) from ground

18
quill composites was 71% greater than jute composite. Factors such as inner and
outer void content of the quill contributed to better sound absorption, making
quills a promising noise insulation panel and head-liner substrate [38].

Building material containing chicken feather fibre, recycled kraft paper and
newspaper was made into non-woven mats by a wet-lay process. The fibre mats
were infiltrated with acrylated epoxidised soybean oil (AESO) resin using vacuum
assisted resin transfer moulding (VARTM). The product is cheap, environmentally
friendly, and energy efficient. Product applications include temporary building
structure and light vehicle bodies [39].

Recent work by the US Agricultural Research Services produced dashboards,

soundproof materials, plant pots and 5 m mesh depth filters. The patented and
licensed technology has a pilot plant production capacity of 90 kg fibre per hour.
The planned large-scale facility is set to produce 5 tonnes per hour and will be
located in Southwest Missouri or East Maryland [7].

Printed circuit boards for the electronics industry can be made from keratin
feather fibres infused with acrylated epoxidised soybean oil (AESO). In the study
it was shown that keratin fibres remained hollow resulting in a dielectric constant
(k) from 1.7 2.7. This value is lower than silicon dioxide, epoxy and polymer
dielectric insulators [40]. The chicken feather fibre-epoxy composite was
compared with E-glass-epoxy composites. Boards containing chicken feather fibre
had a 2 to 4 magnitude greater electrical resistance than their E-glass counterpart.
The dielectric constant values were comparable to commercial printed circuit
boards [41].

It has been shown that potassium hydroxide hydrolyses feather, making it suitable
for applications such as biopolymer films, foils and drug encapsulation materials
[24]. Furthermore, bacteria such as Streptomyces and Bacillus sp. can hydrolyse
keratin to enhance drug delivery as biocompatible materials typically inserted into
animal tissue. It also has a novel potential to hydrolyse prions, misfolded proteins
that cause bovine spongiform encephalopathy and CreutzfeldtJakob disease in
humans [42].

19
2.4.2 Adsorbent in Bioremediation
Chicken feather has exceptional adsorption properties. Chicken feather has been
utilised as low-cost adsorbents on a fixed-bed system for phenol removal from
aqueous solutions [43]. Filter media made from chicken feather accumulated
gold-cyanide ion from aqueous solutions with a high yield [44]. Chicken feather
also complexes with palladium (Pd), a precious metal [45]. Regenerated chicken
feather adsorbed hexavalent chromium (Cr6+) ions in wastewater with high
efficiency [46]. Arsenite (AsO43-) can be removed from aqueous solution by direct
adsorption onto chicken feather, while traditional bioremediation required an
oxidation step to form arsenic first. The sulfhydryl groups on cystein/cysteine
molecules forms a complex to take up As3+ [47, 48].

Keratin pyrolysis produced porous carbon-nitrogen fibres for nanotechnology. The

heat-resistant adsorbent product had accessible pores with a diameter of less than
1 nm and can also be used for hydrogen storage or small gas molecule separation.
The micro-pores sizes were more uniform than commercial activated carbon [49,
50].

2.4.3 Enzymatic Applications

Chicken feather substrate can be used to produce enzymes other than keratinase.
Azocasein, a dye used to induce inflammation in animal testing was degraded by
Bacillus licheniformis ER-15 cloned into E. coli HB101 via vector pEZZ18 ker
BL2 [51]. B. halodurans JB 99[52], B. pumilus A1 [53, 54] and B. licheniformis
RPk [55] secreted caseinase, used in the human body to digest milk. B.
licheniformis NH1 secreted -amylase to convert starch to sugar [56].

Feather used as a substrate in microbial cultures that secrete extracellular enzymes

for detergent formulations. Serine protease from Bacillus mojavensis A21 is
compatible with a variety of alkaline surfactants at 60oC. The enzyme can be used
with non-ionic and anionic surfactants as both solid and liquid detergent. The
formulations remain active and stable at warm temperatures [57].

2.4.4 Crop Fertiliser and Plant Antimicrobial Control

Bacteria feeding on as little as 0.1% w/v feather is able to promote plant growth
such as siderophore (iron transport), indoleacetic acid (phytohormone for cell

20
division), ammonification, hydrolytic enzyme activity and fungal growth
inhibition. Examples include Xanthomonas sp. P5 [58] and Stenotrophomonas
maltophilia from rhizospheric soil of reed growing on chicken feather [59].
Bacillus subtilis from forest soil also exhibited phosphate solubilisation and
broad-spectrum antimicrobial activity [60].

2.4.5 Animal Products

Synthetic tanning methods use toxic chemicals, but the naturally occurring
enzymes from Bacillus halodurans JB 99 can remove hair from animal during
leather manufacture [52]. Another enzyme, secreted by B. halodurans PPKS-2,
was isolated from rice mill effluent and immobilised. Its activity was not inhibited
by SDS, EDTA or H2O2. The immobilised enzyme was used on goat skin and
collagen integrity remained after treatment [61]. Bacillus pumilus CBS degraded
80% of goat hair and 68% of bovine hair after 24 h at 37oC [62]. In a related
process, chicken feather was transamidified, or had its peptides exchanged with
amines. The polypeptides were acylated with maleic anhydride and co-
polymerised with acrylic acid. The product was applied to goat skin wet-blues,
hides that have been processed with chromium tanning agents. The crust, or
leather before colouring, had improved strength and were not damaged by dyes
[63]. Aspergillus oryzae secreted keratinase that can be immobilised on sintered
glass beads. The product acted on breaking down bovine serum albumin, casein,
keratin, chicken feathers, collagen, duck feather and sheep wool [64].

Candida parapsilosis feeding on feather secretes a serine peptidase that

hydrolyses collagen into gelatine [65]. Bacillus megaterium F7-1 feeding on
autoclaved chicken feather produced keratinase that readily degraded feather,
meat, duck feather and human nail, but not human hair and sheep wool [66].
These enzymes are being used in wool wash in the laundry detergent industry and
keratin hair treatments.

2.5 Fibre Production from Feather

Recovering fibre from chicken feather that are usable for the aforementioned
applications require some degree of processing. Most importantly, feathers require
decontamination to remove pathogens, cleaning to remove residues from chicken

21
processing and in some cases size reduction; these are discussed in more detail in
the following sections.

2.5.1 Decontamination
Unprocessed feather requires pre-treatment, starting with decontamination to
ensure process hygiene and cleaning to remove impurities that cause objectionable
odour, discolouration and equipment fouling.

Chickens are warm-blooded, leading to a variety of microorganisms on chicken

parts, including mesophilic or psychrotrophic organisms that originate from the
animal, its habitat, processing equipment and human handlers. Contamination
with pathogenic bacteria described in Table A 4 may lead to spoilage and
gastroenteritis; a carcass surface could contain as much as 100 to 1,000
mesophilic bacteria per cm2 [6].

Two common Campylobacter species are C. jejuni and C. coli, where growth
occurs between 25 to 43oC. Gastroenteritis could result from contact with as few
as 500 cells of C. jejuni [67]. They are found on most animals, especially poultry,
but the hosts do not show symptoms [68]. The Enterobacteriaceae family is found
in the intestines of chickens, but handling during slaughter causes cross-
contamination. Common genera include Enterobacter, Escherichia and
Salmonella. The psychrotrophic Salmonella persist even when the meat is chilled
or frozen [69]. Enterobacteriaceae detection is used to indicate food quality and
safety [70].

Chlorine bleach such as 1 5% w/v sodium hypochlorite (NaOCl) at pH 10.0 to

12.0 at 25oC can make unprocessed feather bacteriostatic, where colony
proliferation is halted or delayed [71].

2.5.2 Cleaning
Feather is traditionally washed by organic liquids such as chlorinated
hydrocarbons, and has been described in patented literature. The chemicals are
reused after distillation and solid removal. C2 hydrogen derivative mixtures are
used such as trichloroethylene, perchloroethylene, tetrachloroethylene, and
tetrachloroethene. These chemicals are used for dry-cleaning but discharge has

22
detrimental effects on the environment [36].

The commercialised process filed by the United States Department of Agriculture

in 1998 washed feathers in polar organic solvents such as ethanol. This process
required dual stage leaching, and solvent recovery required additional chemicals
to break the ethanol-water azeotrope [32].

Another patent has recommended using an inorganic solvent such as hydrogen

peroxide (H2O2), chlorine bleach (such as sodium hypochlorite) or detergents such
as sodium dodecyl sulfate (SDS) [72]. Hydrogen peroxide is routinely used in
waste water treatment, where proper use allows it to break down into innocuous
water and oxygen [12].

Feather cleaning methods are often adapted from unit operations in the pulp and
paper, waste treatment or food industry. Pulp washing separates dissolved
organics and spent chemicals from wood fibres. Dissolved contaminants are
removed from the suspension for further treatment. Well-designed pulp washing
minimise effluent discharge, chemical use and spillage. The cellulose fibre
suspension is mixed, filtered and fed to another washing steps in series [73]. Pulp
is also bleached to alter the fibres optical properties by removing
chromatophores, or light absorbing components. Bleaching also adjusts the degree
of cellulose polymerisation, but it is important that when brightness is increased
that material strength is preserved [74].

In the food industry, lipids are routinely separated from plant material. Products
include vegetable oils, sugars, carbohydrates, aromatics, antioxidants and
cholesterol removal, which is similar to lipid removal from chicken feather. Solid-
liquid extraction (leaching) is commonly used, where a solute transports fluid
from a solid matrix to a solvent. It is a multi-component, multiphase, un-steady
state mass transfer operation [75]. The most important steps in leaching include:

entrance of the solvent into the solid matrix,

solubilisation and/or breakdown of components,
transport of the solute to the exterior of the solid matrix,
migration of the solute to the exterior of the solid matrix,

23
 movement of the extract with respect to the solid (i.e. extract
displacement), and
separation and discharge of the extract and solid.

Leaching is classified into percolation or immersion methods; batch or continuous

modes; and whether the process is single stage or uses multiple counter-current
stages [76]. A suitable solvent for feather leaching must meet the following
criteria [77]:

compounds to be removed must be soluble in the solvent,

the product must not contain solvent residues,
the chemical should be recovered by distillation or evaporation without
azeotrope formation and has a small latent heat of vapourisation,
interfacial tension, viscosity and wettability should allow solvent to flow
and penetrate pores and capillaries in the feather, and
ideally, the solvent should be non-toxic, stable, non-reactive, non-
flammable, harmless to the environment and cheap.

Chicken feather has an intricate heterogeneous microstructure. The rachis forms

capillaries and hooked barbicels that add to resistance to lipid diffusion. Leaching
is more effective when feather components have full contact with the solvent,
which would otherwise lead to lipid entrainment. The solids should be agitated
and reduced to uniform size by comminution to ensure a high surface area
required for efficient extraction. It has also been shown that temperature control
during leaching maintained feather cell membrane permeability; heating reduced
lipid viscosity and ensured micro-organisms remaining inactive; however, the
temperature should be low enough to prevent keratin degradation [78].

2.5.3 Effluent
The waste purged from the fibre production process should be free from
pathogens. Feather treatment effluent should meet regional by-laws summarised
in Table A 3. Conditional trade waste consent is required for volumes greater than
5 m3 per day, instantaneous flow rates greater than 2 L s-1 or discharge
temperature above 40oC and the waste must not interfere with sewage flow. As an
example, a solvent layer settled or floating in the discharge cannot be allowed
because the sewer could be subjected to corrosion or harmful gas emissions.

24
Furthermore, wastewater treatment can be impaired if the ratio between Chemical
Oxygen Demand (COD) and five-day Biochemical Oxygen Demand (BOD5) fall
beyond 1:1 to 100:1 [79, 80].

2.5.4 Comminution
Size reduction, or comminution, is the process of applying mechanical stress to a
solid to reduce particle size. Comminuted feather has a lower mean particle size
and a narrower particle size range implying a more uniform product.
Comminution is an energy intensive preparation step and up to 99% of the energy
input is converted into heat and absorbed by carrier medium, product and
equipment. Suitable methods include crushing, impact, cutting and using shear
induced from the surrounding medium. Factors affecting comminution equipment
selection include stressing mechanism, feed and product size, material properties,
carrier medium, operation mode and unit operation integration [81]. Some
challenges during comminution are that a heterogeneous feather mixture could
result in reduced average fibre length and changes in fibre length distribution.
Also, lipids remaining after cleaning could cause product discolouration.

Feather comminution could broadly be divided into dry and wet processes. Dry
comminution using a comb/brush separator has been described in a patent. Fibres
were combed by rotating brushes and air pulled through a screen. The screen
aperture size prevented rachis from passing through [32]. One complication was
that dry barbs had a static charge and stuck to dry equipment surfaces.

Wet comminution would overcome these problems. A patent recommended wet

comminution by using refiners, pulpers or disc mills. A high shear pulper used a
coupled rotor and stator with a close tolerance to remove fibre from rachis. The
refiners, or disc mill, grinded, sheared, shredded, pulverised, rubbed and fluffed
the feather into a suitable product [72].

