Nanostructured Biopolymeric Materials - Diva1286485/FULLTEXT01.pdf · I. Flow-Assisted Organization...
Transcript of Nanostructured Biopolymeric Materials - Diva1286485/FULLTEXT01.pdf · I. Flow-Assisted Organization...
Nanostructured Biopolymeric Materials
Hydrodynamic Assembly, Mechanical Properties and Bio-
Functionalities
NITESH MITTAL
DOCTORAL THESIS IN ENGINEERING MECHANICS
STOCKHOLM, SWEDEN 2019
Nanostructured Biopolymeric Materials
Hydrodynamic Assembly, Mechanical Properties and Bio-
Functionalities
Nitesh Mittal
Doctoral Thesis
Department of Mechanics
School of Engineering Sciences
Wallenberg Wood Science Center
KTH Royal Institute of Technology
Stockholm, 2019
Principal advisor
Docent Daniel Söderberg
Co-advisor
Professor Fredrik Lundell
Copyright © Nitesh Mittal, Stockholm, 2019
All Rights Reserved. No part of this thesis may be reproduced by any means
without permission of the author.
Cover Image © 2018 American Chemical Society
Paper I © 2018 American Chemical Society
Paper II © 2018 American Chemical Society
Paper III © 2019 American Chemical Society
Paper IV © 2017 American Chemical Society
Paper V © 2017 National Academy of Sciences USA
ISBN: 978-91-7873-088-9
TRITA-SCI-FOU 2019:5
AKADEMISK AVHANDLING
Som med tillstånd av Kungliga Tekniska Högskolan i Stockholm
framlägges till offentlig granskning för avläggande av teknologie
doktorsexamen fredag den 15e mars 2019, kl. 10.00 i sal F3,
Lindstedtsvägen 26, KTH, Stockholm. Avhandlingen försvaras på
engelska.
Opponent: Professor Chinedum Osuji, University of Pennsylvania, USA.
To my parents
मरी माता और पिता को समपिित
“Two things are infinite: the universe and human stupidity;
and I’m not sure about the universe”
−Albert Einstein
ABSTRACT
The need for high-end multifunctional materials from renewable resources has
evolved given a rapidly increasing population and accompanying environmental
concerns. Scalable assembly methods are and will be imperative in designing high-
performance environmentally friendly materials, requiring new processes allowing
control on all hierarchical levels. In this thesis, engineering concepts for
manipulation of nanoscale components from biopolymeric resources have been
applied to achieve extraordinary macroscale performance. The route chosen has been
fluid-phase assembly as it is one of the most promising methods for producing large,
ordered structures from nanoscale objects.
The thesis has three main parts; assembly of cellulose nanofibrils (CNFs) and
fundamentals associated with the processing technique, the combination of CNFs
with silk fusion proteins and finally the assembly of amyloid-like protein nanofibrils
(PNFs). In the CNFs assembly part, we have pursued the challenge of transferring
the full potential of CNFs to macroscale materials. CNFs are the most abundant
structural elements in biological systems and have impressively high strength and
stiffness, yet natural or man-made cellulose composites are much weaker than the
CNFs. We fabricated nanocellulose fibers in pursuit of maximal mechanical
performance by hydrodynamically controlling the structural ordering of nanofibrils,
resulting in continuous fibers with mechanical properties higher than any natural or
man-made macroscale biopolymeric material (Young’s modulus 86 GPa and a
tensile strength 1.57 GPa). As the hydrodynamic assembly process is largely
dependent on fundamental phenomenon controlling rotational and translational
diffusion, we have applied a novel methodology based on birefringence allowing
time-resolved in-situ investigations of diffusion and network dynamics of nanofibrils
including effects of anisotropic orientation distributions.
Genetic engineering enables the synthesis of bioengineered silk fusion proteins that
can serve as a foundation of new biomaterials. However, silk proteins are difficult
to process and cannot be obtained in large quantities from spiders. By combining
CNFs with recombinant spider silk proteins (RSPs) we have fabricated strong, tough
and bioactive nanocomposites. We demonstrate how small amounts of silk fusion
proteins added to CNFs give advanced bio-functionalities unattainable to wood-
based CNFs alone. Finally, flow-assisted assembly is applied to fabricate a material
from 100% non-crystalline protein building blocks with whey protein, a mixture with
β-lactoglobulin as the main component, which self-assemble into amyloid-like PNFs
stabilized by hydrogen bonds. We show how conditions during the fibrillation
process affect properties and morphology of the PNFs. Furthermore, we compare the
assembly of whey PNFs of distinct morphologies and show that PNFs can be
assembled into strong microfibers without the addition of plasticizers or crosslinkers.
Keywords: cellulose nanofibrils, recombinant spider silk proteins, whey protein
nanofibrils, fiber, polarized optical microscopy, orientation, diffusion, extensional
flow, tensile mechanical properties, bio-functionalities
SAMMANFATTNING
Behovet av avancerade multifunktionella material från förnyelsebara råvaror ökar
med en snabbt växande befolkning i världen och behovet av att värna om vår miljö.
Skalbara framställningsprocesser är och kommer att vara avgörande för
utformningen av dessa hållbara och miljövänliga material med hög prestanda, vilket