2.5.5 Filtration
After comminution fibre can be recovered using three steps:

feather fraction less dense than water float to the top and is skimmed off,
feather in suspension is separated by gravity filter, and

25
 wet feather is dried.

Fibre separation classifies particles by size in a continuous process and separation

is affected by particle density and shape. These properties need to be constant for
effective separation and reproducible results. Other factors that need to be
considered during filtration are [82]:

particle shapes and size,

volume and mass fraction,
filter performance,
whether the process is by cake and/or depth filtration,
number of stages for filtration in series and number of filter cycles for
circuits,
the type of separation equipment such as vacuum filter, batch pressure
filter, continuous pressure filter and centrifugal filters, and
filter media selection.

A Beloit pressure screen in Figure 8, removes debris from slurry. Several units are
set up in series or parallel to form a multi-stage cascade. The pressure screen
shown in consists of a rotor, screen cylinder, pipes and a chamber. The rotor
increases consistency, or solids concentration. The screen cylinder, or basket, can
be slotted, perforated or contain wedge wires. Perforations and wedge wires allow
100% screen plate effective area. Slots usually have 60 65% effective area or
can be up to 75% or 80%. An optimised pressure screen process improves
screening efficiency, increases capacity and minimises equipment downtime due
to blockage [83].

26
Figure 8: Beloit pressure screen. A - The open pressure screen chamber. B - Control panel. C -
Front view of pressure screen chamber with basket. D Side view. E Inlet. F, G and H Outlet.

2.5.6 Drying
Pulp fibre is dried by heated air circulation and wet fibres are often fluffed before
drying to disperse particles. The dispersed fibre has greater specific surface and

27
contact with drying medium, which facilitates efficient evaporation. Fibre
dispersion also allows transport through a continuous drying system.

Wet fibre surfaces are covered with moisture at the start. The drying rate during
this period is high and the fibres warm to wet-bulb temperature of the air. Internal
moisture evaporate when the surface dry out. Resistance to drying increases
during this phase and surface temperature rises. The drying rate depends on how
liquid migrates to the surface for the specific material. The exhaust air is reheated
and recycled to reduce fresh air intake. Dry fibre is cooled to prevent brightness
reversion [84].

28
3 Experimental

29
3 Experimental
3.1 Materials
Chicken feather used in this project was obtained from Wallace Corporations
Waitoa rendering plant. The rendering department operates 5 days a week for 16
hours a day, processing about 200 tonnes of wet feather per week, most of which
come from a close by Ingham processing plant (Figure 9). Feathers are rendered
into feather meal, yielding 50 tonnes of feather meal per week. Feather generation
increases by about 3% every year.

Figure 9: Raw feather receiving at Wallace Corporations Waitoa rendering plant.

Chemicals used for decontamination were sodium hypochlorite (NaOCl) and

sodium hydroxide (NaOH). For the purpose of initial scoping trials, commercial
bleach with trade name Janola was used and contains 42 g L-1 NaOCl (equivalent
to 4% w/v available chlorine at the time of manufacture), and 4 g L-1 sodium
hydroxide (NaOH).

Chemicals used for feather treatment were 15% sodium hypochlorite, ethanol
absolute, 30% hydrogen peroxide, Tween-80, and n-hexane supplied by Univar.
Sodium hydroxide pellets, sodium dodecyl sulphate powder and peptone powder
were supplied by Sigma Aldrich.

3.2 Equipment
Equipment used in this study is presented in three parts, equipment used for lab
scale experiments, large scale equipment and analytical equipment.

30
3.2.1 Lab-scale
A Boltac 6-unit mixer was used for initial scoping trials for feather
decontamination and cleaning (Figure 10). Each unit had a working volume of 1 L
and stirring speed up to 100 rpm without temperature control.

Figure 10: Boltac 6-unit mixer (left).

A Breville BFP300 and a Kenwood FP920 blender processor (Figure 11) was used
for lab scale comminution.

A B

31
 E F

Figure 11: Lab-scale comminution equipment. A Breville processor blade. B Breville 3 L

processor and 2 L blender. C Kenwood 2 L blender jug and blade bowl. D Kenwood 3 L
blender bowl on rotor unit. E Kenwood dry mill. F Kenwood processor bowl.

3.2.2 Large Scale

Feather was decontaminated and cleaned using a 30 L Lamort Pulper (Figure
12A). The equipment specification is summarized in Table A 11. The number and
spacing of rotor teeth can be varied for fine or coarse fibre removal and angled
baffles deflect solids coming into contact with the side of the tank [1]. In this
study, a flat disc agitator was used to prevent feather entanglement around the
agitator (Figure 12B).

32
Figure 12: Lamort pulper. A Motor and tank and B Disc impeller.

Feather comminution was performed using a 600 L stainless steel tank combined
with a centrifugal pump, operated under total recycle (Figure 13). The tank had an
inner diameter of 0.7 m and an inner height of 1.8 m. The outlet of the tank was
directly mounted to the suction end of the pump with a diameter of 100 mm. The
discharge side was connected to a 101.6 mm diameter, 8 m long PVC pipe and
was returned directly back to the tank.

33
 C D

Figure 13: Comminution system. A 600 L stainless steel tank. B Tank filled with 300 L water.
C Scanpump connected to side of tank. C Side access panel and discharge valve.

A centrifugal pump (Scanpump 310-0133) was to circulate the feather suspension,

capable of delivering 250 m3 h-1 (Figure 14). Technical data are shown in Table A
12. The pump was controlled by variable speed drive (Altivar-61). The
volumetric flow rate was monitored using a Promag 35 electromagnetic flow
meter from Endress+Hauser.

Figure 14: Scanpump 310-0133.

Chicken feather suspensions were filtered using filters shown in Figure 15 to

separate fibre from partially cut and uncut feathers. Only the fine hand-held filter
is shown in Figure 15 as both had a similar appearance.

34
Figure 15: Tall, wide and fine filters used during fibre recovery.

Details of filters used are shown in Table 11.

Table 11: Filter dimensions

Filter Type Tall Wide Coarse Fine
Diameter (m) 0.3 1.0 0.20 0.15
Container Height (m) 1.0 0.5 0.30 0.05
Aperture location Side Base Base Base
Aperture diameter (m) 0.0254 0.001 0.05 0.001

35
3.3 Methods
3.3.1 Lab-scale Decontamination and Cleaning
10 g unprocessed chicken feather was agitated using the Boltac mixer at 60 rpm in
500 mL diluted Janola, ethanol, hydrogen peroxide (H2O2), or sodium dodecyl
sulphate (SDS) for 1 hour. Liquid were filtered away with 1 mm mesh hand held
filter. The wet feathers were dried to constant mass in Contherm air-forced dryer
at 70oC.

3.3.2 Lab-scale Comminution

10 g chicken feather suspended in 500 mL water was cut with a Kenwood FP920
(Figure 11) horizontal double knife unit in the processor bowl, alternating angled
4-blade unit in the blender jug, and a dry mill with blades similar to the blender.

3.3.3 Large Scale Decontamination

The Lamort Pulper was filled with 25 L water and 2.5 kg unprocessed feather. 1M
NaOH was added to keep suspension at pH 10.0 before 250 mL 15% NaOCl was
added. The suspension was agitated at 10 Hz for 30 minutes. The liquid phase was
drained, replaced with 250 mL 15% NaOCl adjusted to pH 10.0, and mixed for
another 30 minutes. The liquid phase was drained and the decontaminated feather
was rinsed with about 25 L water to remove remaining NaOCl. The process is
summarised in Figure 16.

3.3.4 Large Scale Cleaning

The cleaning step was a three stage process. The decontaminated feather in the
Lamort Pulper was suspended with 25 L water and 125 mL 30% H2O2 was added.
The suspension was agitated at 10 Hz for 30 minutes. The liquid phase was
drained and replaced with 25 L water and 125 mL 30% H2O2 before agitation at
10 Hz for another 30 minutes for stages two and three. The clean feather was
rinsed with 25 L water to remove remaining H2O2. The process is summarised in
Figure 17.

36
 Feather after stage 2
Pulper decontamination
25 L Water
pH 10.0
Raw feather
1 M NaOH
250 mL NaOCl

1 mm mesh
Filter

Waste Water
Feather after stage 1
decontamination
Cleaning

Figure 16: Decontamination block flow diagram.

Feather after stage 3

cleaning
Pulper
25 L Water

Decontaminated feather
125 mL H2O2

1 mm mesh
Filter

Waste Water
Feather after stage 1 and 2
cleaning
Comminution

Figure 17: Cleaning block flow diagram.

3.3.5 Large Scale Comminution

The 600 L tank was filled with 300 L water and 1.33 kg clean feather. The
centrifugal pump was set to 30 Hz impeller speed. The feather-water slurry
circulated through the system for four hours. 4 L samples were taken every hour,
weighed and filtered into feather fraction (> 5 mm mesh) and fibre fraction (> 1

37
mm mesh) and liquid phase (< 1 mm mesh). The feather and fibre were dried to
constant mass in Contherm air-forced dryer at 70oC. The slurry temperature was
recorded at every sample interval. The process is summarised in Figure 18.

1.5 kg
Clean Feather

Tank 300 L
Centrifugal
Water
Pump
30 Hz

Cut Feather Recovery

Figure 18: Comminution block flow diagram.

3.3.6 Large Scale Fibre Recovery

Fibre was recovered from 5 mm and 1 mm filters set up like Figure 19. The slurry
was allowed to float to the surface while bulk slurry cooled to a temperature safe
to handle. The floating fraction was skimmed off with a net. The suspension was
discharged over the 1 mm mesh wide filter from a horizontal valve at the side of
the tank near the base. Sediments were discharged through a vertical valve at the
tank base and filtered by a 1 mm mesh filter. Floating feather and suspension were
divided into manageable portions and washed over a 5 mm filter placed above a 1
mm filter with high pressure water.

The feather fraction remained on the 5 mm filter and the fibre fraction remained
on the 1 mm filter. Feathers and fibres were laid flat on plastic trays lined with
aluminium foil and dried to constant mass at 70oC in a Contherm air-forced dryer.
Dry feather and fibre were weighed and sealed in plastic bags.

38
 Floating feather

Tank 300 L Feather Feather

Water suspension 1 suspension 2

1 mm mesh
filter

Feather Waste Water

Sediment
High pressure water

Feather to
dryer 70oC
Floating Feather and
5 mm mesh
Feather suspension 2 filter

Product to
dryer 70oC

1 mm mesh
filter

Waste Water

Figure 19: Fibre recovery block flow diagram.

39
3.3.7 Analysis
Feather samples were sent to NZ Lab to test for Campylobacter,
Enterobacteriaceae and Salmonellae. Each specimen was weighed, rinsed and
removed from 400 mL of 0.1% peptone water according to APHA (4) 2.527. The
bacteriological test procedures are summarised in Table A 5.

Feather morphology was observed using a Wild-Leica M2B stereo microscope

with 6,4x, 16x and 40x click magnification mounted on adjustable stand and
illuminated by halogen light [85]. Fibres were photographed using an Olympus
PM-10M digital camera. Technical specifications for the microscope are listed in
Table A 6.

A Hitachi S4100 Field Emission Scanning Electron Microscope (FE-SEM),

similar to Figure 20, was used to examine feather surface properties. The
specimens were coated with platinum and examined at 20 kV accelerating
voltage. Micrograph magnifications used were 1,000x and 3,500x [86].

Figure 20: Hitachi S4000 series FE-SEM [3].

Soxhlet extraction was used to extract soluble impurities from chicken feather
(Figure 21). The main apparatus consisted of the Soxhlet extractor, a glass
chamber with siphon, a water cooled condenser and a 125 mL reservoir flask The
Soxhlet extractor was connected to the reservoir flask with a condenser and

40
clamped above a heating element (Thermo Scientific).

Figure 21: Soxhlet extractor apparatus [4].

Solid specimens were dried to constant mass in a Contherm air-forced oven after
extraction and solvents were recovered using vacuum evaporation in a Buchi
Rotavapor consisting of a rotating angled flask on a water bath, coil condenser,
pressure release tap and a clamped round flask. The tests were carried out in
triplicate.

Material colour was determined by Konica Minolta CR-410 Chroma Meter

(Figure 22) with technical specifications in Table A 7. The pulsed xenon arc lamp
(a) light source with baffle (b) and diffuser (c) allow light to travel only in the
vertical direction to illuminate sample (f). The optic fibre cables (d) and (e)
monitors colour and illuminant, respectively (Figure 22) [87]. The meter was
calibrated with a white plate. Absolute colour readings were recorded in L*a*b*
space and converted to RGB using method described in Table A 8.

41
Figure 22: Konica Minolta CR410 Chroma meter sensor [6].

3.4 Experimental Plan

3.4.1 Lab-Scale Decontamination
Information on the Janola container suggested household cleaning in 1:4, 1:8 or
1:20 dilutions for 5 to 10 minutes. 10 g unprocessed feather was treated in 500 mL
diluted Janola in the concentrations listed in Table 12 for 5 or 10 min.

Table 12: Janola concentrations and residence time

Sample Diluted Ratio Concentration (% w/v NaOCl) Time (min)

J1 1 in 20 0.21 5
J2 1 in 8 0.525 5
J3 1 in 4 1.25 5
J4 1 in 20 0.21 10
J5 1 in 8 0.525 10
J6 1 in 4 1.25 10

3.4.2 Lab-Scale Cleaning

10 g unprocessed chicken feather were mixed with 500 mL cleaning agents for 1
h. Initial trials with ethanol dilutions listed into Table 13 were carried out to gauge
the mixing effect of Breville blender-processor.

Table 13: Low concentration ethanol cleaning

Concentration (% v/v) Volume* (mL) Mass (g) Concentration (wt%)
0 0 0 0
1 5 3.95 0.79
3 15 11.85 2.39
5 25 19.75 3.99
7 35 27.65 5.61
10 50 39.50 8.07
*ethanol pipetted into volumetric flask and filled up to 500 mL with water

42
The surfactant Tween-80 (Table 14), recommended by patent [32], was also used
in the blender to gauge mixing effects.

Table 14: Tween-80 cleaning

Tween-80 Volume (mL) Concentration (% v/v)
2.5 0.005
3.5 0.007

30% hydrogen peroxide was diluted to concentrations in Table 15 recommended

by patent [72].

Table 15: Hydrogen peroxide cleaning

Volume (mL) mass (g) Concentration (wt%)
2.75 3.0525 0.6101
2.25 2.4975 0.4993
1.75 1.9425 0.3884
1.25 1.3875 0.2774
0.75 0.8325 0.1665

Sodium dodecyl sulfate, recommended by patent [88], was diluted to

concentrations listed in Table 16.

Table 16: Sodium dodecyl sulfate cleaning

Label Mass (g) Concentration (wt%)
DA 2 0.5714
DB 4 1.1429
DC 6 1.7143
DE 8 2.2857
DF 10 2.8571

Ethanol concentration in the range listed in Table 17 was used to confirm that
chemicals chosen in this project are feasible alternatives to state of the art
technology.

Table 17: Higher concentration ethanol cleaning

Label Volume (mL) Concentration (% v/v)
ED 50 14.29
EE 100 28.57
EF 150 42.86
EG 200 57.14
EH 250 71.43
EI 300 85.71

43
3.4.3 Lab-Scale Comminution
The comminution configurations were Kenwood FP920 horizontal double knife
unit in the processor bowl, alternating angled 4-blade unit in the blender jug, and a
dry mill with blades similar to the blender. All three configurations were tested for
1 h residence time at three speed settings, except the test using dry mill, which
was for 10 seconds.

3.4.4 Large Scale Decontamination and Cleaning

2.5 kg batches of feathers were decontaminated using 2 stages of 0.1485% NaOCl
adjusted to pH 10.0 with 1 M NaOH in 25 L water mixed at 10 Hz by a Lamort
pulper for 30 minutes. Decontaminated feather was cleaned for 3 stages using
0.15% H2O2 in 25 L water mixed at the same conditions as decontamination. 10 g
samples were taken after each step for tests listed in Table 18.

Table 18: Large scale treatment analysis

Analysis
Time
Treatment Optical
(h) Bacteria Colour SEM
Microscope
0 Untreated X X X X
0.5 1 stage NaOCl X X
1 2 stage NaOCl X X X X
1.5 1 stage H2O2 X X
2 2 stage H2O2 X X
2.5 3 stage H2O2 X X X X

3.4.5 Large Scale Comminution

Two experiments were performed: A) 1.33 kg dry feather in 300 L water
processed for 3 hours circulating at 100 m3 h-1. B) 0.947 kg dry feather in 300L
water processed for 4 hours circulating at 100 m3 h-1.

44
4 Results and Discussion

45
4 Results and Discussion
Unprocessed feather appeared straw-like and the barbs were stuck to the rachis in
a greasy tangle (Figure 23). Raw wet feather turned brown after 2 days if left at
room temperature and had a distinct putrid smell. After washing in regular tap
water feathers had their barbs fanned out, but a significant odour remained. Water
holding the feather turned black after a few days from bacterial decomposition.

A B

Figure 23: Raw feather. A Rachis with lipids. B Tip with brown impurities.

4.1 Lab-scale Decontamination

The main contaminants on the surface of feathers are preen oil, offal fat and faecal
matter. Initially, full strength Janola (4% w/v NaOCl) was used to decontaminate
unprocessed feather, however, the feathers completely degraded in less than a
minute. Feathers treated with diluted Janola containing 1.05% NaOCl did not turn
brown or retained their unpleasant odour.

Sodium hypochlorite (NaOCl), the sodium salt of hypochlorous acid, caused lipid
bilayer of bacterial cell membranes to lyse. Hypochlorous acid hydrolyses
unsaturated fatty acids by adding chlorine and a hydroxyl group to carbonyls. At
high concentrations, hypochlorous acid oxidises sulfhydryl groups on cysteine to
denature keratin and form hydrochloric acid (HCl). NaOCl reacts with HCl to
form chlorine gas. This was prevented by maintaining a pH 10.0 suspension by
adding sodium hydroxide (NaOH). The effluent must be thoroughly rinsed away
because NaOH degrades fibre and NaOCl releases oxygen violently on contact
with hydrogen peroxide.

Soxhlet extraction was used to determine residual fat content, after treating feather
using different concentrations of NaOCl (Table A 10). Feather treated with 0.21%

46
w/v Janola unexpectedly caused final specimen weight to increase; while a small
degree of impurities was removed using a concentration ranging from 0.525 to
1.25% w/v. A sample of feather treated using 0.525% Janola is shown in Figure
24. Consequently, 15% sodium hypochlorite diluted to 1% w/v was used to
decontaminate feather in large scale processing.

Figure 24: Feather treated by 0.525% Janola for 5 minutes.

4.2 Lab-scale Feather Processing

Candidate washing solutions representative of solvents suggested by patents were
trialled; these were ethanol, Tween-80, sodium dodecyl sulfate (SDS) and
hydrogen peroxide. In these experiments, cleaning and feather cutting were
trialled as a combined process using the lab scale blenders described in the
previous chapter.

Initially, a Breville BFP30 processor, shown in Figure 11B, was used to mix the
feather suspensions, but the rachis tangled around the rotor and blade, causing
undue resistance to the motor. In an effort to resolve this, raw feathers were pre-
cut to 15 25 mm using scissors. The processor could only be used for 5 min at a
time to prevent the motor overheating and intermittent mixing meant the feather
settled or floated, resulting in inadequate contact with the cleaning solution. Since
feather were pre-cut with scissors before using the blender-processors, it was
impossible to determine whether feather was further reduced in size during
processing.

At low agitation speed and a feather to liquid ratio of 1:50, feather floated and
swirled around with little contact with the blades. Adjusting to 1:1 feather to
liquid ratio caused long flight feathers to wrap around the blade causing
equipment failure.

47
Using a blender with different height to diameter ratio and spike blades at the base
(Figure 11C) was capable of shredding feather at 1:1 feather to liquid ratio. At low
solid concentrations, the suspension spun from the middle to form a vortex and
travelled up the side of the container. At high solid concentrations, long rachis
wrapped around the blades after a few minutes. A slurry formed, where down
feather was reduce in size, but long flight feathers interfered with the rotating
blade.

The wider bowl (Figure 11F in page 32) could accommodate the occasional long
rachis as the blade was located at the base of the rotor causing less vortex
formation. However, agitation caused feather to align parallel to the flat blades so
cutting was limited to vertically oriented feathers. The Kenwood FP920 (Figure
11D) as capable of superior mixing, its blades were inefficient in reducing feather
into fibre.

The dry mill in Figure 11E had the same shaped blades as the blender. It readily
pulverised dry feather but produced a significant amount of powder and only 1 g
of completely dry feather could be processed at one time.

Feather cleaned by up to 10% v/v ethanol was generally found to be ineffective

and resulted in fluffy light brown feather with an unpleasant odour. Tween-80
caused excessive foaming, and can only be used with gentle mixing. Hydrogen
peroxide imparted a sickly sweet bleach odour to the feather. SDS imparted a soap
smell to the feather and ethanol imparted a perfume-like odour to washed feather.
It was found that feather had to be properly rinsed to remove residual cleaning
chemicals and stop further chemical degradation that lead to foreign odour
development in the product.

4.3 Lab-scale Analysis

4.3.1 Impurity Evaluation
The six-unit Boltac mixer was used to agitate 10 g feather suspended in 500 mL
cleaning solutions at 100 rpm for 1 h. The impurities expressed as n-hexane
extractable content were plotted against concentration in Figure 25 to Figure 27.

It can be seen from Figure 25 that ethanol cleaning reduced the hexane soluble
content from about 15% in raw feathers, to about 10% using 80 % ethanol. This

48
would suggest that a single equilibrium stage would be insufficient in reducing
hexane extractable impurities to an acceptable level. Assuming that one third
impurities are removed in single stage, it would require at least 3 cleaning stages
to produce fibre containing approximately 1% impurities. It was also found that
using more than 80 % ethanol resulted in no additional benefit.

20
18
16
14
Hexane 12
Extractable
10
Content
(%) 8
6
4
2
0
0 20 40 60 80 100
Ethanol concentration (wt%)

Figure 25: Hexane extractable content after ethanol cleaning.

Using hydrogen peroxide at a concentration ranging from 0.15 0.55% w/v gave
similar results to that of ethanol and cleaning efficiency levelled off after 0.25 %
hydrogen peroxide. Impurity content (Figure 26) also remained above 10% and
would therefore require more than one equilibrium stage

49
 20
18
16
14
Hexane 12
Extractable
10
Content
(%) 8
6
4
2
0
0 0.1 0.2 0.3 0.4 0.5
Hydrogen peroxide concentration (wt%)

Figure 26: Hexane extractable content after hydrogen peroxide cleaning.

From Figure 27 it can be seen that using SDS at concentrations from 4 x 10-4 to 2
x 10-3 % was ineffective at reducing hexane extractable impurities. It was thought
that SDS encapsulated lipids and keratin as micelles but did not separate the two.
Another factor may also be that it was more difficult for this amount of SDS to
interact with oil adsorbed into feather barbs or rachis and would therefore require
a solvent that can easily penetrate the rachis or barbs.

22
20
18
16
14
Hexane
Extractable 12
Content 10
(%)
8
6
4
2
0
0.00E+00 5.00E-04 1.00E-03 1.50E-03 2.00E-03
Sodium dodecyl sulfate concentration (wt%)

Figure 27: Hexane extractable content after SDS cleaning.

From these results, it was concluded that equilibrium quickly sets in between the
hexane extractables in the feather and the amount that can be dissolved in

50
solution. It would therefore be required to use more than one equilibrium stage to
obtain a feather product with very low extractable content.

It was found that hydrogen peroxide would be a better choice of cleaning agent,
based on price and safety concern dealing with a solvent. Ethanol has to be
recovered using distillation whereas hydrogen peroxide safely decomposes and
can be discarded. In large scale experiments, three equilibrium stages using
hydrogen peroxide was implemented. Considering the relative insensitivity of
impurity removal to hydrogen peroxide removal, it was decided to use 0.15% in
large scale trials.

4.3.2 Colourimetry
Feather whiteness after treatment with various concentrations NaOCl is presented
in Figure 28. Feathers were whitened to at least 80% after treatment. For
concentrations 0.525 and 1.05% equivalent NaOCl, feathers treated for 5 minutes
were somewhat whiter than those treated for 10 minutes. Table 19 illustrates the
colour of treated feathers using sodium hypochlorite.

100

80

60
Whiteness
(%)
40

20

0
X A B C D E F

Figure 28: Feather whiteness after Janola treatment. X Untreated. A 0.21% NaOCl for 10
minutes. B 0.21% NaOCl for 5 minutes. C 0.525% NaOCl for 10 minutes. D 0.525% NaOCl
for 5 minutes. E 1.05% NaOCl for 10 minutes. F 1.05% NaOCl for 5 minutes.

Table 19: Feather colour after Janola decontamination.

NaOCl Time Average Colour Whiteness

51
 (%) (min) L* a* b* R G B (%)
0 0 78.5 1.4 11.0 207 203 172 76.13

0.21 10 89.5 0.4 8.8 235 236 207 88.57

0.21 5 88.0 0.6 8.8 231 231 202 86.85

0.525 10 85.1 0.7 9.6 224 222 193 83.56

0.525 5 90.9 -0.1 7.1 237 240 214 90.28

1.05 10 85.4 1.7 11.3 228 223 191 83.84

1.05 5 87.8 1.0 11.1 233 230 197 86.34

Feather whiteness after ethanol treatment is presented in Figure 29. Whiteness

values were also above 80%. Ethanol concentrations above 14 % did not produce
increasingly whiter feathers. Table 20 shows 14% ethanol was most effective at
removing colour.

100

80

60
Whiteness
(%)
40

20

0
X A B C D E F

Figure 29: Feather whiteness after ethanol treatment. X Untreated. A 1%. B 3%. C 14%. D
42%. E 57%. F 85%.

Table 20: Feather colour after ethanol cleaning.

Ethanol Average Colour Whiteness

52
 Concentration (%)
L* a* b* R G B
(%)
0 78.5 1.4 11.0 207 203 172 76.13

1 83.6 1.0 10.6 221 218 187 81.78

3 86.6 0.5 7.8 227 227 201 85.52

14 90.9 -0.2 6.6 236 240 215 90.37

42 85.8 -0.3 5.8 221 225 202 84.70

57 88.4 -0.1 6.9 230 233 207 87.48

85 83.8 -0.2 6.8 216 219 195 82.43

Feather whiteness values for two mixing methods were compared in the first part
of Figure 30. Feathers were suspended in 500 mL water and mixed using the
Kenwood processor and the Boltac mixer at 60 rpm. There was no significant
difference in whiteness. Feather treatment with ethanol, hydrogen peroxide and
sodium dodecyl sulfate are also presented in Figure 30. SDS produced the whitest
feathers. Feather cleaned by 500 mL 0.25% H2O2 agitated at 60 rpm for 1 hour in
the Boltac was also able to whiten feather to about 90% whiteness. Table 21
shows feather colour removal by mixing equipment and treatment with different
chemicals.

53
 100

80

60
Whiteness
(%)
40

20

0
X A B C D E F G H

Figure 30: Feather whiteness comparison after mixing and treatment. X Untreated A
Suspended in water and mixed by processor B Suspended in water and mixed by Boltac at 60
rpm. C 1% ethanol. D 3% ethanol. E 0.15% H2O2. F 0.25% H2O2. G 2.28 wt% SDS. H
2.86 wt%.

Table 21: Feather colour after mixing and treatment.

Average Colour Whiteness
Treatment
L* a* b* R G B (%)

No Treatment 78.5 1.4 11.0 207 203 172 76.13

Kenwood Processor 89.6 -0.3 7.3 233 236 210 88.76

Boltac at 60 rpm 88.2 -0.4 5.8 228 232 209 87.44

1% ethanol 83.6 1.0 10.6 221 218 187 81.78

3% ethanol 86.6 0.5 7.8 227 227 201 85.52

0.15% H2O2 80.4 0.2 9.5 209 209 180 78.30

0.25% H2O2 89.4 -0.3 6.3 232 236 211 88.70

SDS 2.28 wt% 90.6 0.1 7.3 237 239 213 90.02

SDS 2.86 wt% 91.0 -0.2 5.9 236 241 216 90.58

54
4.3.3 Morphology
Feather decontaminated using 0.21% Janola for 5 minutes and 10 minutes were examined and
the results are present in Figure 31. Feather processed for 10 minutes were more yellow than
those processed for 5 minutes. Mixing in diluted Janola caused feathers to unfurl improving
impurity removal during further treatment.

5 minutes 10 minutes

6.4X

16X

40X

Figure 31: Feather morphology after 5 min or 10 min Janola treatment.

Feather cleaned by 1% and 3% ethanol were compared under optical microscopy (Figure 32).
Intricate substructures were clearly visible after the 3% ethanol treatment. The barbs were
parallel and barbules were aligned. The barbules were still sticky, presumably as a result of
residual lipids after 1% ethanol.

55
 1% Ethanol 3% Ethanol

6.4X

16X

40X

Figure 32: Feather morphology after ethanol lab-scale treatment.

Feather cleaned by 0.15 and 0.25% hydrogen peroxide in the Breville mixer were compared
under an optical microscope (Figure 33). In these images, feather fibres are clearly visible.
These fine fibres have unfolded away from clumps. Some layers of lipids have been removed
to reveal the white surface of feather keratin.

56
 0.15% H2O2 0.25% H2O2

6.4X

16X

40X

Figure 33: Feather morphology after hydrogen peroxide lab-scale treatment.

From the results presented here it can be concluded that feather treatment can improve colour
to 90% whiteness. Colour improvement using ethanol was best at 14.29%, but using
concentrated ethanol only 80% whiteness could be reached. SDS produced the whitest
feathers, and 0.25% hydrogen peroxide was able to whiten feather to a comparable degree. It
was also confirmed that mixing using the Boltac at 60 rpm was just as effective as mixing
with Kenwood processor in terms of removing colour.

4.4 Large Scale Treatment

Initial large scale decontamination used 1% sodium hypochlorite. Excessive
chlorine gas was generated at this concentration.

57
The chemical equilibrium was upset by acidic pH of the feather. The suspension
must remain alkaline for effective decontamination so sodium hydroxide was
added in subsequent experiment to increase pH to 10.0.

The suspension became saturated with impurities in 30 minutes when the mixing
speed was set at 10 Hz. The liquid fraction was drained and refilled with 25 L
water and 250 mL 15% NaOCl adjusted to pH 10.0, or 0.1485% NaOCl.
Bacteriological test carried out by NZ Labs found that two decontamination stages
at this concentration were adequate. Sodium hypochlorite solution was rinsed
after decontamination to prevent further chemical degradation.

The decontaminated feather was suspended in 25 L water and 125 mL of 30%

hydrogen peroxide for 30 minutes. The murky liquid fraction was drained. Feather
whiteness by colourimetry suggested three stages of 125 mL 13% H2O2 in 25 L
water 0.15% H2O2 sufficiently removed colour from the feather. The wet feather
appeared white to visual inspection. The barbs and barbules were fully spread out
from the rachis. The feather no longer wet rancid during storage.

4.4.1 Bacteriological Test

The raw feather tested positive for C. jejuni. No Salmonella was detected in any
sample. Enterobacteriaceae decreased by a thousand fold after decontamination
and was eliminated after cleaning, as shown in Table 22.

Table 22: Bacterial test results for feather treatment.

Untreated Decontaminated Cleaned
Sample ID
Feather Feather Feather
Salmonellae Not Not Not
in 100 mL Detected Detected Detected
Campylobacter C. jejuni Not Not
in 100 mL Detected Detected Detected
Enterobacteriaceae
20,000 23 <1
cfu per mL

4.4.2 Colourimetry
Feather colours were measured and converted to percentage whiteness shown in
Figure 34. Average whiteness increased after each decontamination and cleaning
stage. Sodium hypochlorite did not remove pigments, while the third stage of
hydrogen peroxide affected noticeable change. Table 23 shows feather colour after
each treatment step.

58
 100

80

60
Whiteness
(%)
40

20

0
X D1 D2 C1 C2 C3

Figure 34: Large scale treatment average feather whiteness. X Untreated. D1 1 stage NaOCl
decontamination. D2 2 stage NaOCl decontamination. C1 1 stage H2O2 cleaning. C2 2 stage
H2O2 cleaning. C3 3 stage H2O2 cleaning.

Table 23: Feather colour after decontamination and cleaning steps.

Average colour from 15 readings Whiteness
Treatment
L* a* b* R G B (%)
No
78.5 1.4 11.0 207 203 172 76.13
Treatment
1 stage
83.7 0.8 12.6 222 218 184 81.57
NaOCl
2 stage
84.3 1.2 12.8 225 219 185 82.19
NaOCl
1 stage
85.9 0.5 12.0 227 225 190 83.98
H 2O 2
2 stage
86.9 0.4 11.2 230 228 195 85.26
H 2O 2
3 stage
87.4 0.7 11.9 232 229 195 85.79
H 2O 2

The large variation was due to the biological nature of the sample. Feather colour
varies due to the individual nature of the animal and how it was processed.
Different aged individuals has varying amount of preen oil in their feather, where
the fatty acid composition contributes to the yellowness. Particulates from dust
bathing, pieces of feed and faecal matter may also be present, resulting in a
heterogeneous mixture. The untreated feather may have been harvested several
hours before decontamination, where bacteria have started to excrete waste
material that cause feather to darken. The untreated feather also contains blood,
which look pink just after collection but turns brown as it dries. The method of

59
transporting the feather, such as by flume water, may affect amount of impurities.
If the fluids were warm, the pigment from blood may set and become hard to
remove.

Table 24 shows feather colour after mixing with Lamort pulper. Both types of
impeller achieved about 88% whiteness after 10 minutes. It was difficult to drain
the suspension while using a spiral impeller so the disc impeller was used for
subsequent large scale treatments.

Table 24: Feather colour after mixing with Lamort pulper impellers.

Time Average Colour Whiteness

Impeller Speed
(min) L* a* b* R G B (%)

None 0 0 78.5 1.4 11.0 207 203 172 76.13

Disc Fast 10 89.4 0.0 7.1 233 236 210 88.64

Disc Slow 10 89.2 0.0 7.0 232 235 209 88.37

Spiral Fast 10 90.6 0.0 5.8 235 239 215 90.15

Spiral Slow 10 88.7 0.0 7.1 231 233 207 87.81

-
Spiral Fast 5 89.3 7.0 232 235 209 88.50
0.1

Spiral Slow 5 84.5 0.3 7.5 220 221 195 83.11

4.4.3 Macroscopic Morphology

Dried feather ranged from pale yellow for untreated feather to off white for
decontaminated and cleaned feather. The bulk of the clean feather became fluffy
and tufts of white down fibre have a soft texture. The flight feather barbs with a
waxy waterproof surface reflected light. The dirty feather was in a clumpy, lank
and stringy form with debris due to remaining lipids. The decontaminated feather
started to unfurl. Each successive stages of cleaning made the feather whiter and
the substructures spread out.

60
 Packaged On contrasting background

Untreated

1 stage
NaOCl

2 stage
NaOCl

1 stage
H 2O 2

61
 2 stage
H 2O 2

3 stage
H 2O 2

Figure 35: Feather photographs after each treatment step.

62
4.4.4 Stereoscopic Morphology
Feather structure and form were examined under an optical microscope. Two
pieces of untreated feathers are shown in Figure 36. The entire feather is tangled
and curled into a clump. Figure 36E shows flecks of yellow debris. Figure 36E
and Figure 36F show bundles of fine fibre.

A B

C D

E F

Figure 36: Morphology of untreated feather.

Figure 37 shows two feathers after the first stage of decontamination. The outer
edges of the barbs in Figure 37B fold inwards. Figure 37D shows a relatively
clean rachis. On the left column Figure 37C and Figure 37E show barbs and
barbules still coated with lipids.

63
 A B

C D

E F

Figure 37: morphology after first stage decontamination.

Figure 38 shows two decontaminated feathers after 2 stages of NaOCl. Both are
still folded due to lipid residues. Figure 38E and Figure 38F show pieces of
yellow debris coating the barbs and the tip of the feathers.

64
 A B

C D

E F

Figure 38: Morphology after second stage decontamination.

Figure 39 shows two feathers cleaned by 1 stage of hydrogen peroxide. Figure

39E and Figure 39F indicate nodular structures are intact.

65
 A B

C D

E F

Figure 39: Morphology after first stage cleaning.

Figure 40A has a lot of fine down fibre, while Figure 40B has ordered barbs and
barbules, but a large piece of debris where the barb is folded over.

66
 A B

C D

E F

Figure 40: Morphology second stage cleaning.

Figure 41 shows two feathers after 3 stages of hydrogen peroxide cleaning. Figure
41A and D has tufts of fine fibres on the outer surfaces. The feather in Figure 41B
has a thin rachis and slightly folded barbs. On closer inspection, the barbs
extending at an angle at even spaces (Figure 41D). The fibres were found to
scatter light at 40X magnification (Figure 41E).

67
 A B

C D

D F

Figure 41: Morphology after third stage cleaning.

4.4.5 Histological Morphology

Figure 42 shows surface morphology for untreated, decontaminated and cleaned
feather. Untreated feather Figure 42D had copious debris on the left side and
closely linked barbules. Figure 42E in the upper middle section showed hooked
barbicels linking to adjacent substructure and a smooth surface in the background.
The feather microstructure was not damaged after decontamination, where
barbicels at 5 m diameter were found intact. Feathers cleaned by 3 stages of
0.15% H2O2 had very few particulate impurities. Figure 42I shows the particles
are located between the layer of barbicels and barbules.

68
 A B

D E

F G

Figure 42: Untreated, decontaminated and clean feather surface morphology.

There were challenges to fully immerse the feather matrix in hydrogen peroxide
solution. The curved hollow cylinder of the rachis acted as a capillary. Its base
was often connected to dermal fragments, which blocked the entrance to the lipid
solute. The barbs, barbules and barbicels were folded together due to the waxy
preen oil coat, so lipids must be removed from the exterior to improve wettability
before the solvent could gain contact with impurities.

Hydrogen peroxide was used for lipid peroxidation, but not to cause keratin
denaturation. Denaturation occurs when disulphide bridges form on cysteine thiol
functional groups and the protein aggregates. It was essential to remove the
extract from wetted feather because it contains fatty acid radicals, lipid peroxyl
radicals, lipid peroxides and cyclic peroxides. The melanin pigments, or

69
chromatophores, were removed. Feather still appeared off-white to pale yellow
due to light diffraction.

4.5 Large Scale Comminution

Results from feather comminution are presented in this section. Two comminution
experiments were performed, using either 0.44% feather consistency or 0.31%.
The pump speed was kept constant at 30 Hz, resulting in a volumetric flow rate of
100 m3 h-1. The first comminution experiment was run for only three hours, and it
was thought that allowing an extra hour could increase the yield; the results are
shown in Table 25.

Table 25: Comminution mass balance on dry basis.

Mass (g)
Time
Stream Experiments 1 Experiment 2
(h)
(0.44% consistency) (0.31% consistency)
Total IN Feather 1325.25 944.47
End Fibre 195.16 248.85
Feather 1004.26 544.45
Samples removed
0 Fibre 0.10 0.07
Feather 10.50 7.09
1 Fibre 2.52 0.26
Feather 13.90 3.98
2 Fibre 0.86 0.25
Feather 11.90 2.17
3 Fibre 2.48 0.27
Feather 9.87 3.93
4 Fibre 0.29
Feather 15.2
Residuals* Feather 6.43 83.76
Total feather remaining
Total fibre and sample 200.96 250.138
Total OUT 1296.19 910.7853
Yield (%) 14.7 26.3
Recovery (%) 97.8 96.4

A 97.8 % and 96.4 % solids recovery was achieved in experiments 1 and 2

respectively suggesting minimal losses during separation and fibre recovery.
About half the mass of feathers is rachis, or there should be a 50 % theoretical
fibre yield. The fibre yield for experiment 1 was 14.7%, or 30% of possible fibre
was recovered. The second comminution used a lower consistency of 0.31 % and
was had an extra hour of operation, leading to a 27 % yield, or 55% of possible
fibre recovered. A lower consistency and longer comminution would therefore be
preferred for increased fibre recovery.

70
Solids were allowed to float to the top of the slurry at the end of comminution.
Barbs detached from the rachis would agglomerate due to hooked barbicels
resulting in highly entangled wet fibres after comminution. This posed significant
problems for fibre recovery using gravity filtration. Pouring water over wet
feather-fibre mixtures did not produce enough agitation or pressure for filtration.
Fibres were mostly curved and angled in a way that exposed barbules and
barbicels as hooks. The 1 mm mesh filter would allow vertically oriented particles
to pass through, but particles that made contact at an angle were trapped. To
overcome this, each batch of feather-fibre mixture was washed several times using
the combination of filtration steps as discussed in the previous chapter. However,
it is believed that not all fibre recovered and an optimised separation sequence
could potentially increase fibre yield further.

After comminution and separation, feather was discovered stuck between the
pump impeller and pump chamber (Figure 43); Most of this was uncut rachis.
Theses feathers were accounted for as residuals in the mass balance. It was found
in subsequent trials that the position of the discharge pipe into the tank strongly
influenced the flow pattern in the tank. If adequate mixing is forced by inducing a
strong vortex, fibre entrapment in the impeller chamber was also reduced. The
chamber was opened to remove all traces of rachis. It would be good practice not
to allow rachis to get stuck between impeller blade and chamber because any
abrasive material could wear down the impeller and reduce pump efficiency.
There was no permanent damage on the impeller shown in Figure 44.

Other problems encountered during processing were feathers floating on the

surface and cavitation in the pump. Floating feather meant that they were not
circulated through the comminution cycle lading to low fibre yield. This problem
was solved by adjusting the position of the discharge pipe to create a vortex
dragging feathers down to the suction end of the pump. However, at high pump
speeds a significant air would be introduced into the tank leading to cavitation,
making the process inefficient.

71
Figure 43: Feather deposition in impeller chamber.

Figure 44: Comminution centrifugal pump impeller.

72
It was interesting to note that the moisture content of fibre were higher than
feather (Table 26). Fibre contained (90 5) % moisture and the uncut feather
contained 81 4 %. This indicated that comminution opened voids in the feather
by cutting apart the rachis and reducing the amount of hidden surface area
between parallel barbs.

Table 26: Comminution moisture content and fibre yield.

Moisture Content (%) Fibre Yield (%)
Time (h) Run 1 Run 2 Run 1 Run 2
Run 1 Run 2
Fibre Fibre Feather Feather
0 74.36 96.83 81.69 88.73 0.95 0.98
1 87.99 96.21 79.78 88.08 15.38 6.02
2 90.10 94.23 83.37 90.23 6.77 10.40
3 93.58 95.57 87.11 91.39 20.08 6.36
4 95.36 80.71 1.90

The average fibre yield increased over time for the first experiment, but decreased
at hours 3 and 4 during the second experiment. Considering that the second
experiment had an overall higher fibre yield, this is a somewhat unexpected result.
It was thought that the inherent heterogeneity of the system was highlighted in
this result. At any given time a significant amount of feather may be contained in
the pipes or pump chamber leading to a feather consistence lower than expected in
tank. Ensuring good fibre recovery would imply fairly long processing times to
overcome this.

Bulk feather suspension temperature increased over time (Figure 45). Slurry
characteristics are expected to be highly dependent on temperature. Bulk slurry
viscosity, Reynolds number and head loss were calculated and graphed in Figure
A 5 to Figure A 7, respectively. Furthermore, it can be expected that comminution
may also be different at higher temperatures. According to literature, only around
1% of energy during comminution is imparted to the material. The rest ends up
heating the surrounding medium. It is evident from the large temperature
difference between run 1 and 2 and the experiment without feather (Figure 45).
Feathers are mostly protein and changing the operating temperature could lead to
degradation. It may therefore be required to cool the comminution circuit when
done on a large scale, or if longer comminution times are required.

73
 Experiment 1 Experiment 2 No Feather

80
70
60
50
T
40
(oC)
30
20
10
0
0 1 2 3 4
t (h)

Figure 45: Comminution slurry temperature change through time.

4.5.1 Macroscopic Morphology

The fibres and feathers from comminution are displayed in Figure 45.

Fibre Feather

Figure 46: Dried and packaged fibre and feather from comminution.

Figure 47 shows photographs of fibre and feather fraction sampled at each hour of
comminution. The amount of fibre recovered from samples increased over time,
however, the feather fraction still had significant amount of entrained fibre.

74
 t (h) Fibre Feather
A B

C D

E F

G H

I J

Figure 47: Comminution feather and fibre photographs.

75
4.5.2 Stereoscopic Morphology
The fibre and feather fractions were examined under an optical microscope.
Morphology prior to comminution is shown in Figure 48 and feathers were also
filtered before comminution to qualitatively assess the degree to which fibres
detached from the rachis after decontamination and cleaning.

Fibre Feather
A B

6,4X

C D

16X

E F

40X

Figure 48: Feather morphology before comminution.

In Figure 5A feather with most of its fibre intact are shown. Figure 48C contains a
highly concentrated area of fibre. Figure 48E is the same area, where the nodes
are barbules and barbicels. The bundled feather shown in in Figure 48B did not go
pass through the 1 mm filter. Figure 48D shows that particles in the feather
fraction were larger and had more tortuous bends. Nodular substructures were also
present in Figure 48F.

Fibre and feather after 1 hour of comminution are shown in Figure 49.

76
 Fibre Feather
6,4X A B

16X C D

40X E F

Figure 49: Feather and fibre morphology after 1 hour of comminution.

Fibres in Figure 49A were about 10 mm long and very fine. Short and relatively
straight rachis of about 8 mm long passed through the 1 mm filter because its
diameter (labelled with arrows in Figure 49E) was only 100 m. Figure 49C
showed how similar this group of rachis were compared to the barbs. A bundle of
barbs has been partially sheared in B and the faint line down the middle of Figure
49D indicates where the impeller has made contact. From Figure 49F it can be
seen that the width of a cut was less than 500 m.

Figure 50 are images of feather and fibre after 3 hours comminution.

77
 Fibre Feather
A B

6,4X

C D

16X

E F

40X

Figure 50: Feather and fibre morphology after 2 hours of comminution.

The centre of Figure 50A shows a sheared rachis, bent in an obtuse angle. In
Figure 50C a 2 mm wide barb with a few missing barbules is shown. The 10 mm
rachis in Figure 50E has had its barbs sheared to approximately 200 m wide. The
4 mm wide, over 1 cm long semi-plume in Figure 50B still contains a lot of fibre
after 2 hours of rigorous cutting. Figure 50D and Figure 50F highlights the
durability of rachis against mechanical deformation, still showing highly uniform
barbules. The translucent rachis width was 200 m, while the spaces between
parallel barbs were 100 m.

Figure 51 are feather and fibre after 3 hours of comminution.

78
 Fibre Feather
6,4X A B

16X C D

40X E F

Figure 51: Feather and fibre morphology after 3 hours of comminution.

Figure 51A shows a small 200 m wide, 2 mm long bundle containing only barbs.
The down fibre to its right curls in different directions. Figure 51C shows plenty
of nodules. Figure 51E is a bundle showing a wealth of fibre in its early
developmental stage. Figure 51B shows a 12 mm rachis bent in an acute angle
with barbs sheared off. Figure 51D is a 6 mm wide developing bundle with
differentiated barbules with a pinched mid-section and Figure 51F is the bent
rachis, where remaining barbs act as hooks to prevent the rachis from passing
through the filter.

Fibres after comminution shown in Figure 52 were finer and shorter.

79
 Fibre Feather
A B

6,4X

C D

16X

E F

40X

Figure 52: Feather and fibre morphology after 4 hours of comminution.

Figure 52A shows a down feather, a thin rachis and barbules. Figure 52C is a barb
fibre with a substantial amount of barbules and sub-structural nodes are present in
Figure 52E. Feathers such as the one shown in Figure 52B did not get sheared
because it was smaller than the impeller clearance. In Figure 52D shows some
fibre that should have passed through the filter; the very small piece of down
feather remained due to its width. The sickled-shaped rachis was bent at a sharp
angle. Figure 52E and F shows the down feather maintained its structural
integrity.

From these images it is clear that comminution has reduced average feather size,
and that some rachis was recovered with fibres. Also, some usable fibre remained
in the feather fraction. Comminution was found to be a process whereby fractions
of feather were stripped from larger rachis, as opposed to fracturing distinct
feather rachis. Over time, the fibre fraction reduced in size, although qualitative

80
data was not presented in this section. It is clear that longer comminution times
are required to produce a fibre product of reasonable quality.

4.5.3 Histological Morphology

Figure 53 shows fibre surface for every hour of comminution. The parallel barbs
in Figure 53A have a diameter of 10 um. In Figure 53B, the barbs have chipped
edges. After 2 hours, the rachis still had smooth surfaces as shown in Figure 53C
and a piece of powder feather remains in the lower middle section of the image.
Figure 53D shows where the barbs are slightly torn from the rachis.

T (h) 1000X 3500X

A B

C D

E F

81
 G H

Figure 53: Fibre surface morphology in comminution.

At 3 hours, Figure 53E shows a bundle of barbs and Figure 53F is the barb with
peeled layers. After 4 hours of comminution, the barbules in Figure 53G showed
serrated barbicels. The barbule in Figure 53H had surface striations. This artefact
is unlikely to be from pump impeller shear, but may be the effect of drying.

Many stages of filtration in series were necessary, which indicate continuous

filtration would result in better separation. The proposed method is to cycle the
suspension through a pressure screen controlled by valves that thickens the fibre
fraction. A process diagram was devised to summarise decontamination, cleaning,
comminution, filtration and drying Figure A 8.

82
5 Conclusions
Raw feather supplied by Wallace Corporation contained about 15% hexane
extractables such as preen oil, as well as other contaminants such as offal, faecal
matter and process water. Treating feather with sodium dodecyl sulphate solutions
did not reduce the hexane extractable content, while high concentration ethanol or
hydrogen peroxide solutions were able to reduce hexane extractables to about
10% in one equilibrium stage.

Large-scale feather treatment required 2 equilibrium stages for decontamination

using sodium hypochlorite and 3 stages for cleaning. Each equilibrium stage
contained 25 L solution at a concentration of 0.1485% sodium hypochlorite
adjusted to pH 10.0 with 1 M sodium hydroxide for decontamination and 0.15%
hydrogen peroxide for cleaning. A Lamort pulper with a disc impellor was used to
mix the feather suspension at 10 Hz for 30 minutes in each stage.

Bacteriological testing found no Salmonella, Campylobacter or

Enterobacteriaceae on treated feather. Using colour analysis it was revealed that
raw feather showed 76% whiteness, which was increased to 86% whiteness after
three stages hydrogen peroxide washing. Clean feathers were fluffy and off-white,
compared to a brownish yellow colour of raw feather with a foul odour.

Clean feather was comminuted in a stainless steel tank containing 300 L water
using a centrifugal pump at 30 Hz impeller speed under full recycle for up to 4
hours. At the end of the process partially cut feather was separated from fibre
using a series of gravity filters and dried.

Half of the mass of feather is rachis, so the theoretical fibre yield after
comminution was 50%. It was found that the yield after three hours comminution
was 15% and increased to 27% after four hours. Morphological studies revealed
that the fibre surface was not damaged by this comminution method. Fibre
product consists of barbs with intact barbules and thin rachis with barbs sheared
off.

83
5.1 Recommendations
The large scale process could be improved to produce fibre quality that is more
consistent with that obtained in lab scale processing, in terms of fibre whiteness
and hexane extractable content.

Information on fibre diameter and length distribution at different operating

conditions would be helpful for optimising pump impeller selection as well as
pipes and valve arrangement that would reduce cavitation and fibre deposition.

The recovery process should use continuous filtration, such as the Beloit pressure
screen.

Production scale development should consider equipment available within the

rendering industry and can be supplemented with those available from the pulp
and paper industry. These could include continuous filtration equipment, such as a
Beloit pressure screen.

Further testing such as single fibre tensile tests would allow comparison with
other types of fibre. Lastly, the product should be tested in applications such as
insulating materials and composites to determine its suitability.

84
6 References
1. Schmidt, W.F. and S. Jayasundera, Microcrystalline avian keratin protein
fibres, in Natural fibers, plastics and composites, F.T. Wallenberger and
N.E. Weston, Editors. 2004, Kluwer Academic Publishers: Boston.
2. The Poultry Industry Association of New Zealand. [Website] 2006 [cited
2010 June 10]; Available from: http://www.pianz.org.nz/.
3. Meat in focus: a closer look at a key New Zealand industry. 2009,
Wellington, New Zealand: The Meat Industry Association. 28.
4. Roff, W.J., J.R. Scott, and J. Pacitti, Fibres, Films, Plastics and Rubbers:
A handbook of Common Polymers. 1 ed. 1971, London, UK: Buterworths.
688.
5. Shah, H. and Y. Yang, Feather Fiber Reinforced Light-Weight Composites
with Good Acoustic Properties. Journal of Polymers and the Environment,
2009. 17(2): p. 131-142.
6. Tompkin, R.B., A.M. McNamara, and G.R. Acuff, Meat and Poultry
Products, in Compendium of Methods for the Microbiological
Examination of Foods, F.P. Downes and K. Ito, Editors. 2001, American
Public Health Association: Washington, DC. p. xxi, 676.
7. Durham, S., Feather Fiber Technology Receives a "World's Best" Award.
2004, Agricultural Research Service.
8. Parkinson, G., Chementator: a higher use for lowly chicken feathers, in
Chemical Engineering. 1998. p. 21.
9. USDA, USDA wants partner for feather fiber development. Industrial
Bioprocessing, 2003. 25(Compendex): p. 3-4.
10. Stuart, T., et al., Structural biocomposites from flax - Part I: Effect of bio-
technical fibre modification on composite properties. Composites Part a-
Applied Science and Manufacturing, 2006. 37(3): p. 393-404.
11. Shibata, M., et al., Biocomposites made from short abaca fiber and
biodegradable polyesters. Macromolecular Materials and Engineering,
2003. 288(1): p. 35-43.
12. Jones, C.W., Applications of Hydrogen Peroxide and Derivatives. RSC
Clean Technology Monographs. 1999, Cambridge, UK: The Royal Society
of Chemistry.

85
13. Ministry of Agriculture and Forestry. About MAF. 2011 [cited 2011
August 13]; Available from: http://www.maf.govt.nz/about-maf.
14. Montazer, M., M. Zargaran, and A. Rahimi, Depigmentation of Pigmented
Wool. Textile Research Journal, 2009. 79(3): p. 261-267.
15. Inghams. Inghams Enterprises New Zealand: a New Zealand Chicken
company. 2011 [cited 2011 29 July]; Available from:
www.inghams.co.nz.
16. Tegel. Tegel New Zealand. 2011 [cited 2011 29 July]; Available from:
www.tegel.co.nz.
17. Brink, P.v.d. Brink's: Enjoy the difference care makes. 2011 [cited 2011
29 July].
18. Turk's. TURK's - from our family to yours. 2011 [cited 2011 29 July];
Available from: www.turkspoultry.com.
19. Appleby, M.C., B.O. Hughes, and H.A. Elson, Poultry production systems:
behaviour, management and welfare. 1992, Wallingford: C.A.B
International. xvi, 238.
20. Whole Chicken and Edible Cuts. 2009, Sun Belt Foods. Retrieved from
http://sunbeltfoods.com/: Boca Raton, Florida.
21. Papadopoulos, M.C., Processed chicken feathers as feedstuff for poultry
and swine. A review. Agricultural Wastes, 1985. 14(4): p. 275-290.
22. Cheung, H.-y., et al., Natural fibre-reinforced composites for
bioengineering and environmental engineering applications. Composites
Part B: Engineering, 2009. 40(7): p. 655-663.
23. Burton, C.H., T.R. Cumby, and D.B. Tinker, Poultry Meat Processing and
Quality. 2004, Cambridge, , GBR: Woodhead Publishing, Limited. 408.
24. Mokrejs, P., et al., Processing poultry feathers into keratin hydrolysate
through alkaline-enzymatic hydrolysis. Waste Management and Research,
2011. 29(Compendex): p. 260-267.
25. Bock, W.J., Birds, in Encyclopaedia of Life Support Systems, G.
Contrafatto and A. Minelli, Editors. 2004, EOLSS: Oxford, UK.
26. Yu, M., et al., The morphogenesis of feathers. Nature, 2002.
420(Compendex): p. 308-312.

86
27. Harrop, B.S. and E.F. Woods, Soluble Derivatives of Feather Keratin 1.
Isolation, Fractionation and Amino Acid Composition. Biochemical
Journal, 1964. 92(8): p. 8-18.
28. Fraser, R.D.B. and D.A.D. Parry, Molecular packing in the feather keratin
filament. Journal of Structural Biology, 2008. 162(1): p. 1-13.
29. Richardson, M.K. and L.G.-d. Roo, Cell and tissue structure in animals
and plants, in Encyclopaedia of Life Support Systems, G. Contrafatto and
A. Minelli, Editors. 2003, EOLSS: Oxford, UK.
30. Rajchard, J., Biologically active substances of bird skin: a review.
Veterinarnin Medicina, 2010. 55(9): p. 413-421.
31. Wertz, P.W., P.M. Stover, and D.T. Downing, A survey of polar and
nonpolar lipids from epidermis and epidermal appendages of the chicken
(Gallus domesticus). Comparative Biochemistry and Physiology Part B:
Comparative Biochemistry, 1986. 84(2): p. 203-206.
32. Gassner III, G., et al., US5,705,030 Fiber and fiber products produced
from feathers. 1998, The United States of America as represented by the
Secretary of (Washington, DC): United States.
33. Thomas, G. International Year of Natural Fibres. 2009 [cited 2011
August 8]; Official web site of the United Nations International Year of
Natural Fibres 2009]. Available from: http://www.naturalfibres2009.org/.
34. Li, Y., Processing of hemp fibre using enzyme/fungal treatment for
composites, in the department of science and engineering. 2009,
University: Hamilton. p. 173.
35. Wollerdorfer, M. and H. Bader, Influence of natural fibres on the
mechanical properties of biodegradable polymers. Industrial Crops and
Products, 1998. 8(2): p. 105-112.
36. Kruchen, E., US4,169,706 Method of cleaning poultry feathers. 1979:
United States.
37. Fleet, C.R.W. and V.K. Hewinson, US4,537,594 Processed feathers. 1985,
Fogarty P. L. C. (Lincolnshire, GB2): United States.
38. Shah, H. and Y. Yang, Composites from ground chicken quill and
polypropylene. Composites Science and Technology, 2008. 68(3-4): p.
790-798.

87
39. Dweib, M.A., et al., All natural composite sandwich beams for structural
applications. Composite Structures, 2004. 63(2): p. 147-157.
40. Hong, C.K. and R.P. Wool, Development of a bio-based composite
material from soybean oil and keratin fibers. Journal of Applied Polymer
Science, 2005. 95(6): p. 1524-1538.
41. Zhan, M.J., R.P. Wool, and J.Q. Xiao, Electrical Properties of Chicken
Feather Fiber Reinforced Epoxy Composites. Composites Part A: Applied
Science and Manufacturing. 42(3): p. 229-233.
42. Brandelli, A., Bacterial Keratinases: Useful Enzymes for Bioprocessing
Agroindustrial Wastes and Beyond. Food and Bioprocess Technology,
2008. 1(2): p. 105-116.
43. Ahmaruzzaman, M., Adsorption of phenolic compounds on low-cost
adsorbents: A review. Advances in Colloid and Interface Science, 2008.
143(1-2): p. 48-67.
44. Ishikawa, S. and K. Suyama, Recovery and refining of Au by gold-cyanide
ion biosorption using animal fibrous proteins. Applied Biochemistry and
Biotechnology, 1998. 70-2: p. 719-728.
45. Zheng, M., et al., Hydrogenation of anisol and benzaldehyde catalyzed
bay chicken feather-palladium complex. Polymers for Advanced
Technologies, 1997. 8(11): p. 638-640.
46. Sun, P., Z.T. Liu, and Z.W. Liu, Particles from bird feather: A novel
application of an ionic liquid and waste resource. Journal of Hazardous
Materials, 2009. 170(2-3): p. 786-790.
47. Teixeira, M.C. and V.S.T. Ciminelli, Development of a biosorbent for
arsenite: Structural modeling based on x-ray spectroscopy. Environmental
Science and Technology, 2005. 39(Compendex): p. 895-900.
48. Teixeira, M.C., et al., Raman spectroscopy and DFT calculations of As(III)
complexation with a cysteine-rich biomaterial. Journal of Colloid and
Interface Science, 2007. 315(1): p. 128-134.
49. Senoz, E. and R.P. Wool, Microporous carbonnitrogen fibers from
keratin fibers by pyrolysis. Journal of Applied Polymer Science, 2010.
118(3): p. 1752-1765.
50. Senoz, E. and R.P. Wool, Hydrogen storage on pyrolyzed chicken feather
fibers. Journal of Applied Polymer Science, 2011(Compendex).

88
51. Tiwary, E. and R. Gupta, Extracellular Expression of Keratinase from
Bacillus licheniformis ER-15 in Escherichia coli. Journal of Agricultural
and Food Chemistry, 2010. 58(14): p. 8380-8385.
52. Shrinivas, D. and G.R. Naik, Characterization of alkaline thermostable
keratinolytic protease from thermoalkalophilic Bacillus halodurans JB 99
exhibiting dehairing activity. International Biodeterioration &
Biodegradation, 2010. 65(1): p. 29-35.
53. Fakhfakh-Zouari, N., et al., A Novel Serine Metallokeratinase from a
Newly Isolated Bacillus pumilus A1 Grown on Chicken Feather Meal:
Biochemical and Molecular Characterization. Applied Biochemistry and
Biotechnology, 2010. 162(2): p. 329-344.
54. Fakhfakh-Zouari, N., et al., Application of statistical experimental design
for optimization of keratinases production by Bacillus pumilus A1 grown
on chicken feather and some biochemical properties. Process
Biochemistry, 2010. 45(5): p. 617-626.
55. Fakhfakh, N., et al., Production and biochemical and molecular
characterization of a keratinolytic serine protease from chicken feather-
degrading Bacillus licheniformis RPk. Canadian Journal of Microbiology,
2009. 55(4): p. 427-436.
56. Hmidet, N., et al., Chicken feathers: a complex substrate for the co-
production of alpha-amylase and proteases by B. licheniformis NH1.
Journal of Industrial Microbiology & Biotechnology, 2010. 37(9): p. 983-
990.
57. Haddar, A., et al., Characterization of detergent stable and feather
degrading serine proteases from Bacillus mojavensis A21. Biochemical
Engineering Journal, 2010. 51(1-2): p. 53-63.
58. Jeong, J.-H., et al., Keratinolytic enzyme-mediated biodegradation of
recalcitrant feather by a newly isolated Xanthomonas sp. P5. Polymer
Degradation and Stability, 2010. 95(10): p. 1969-1977.
59. Jeong, J.H., et al., Production of keratinolytic enzyme by a newly isolated
feather-degrading Stenotrophomonas maltophilia that produces plant
growth-promoting activity. Process Biochemistry, 2010. 45(10): p. 1738-
1745.

89
60. Jeong, J.H., et al., Characterization of a multifunctional feather-degrading
Bacillus subtilis isolated from forest soil. Biodegradation, 2010. 21(6): p.
1029-1040.
61. Prakash, P., S.K. Jayalakshmi, and K. Sreeramulu, Production of
Keratinase by Free and Immobilized Cells of Bacillus halodurans Strain
PPKS-2: Partial Characterization and Its Application in Feather
Degradation and Dehairing of the Goat Skin. Applied Biochemistry and
Biotechnology, 2010. 160(7): p. 1909-1920.
62. Jaouadi, B., et al., Excellent Laundry Detergent Compatibility and High
Dehairing Ability of the Bacillus pumilus CBS Alkaline Proteinase (SAPB).
Biotechnology and Bioprocess Engineering, 2009. 14(4): p. 503-512.
63. Wang, H.R. and P. Wei, Modification of feather keratin and its retanning
properties. Journal of the Society of Leather Technologists and Chemists,
2006. 90(6): p. 254-259.
64. Farag, A.M. and M.A. Hassan, Purification, characterization and
immobilization of a keratinase from Aspergillus oryzae. Enzyme and
Microbial Technology, 2004. 34(2): p. 85-93.
65. Vermelho, A.B., et al., Identification of a Candida parapsilosis Strain
Producing Extracellular Serine Peptidase with Keratinolytic Activity.
Mycopathologia, 2010. 169(1): p. 57-65.
66. Park, G.T. and H.J. Son, Keratinolytic activity of Bacillus megaterium F7-
1, a feather-degrading mesophilic bacterium. Microbiological Research,
2009. 164(4): p. 478-485.
67. Stern, N.J., J.E. Line, and H.-C. Chen, Campylobacter, in Compendium of
Methods for the Microbiological Examination of Foods, F.P. Downes and
K. Ito, Editors. 2001, American Public Health Association: Washington,
DC. p. xxi, 676.
68. AgResearch, Testing Meats and Meat Produts for Pathogens -
Thermotolerant Campylobacter, in Meat Industry Microbiological
Methods. 2008, AgResearch: Hamilton. p. 7.
69. AgResearch, Testing Meats and Meat Products for Pathogens -
Salmonella, in AgResearch Microbiological Methods for the Meat
Industry. 2011, AgResearch: Hamilton. p. 11.

90
70. Kornacki, J.L. and J.L. Johnson, Enterobacteriaceae, Coliforms, and
Escherichia coli as Quality and Safety Indicators, in Compendium of
Methods for the Microbiological Examination of Foods, F.P. Downes and
K. Ito, Editors. 2001, American Public Health Association: Washington,
DC. p. xxi, 676.
71. Augurt, T.A. and J. van Asten, Sterilization Techniques, in Kirk-Othmer
Encyclopedia of Chemical Technology. 2000, John Wiley & Sons, Inc.
72. Griffith, B.A., US0,148,572 Feather molding method and product. 2002,
Tyson Foods, Inc.: United States.
73. Gullichsen, J., 2. Fiber line operations, in Chemical Pulping:
Papermaking Science and Technology, J. Gullichsen and H. Paulapuro,
Editors. 1999, Finnish Paper Engineers' Associaiton and TAPPI: Helsinki,
Finland. p. 19-244.
74. Gullichsen, J. and C.-J. Fogelholm, 9. Bleaching applications, in Chemical
Pulping: Papermaking Science and Technology, J. Gullichsen and H.
Paulapuro, Editors. 1999, Finnish Paper Engineers' Associaiton and
TAPPI: Helsinki, Finland. p. 603-616.
75. Aguilera, J.M., 2. Solid-Liquid Extraction, in Extraction optimization in
food engineering, C. Tzia and G. Liadakis, Editors. 2003, Marcell Dekker:
New York. p. x, 442.
76. Eggers, R. and P.T. Jaeger, 4. Extraction Systems, in Extraction
optimization in food engineering, C. Tzia and G. Liadakis, Editors. 2003,
Marcell Dekker: New York. p. x, 442.
77. Wypych, G., Solvents, Industrial, in Kirk-Othmer Encyclopedia of
Chemical Technology. 2000, John Wiley & Sons, Inc.
78. Tzia, C., 5. Optimization, in Extraction optimization in food engineering,
C. Tzia and G. Liadakis, Editors. 2003, Marcell Dekker: New York. p. x,
442.
79. Council, W.D., Waikato District Council Water Supply Bylaw 2009. 2009,
Waikato District Council: Hamilton. p. 27.
80. Helleur, H., Matamata Piako Distric Trade Waste Bylaw. 2006, Matamata
Piako District Council: New Zealand. p. 68.
81. Prior, M., Size Reduction, in Kirk-Othmer Encyclopedia of Chemical
Technology. 2000, John Wiley & Sons, Inc.

91
82. Sutherland, K., Filters and filtration handbook. 5th ed. ed. 2008, Oxford:
Elsevier/Butterworth-Heinemann. xii, 523.
83. Ljokkoi, R., et al., 8. Pulp screening applications, in Chemical Pulping:
Papermaking Science and Technology, J. Gullichsen and H. Paulapuro,
Editors. 1999, Finnish Paper Engineers' Associaiton and TAPPI: Helsinki,
Finland. p. 573-602.
84. Gullichsen, J. and C.-J. Fogelholm, 10. Pulp drying applications, in
Papermaking Science and Technology, J. Gullichsen and H. Paulapuro,
Editors. 1999, Finnish Paper Engineers' Associaiton and TAPPI: Helsinki,
Finland. p. 617-666.
85. Leica, Wild Leica M3B Stereomicroscope. 2009: Netherlands.
86. Hitachi, Hitachi Scientific Instrument Technical Data SEM Sheet No. 101.
2003, Tokyo, Japan: Hitachi High-Technologies Corporation. 6.
87. Konica-Minolta, Konica Minolta Chroma Meter Cr-200/CR-210, CR-
221/CR-231. 3 ed. 2006, Osaka, Japan: Konica Minolta Sensing, Inc.
88. Stoltz, M.J., US6,827,948 Method and compositions for processing poultry
feathers. 2004, Steen Research LLC (West Linn, OR): United States.
89. AgResearch, Enteric Indicators for Meat and Meat Processing -
Enterobacteriaceae, in Meat Industry Microbiological Methods. 2006,
AgResearch: Hamilton. p. 3.
90. Irotek. Computer Display Conversion. 2011 [cited 2011 July 8];
Available from: www.easyrgb.com/.

92
Appendices

93
Appendices
Figure A 1: Feather rachis amino acid mass composition [21]. ............................ 81
Figure A 2: Feather barbs amino acid mass composition [21]. ............................. 82
Figure A 3: Feather calamus amino acid mass composition [21]. ........................ 82
Figure A 4: Feather medulla amino acid mass composition [21]. ......................... 83
Table A 1: Rendering equipment manufacturers ................................................... 84
Table A 2: Company in the poultry by-products industry ..................................... 84
Table A 3: Waste water bylaw summary [79, 80].................................................. 84
Table A 5: Procedure used by NZ Lab for bacteriological test. ............................ 85
Table A 6: Olympus BX2 series microscope technical data [6] ............................ 85
Table A 7: Konica Minolta CR-410 Chroma Meter Technical Data [7] ............... 86
Table A 8: L*a*b* conversion to XYZ and RGB [90].......................................... 86
Table A 9: Lab-scale Soxhlet extraction sample treatment description. ............... 88
Table A 10: Lab-scale Soxhlet extraction calculations. ........................................ 88
Table A 11: Lamort pulper specification. .............................................................. 91
Table A 12: Centrifugal pump technical information. ........................................... 91
Table A 13: Comminution enthalpy calculations. ................................................. 92
Figure A 5: Bulk slurry viscosity during comminution......................................... 92
Figure A 6: Slurry Reynolds number during comminution. .................................. 92
Figure A 7: Comminution system head loss due to fittings. ................................. 93
Figure A 8: Process flow diagram. ........................................................................ 94

94
Appendix 1: Chicken Feather

Figure A 1: Feather rachis amino acid mass composition [21].

Figure A 2: Feather barbs amino acid mass composition [21].

95
Figure A 3: Feather calamus amino acid mass composition [21].

Figure A 4: Feather medulla amino acid mass composition [21].

96
Table A 1: Rendering equipment manufacturers
Rendering Equipment Manufacturers Description
Alfa Laval Heat transfer, centrifugal separation and fluid
handling equipment for the poultry processing
plants.
GEA Westfalia Separator Group Mechanical separation products.
(est. 1893)
Napier Engineering and Niven Abattoir equipment for rendering
Contracting Ltd. (est. 1866) plants.
Pacific Terminals NZ Handle and store bulk flammable and
combustible liquids including chemicals,
edible oils and fats.
Pinches Consolidated Industries Processing and drying technology using
(est. 1947) superheated steam dryers.
Rendertech Protein recovery and Rendering, wastewater
(est. 1988) treatment, odour control and bulk storage
tanks.
Transpacific Industries Group Provides waste and environment services by
(NZ) Ltd (TPI) transporting liquid, hazardous and solid waste
to recover resources and manage waste
responsibly.

Table A 2: Company in the poultry by-products industry

Company Description
Tradeskins NZ Ltd (est. 1992) Leather trading company exporting skins, hides,
pelts and rendered animal products
Ritex International Ltd. (est. 1988) Imports industrial and household textile
packaging products
Kemin Industries NZ Pet food manufacturer
Mars Petcare New Zealand Canned, chilled, dry and shelf-stable cat and dog
food and treats.

Table A 3: Waste water bylaw summary [79, 80]

Compound or Test Name Amount
Solids dimensions Less than 15 mm
Suspended solids 2 kg m-3
Settling solids 50 mL L-1
Biodegradable oil, fat or grease1 Less than 500 g m-3
Colour2 Absorbance less than 0.3010
Transmissivity greater than 50%
Acidity or alkalinity pH 6.0 to 10.0
Five-day biochemical oxygen demand (BOD5) Less than 1 kg m-3
Methylene Blue Active Substances 300 g m-3
Chlorine (Cl2)
Free chlorine 5 g m-3
Hypochlorite 10 g m-3
1 When emulsion is stable at 15oC, or the emulsion is diluted by a factor of 10 by raw
sewage at pH 4.5 to 10.0
2 effluent diluted by a factor of 10 with distilled water and tested at 254 nm

97
Appendix 2: Analysis Procedures
Table A 4: Potential pathogens found on chicken feather.
Microorganisms Description
Campylobacter Small, spiral, gram-negative, non-spore-forming rods, cork-screw
[68] motility with polar flagella, micro-aerophilic, oxidase positive,
reduce nitrate
Enterobacteriaceae Gram-negative, non-spore-forming bacilli, facultative anaerobic,
[89] ferments glucose, reduce nitrates to nitrites, grows on bile salts
media and are oxidase negative
Salmonella Motile Enterobacteriaceae, catalase positive, oxidase negative
[69] Biochemistry: H2S production, lysine decarboxylation, negative
urease reaction, uses citrate as carbon source, ferments glucose to
produce gas, no lactose or sucrose fermentation.

Table A 5: Procedure used by NZ Lab for bacteriological test.

Preparation Weigh, dilute and emulsify or suspend as required; dilutions
(APHA (4) 2.527) in 0.1% peptone water
Campylobacter Filter 100 mL of sample using 0.5 m filter. Enrichment in
MIMM 7.3. 4th Ed Bolton Broth, incubation at 37oC for 4 h in microaerophilic
(Modified) atmosphere, then at 42oC for 44 h. Isolation of mCCda, incubate
[68] at 42oC for 48 h in microaerophilic atmosphere. Confirmation by:
oxidase, Gram stain, motility, Growth at 37oC, antibiotic
sensitivity, hippurate hydrolysis
Enterobacteriaceae Count on Petrifilm, incubate at 37oC for 24 h.
MIMM 8.2
(Modified) [89]
Salmonellae Filter 100 mL of sample using 0.45m filter. Resuscitation
MIMM 7.7 Buffered Peptone water, Enrichment Rappaport-Vassiliadis Soya
(Modified) peptone broth, Isolation XLD, Modified Brilliant Green agars,
[69] Confirmation using serabact or Entrotube and serological
screening. If necessary samples are sent to ESR-CDC Porirua for
confirmation. Limits of detection: Not applicable.
Presence/absence test.

Table A 6: Olympus BX2 series microscope technical data [6]

Side cross section

Microscope Optical System UIS optical system
frame Focus Vertical: 25 mm stage stroke with coarse
adjustment limit stopper, torque adjustment
for coarse knobs
Stage mounting positions with high

98
 sensitivity fine focusing knob (minimum
adjustment at 1 m.)
Illuminator Koehler illumination for transmitted light.
12V100W halogen bulb, light preset switch,
intensity LED indicator and filters
Observation Widefield Binocular inclined 30o, Tilting bionocular
Tube inclined 5o 35o, Trinocular inclined 30oand
Ergo binocular inclined 0o 25o
Super Widefield Trinocular inclined 24o
Stage Coaxial, rotational control and torque
adjustment
Condenser Options allow magnification range 1.25x to
100x

Front and Side Line drawings

Table A 7: Konica Minolta CR-410 Chroma Meter Technical Data [7]

Equipment Chroma Meter and its Data Processor
Model CR-410 Data Processor DP-400
Illumination Wide-area Viewing Angle 0o
Light source Pulsed xenon lamp Detector 6 Silicone cells
Display range Y: 0.01 to 160.00% Battery 800 measurements
(reflectance) performance
Measurement 1s Measurement 3s
time interval
Measurement 50 illumination area 53
area
Colour space L*a*b* Illuminant C, D65.
Operating 0 to 40oC Operating 85% relative humidity
temperature Humidity
Observer: 2 degrees, closely matches CIE 1931 standard observers
Repeatability: Within E*ab0.7 standard deviation (white plate measured 30 times at 10 s
intervals)
Inter-instrument agreement: E*ab: within 0.8 average of 12 BCRA series II colours

Table A 8: L*a*b* conversion to XYZ and RGB [90].

Step Formula and Condition

1 var_Y = ( L* + 16 ) / 116
var_X = a* / 500 + var_Y
var_Z = var_Y b* / 200

99
2 if ( var_Y^3 > 0.008856 ) var_Y = var_Y^3
else var_Y = ( var_Y - 16 / 116 ) / 7.787
if ( var_X^3 > 0.008856 ) var_X = var_X^3
else var_X = ( var_X - 16 / 116 ) / 7.787
if ( var_Z^3 > 0.008856 ) var_Z = var_Z^3
else var_Z = ( var_Z - 16 / 116 ) / 7.787
3 X = ref_X * var_X, //ref_X = 95.047, Observer= 2, Illuminant= D65
Y = ref_Y * var_Y, //ref_Y = 100.000
Z = ref_Z * var_Z, //ref_Z = 108.883
4 var_X = X / 100, //X from 0 to 95.047 , Observer = 2, Illuminant = D65
var_Y = Y / 100, //Y from 0 to 100.000
var_Z = Z / 100, //Z from 0 to 108.883
5 var_R = var_X * 3.2406 + var_Y * -1.5372 + var_Z * -0.4986
var_G = var_X * -0.9689 + var_Y * 1.8758 + var_Z * 0.0415
6 var_B = var_X * 0.0557 + var_Y * -0.2040 + var_Z * 1.0570
if ( var_R > 0.0031308 ) var_R = 1.055 * ( var_R ^ ( 1 / 2.4 ) ) - 0.055
else var_R = 12.92 * var_R
if ( var_G > 0.0031308 ) var_G = 1.055 * ( var_G ^ ( 1 / 2.4 ) ) - 0.055
else var_G = 12.92 * var_G
if ( var_B > 0.0031308 ) var_B = 1.055 * ( var_B ^ ( 1 / 2.4 ) ) - 0.055
else var_B = 12.92 * var_B
7 R = var_R * 255
G = var_G * 255
B = var_B * 255

100
Appendix 3: Lab-scale Treatment
Table A 9: Lab-scale Soxhlet extraction sample treatment description.
Label Description Label Description
X1 Unprocessed feather HA H2O2 0.35 wt%
X2 Unprocessed feather HB H2O2 0.45 wt%
X3 Unprocessed feather HC H2O2 0.55 wt%
X4 Unprocessed feather DA SDS 0.00057 wt%
X5 Unprocessed feather DB SDS 0.00114 wt%
X6 Unprocessed feather DC SDS 0.00171 wt%
J1 0.21% NaOCl for 5 min DD SDS 0.00229 wt%
J2 0.525% NaOCl for 5 min DE SDS 0.00286 wt%
J3 1.25% NaOCl for 5 min E1-3 19.9% ethanol
J4 0.21% NaOCl for 10 min E4-6 39.8% ethanol
J5 0.525% NaOCl for 10 min E7-9 59.7% ethanol
J6 1.25% NaOCl for 10 min E10-12 79.6% ethanol
EA 5 % ethanol E13-15 99.5% ethanol
EB 7% ethanol H1-3 0.15% H2O2
EC 10% ethanol H4-6 0.25% H2O2
ED 14% v/v ethanol H7-9 0.35% H2O2
EE 29% v/v ethanol H10-12 0.45% H2O2
EF 43% v/v ethanol H13-15 0.55% H2O2
EG 57% v/v ethanol D1-3 0.0004% SDS
EH 71% v/v ethanol D4-6 0.0008% SDS
EI 86% v/v ethanol D7-9 0.0012%SDS
D10-12 0.0016% SDS
D13-15 0.002% SDS

Table A 10: Lab-scale Soxhlet extraction calculations.

Label Mass (g) Impurity
Thimble Feather Glass Final Final Change (wt%)
Wool Container
+Sample
X1 3.212 3.633 0 6.248 3.036 0.597 16.42
X2 3.829 3.335 0 6.763 2.934 0.401 12.01
X3 4.023 4.084 0 7.716 3.693 0.391 9.57
X4 3.229 2.932 0 5.772 2.543 0.388 13.25
X5 4.073 3.080 0 6.625 2.552 0.529 17.16
X6 3.587 3.038 0 6.112 2.525 0.512 16.87
J1 3.44 2.63 0.07 6.29 2.78 -0.15 -5.79
J2 4.40 2.65 0.07 7.11 2.63 0.03 0.97
J3 3.21 2.13 0.06 5.33 2.06 0.07 3.42
J4 3.85 2.61 0.12 7.67 3.69 -1.08 -41.53
J5 4.49 2.40 0.12 6.88 2.27 0.13 5.54
J6 4.04 2.07 0.13 5.92 1.74 0.33 15.73
EA 3.737 3.681 0.131 6.634 2.766 0.915 24.86
EB 4.109 2.965 0.184 6.943 2.651 0.315 10.61
EC 3.516 3.084 0.143 6.229 2.570 0.514 16.65

101
Label Mass (g) Impurity
Thimble Feather Glass Final Final Change (wt%)
Wool Container
+Sample
ED 3.045 2.702 0.215 5.620 2.575 0.127 4.69
EE 3.087 2.760 0.233 5.776 2.689 0.071 2.56
EF 3.329 2.394 0.295 5.740 2.412 -0.018 -0.74
EG 3.397 2.449 0.313 6.349 2.953 -0.504 -20.56
EH 3.338 2.713 0.144 5.934 2.596 0.117 4.31
EI 3.630 2.933 0.134 5.902 2.273 0.660 22.50
HA 3.199 3.091 0.157 6.217 2.861 0.230 7.44
HB 3.816 3.058 0.189 6.857 2.851 0.207 6.76
HC 3.267 3.079 0.197 6.194 2.730 0.349 11.33
DA 3.316 2.888 0.093 5.882 2.473 0.415 14.37
DB 3.518 2.919 0.134 6.140 2.488 0.431 14.76
DC 3.495 1.821 1.250 6.146 1.404 0.417 22.90
DD 3.515 2.873 0.247 6.146 2.384 0.489 17.03
DE 3.675 3.241 0.094 7.076 3.307 -0.066 -2.04
E1 3.155 1.823 0.154 4.884 1.575 0.248 13.60
E2 3.309 1.013 0.120 4.299 0.870 0.143 14.12
E3 3.985 1.340 0.883 6.004 1.136 0.204 15.22
E5 3.523 1.860 0.179 5.346 1.644 0.216 11.61
E6 3.346 1.045 0.753 5.011 0.912 0.133 12.73
E4 3.512 2.010 0.160 5.439 1.767 0.243 12.09
E7 3.349 0.655 0.092 4.030 0.589 0.066 10.08
E8 4.013 0.515 0.118 4.590 0.459 0.056 10.87
E9 3.430 1.373 0.331 4.979 1.218 0.155 11.29
E12 3.701 1.630 0.428 5.611 1.482 0.148 9.08
E10 3.311 0.365 0.090 3.728 0.327 0.038 10.41
E11 3.301 1.750 0.172 5.051 1.578 0.172 9.83
E13 3.060 0.508 0.197 3.717 0.460 0.048 9.45
E14 3.202 1.080 0.128 4.312 0.982 0.098 9.07
E15 3.786 2.475 0.820 6.826 2.220 0.255 10.30
H1 3.199 1.830 0.154 4.945 1.592 0.238 13.0
H2 3.759 1.120 0.110 4.853 0.984 0.136 12.1
H3 4.305 2.767 0.154 6.858 2.399 0.368 13.3
H4 3.607 1.744 0.152 5.330 1.571 0.173 9.9
H5 3.511 1.208 0.190 4.800 1.099 0.109 9.0
H6 3.023 1.979 0.178 5.000 1.799 0.180 9.1
H7 3.379 2.868 0.126 6.108 2.603 0.265 9.2
H8 3.734 1.112 0.166 4.897 0.997 0.115 10.3
H9 4.132 2.893 0.164 6.919 2.623 0.270 9.3
H10 3.918 0.793 0.114 4.753 0.721 0.072 9.1
H11 3.892 2.696 0.159 6.504 2.453 0.243 9.0
H12 3.493 2.584 0.061 5.924 2.370 0.214 8.3
H13 3.766 1.130 0.163 4.957 1.028 0.102 9.0
H14 3.988 0.923 0.107 4.944 0.849 0.074 8.0

102
Label Mass (g) Impurity
Thimble Feather Glass Final Final Change (wt%)
Wool Container
+Sample
H15 3.584 2.764 0.136 6.264 2.544 0.220 8.0
D1 3.259 0.497 0.054 3.717 0.404 0.093 18.7
D2 3.919 1.567 0.058 5.235 1.258 0.309 19.7
D3 3.542 0.962 0.097 4.422 0.783 0.179 18.6
D4 4.404 1.434 0.010 5.613 1.199 0.235 16.4
D5 3.539 1.981 0.063 5.262 1.660 0.321 16.2
D6 3.978 2.101 0.096 5.824 1.750 0.351 16.7
D7 3.344 1.158 0.016 4.350 0.990 0.168 14.5
D8 3.551 1.677 0.081 5.054 1.422 0.255 15.2
D9 3.277 2.998 0.012 5.855 2.566 0.432 14.4
D10 3.606 0.690 0.032 4.239 0.601 0.089 12.9
D11 3.106 2.959 0.086 5.760 2.568 0.391 13.2
D12 3.836 1.207 0.060 4.924 1.028 0.179 14.8
D13 4.052 1.210 0.043 5.134 1.039 0.171 14.1
D14 3.251 1.956 0.001 4.922 1.670 0.286 14.6
D15 3.245 0.880 0.051 4.054 0.758 0.122 13.9

103
Appendix 4: Large-scale Treatment
Table A 11: Lamort pulper specification.
Part: Motor
NL351365015/5 3 phase motor
50 Hz type MBT 132 SA 4/8
3.7/2.2 kW - hp
1400/700 rpm C|F
400 VYY 7.5 A
400 V 7.2A
Cos 92/0.62 IP 55 IEC 34:1
Calent/Temp rise Class B
Part: 2-speed Variable Speed Drive
Model PDL Microdrive 3
Part: Variable Speed Drive
Model Altivar-21
U (V ~) 380/480 3 380/480 3
F (Hz) 50/60 0.5/1000
I (A) 66 66
U (V ~) 460/480 3 460 3
F (Hz) 50 0.5/1000
I (A) 56 FLA 52
Short circuit withstand 5000 A, 480 V when fitted with fuse

Table A 12: Centrifugal pump technical information.

Variable Value Notes
Pump Scanpump Made in Sweden
Bearing 3 Bearing assembly size
Ds 0.10 m Suction size
Dd 0.11 m Discharge size
Di 0.37 m Impeller size
Phase 3~
F 50/60 Hz Frequency
P 22/25 kW Power
N 1475/1775 rpm Rotational speed
Cos 0.81 Power Factor
Q 250 m3 h-1 4167 LPM
H 35 m
N 1470 rpm
VY 380 420/440 480 VY 44/44 A
V 220 240/250 280 V 76/76 A
Control Altivar 61 variable speed control ATV61HD36N4S337
P 30 kW Motor power
Dimension 240 x 550 x 266 mm Size for model IP20 S7A

104
Table A 13: Comminution enthalpy calculations.
Time Temperature Enthalpy hf T Q h
(h) (oC) (kJ kg-1) (kW) (kJ kg-1)
Experiment 0 14.7 61.67
1 1 37.1 155.39 22.4 2600.89 93.71
2 45.5 190.53 8.4 975.33 35.14
3 54.4 227.76 8.9 1033.39 37.23
Experiment 0 13.3 55.82
2 1 42.5 177.98 29.2 3390.44 122.16
2 54.4 227.76 11.9 1381.72 49.79
3 63.4 265.42 9 1045.00 37.65
4 67.6 282.99 4.2 487.67 17.57

1.E-03 experiment 1
1.E-03 experiemnt 2

8.E-04
Viscosity
6.E-04
(kg/m.s)
4.E-04

2.E-04

0.E+00
0 1 2 3 4
Time (h)

Figure A 5: Bulk slurry viscosity during comminution.

105
 1200

1000

800

Re 600
experiment 1
400 experiemnt 2

200

0
0 1 2 3 4
Time (h)

Figure A 6: Slurry Reynolds number during comminution.

50 experiment 1
45
experiemnt 2
40
35
HL, 30
fittings 25
(m) 20
15
10
5
0
0.00 1.00 2.00 3.00 4.00
Time (h)

Figure A 7: Comminution system head loss due to fittings.

106
Figure A 8: Process flow diagram.

107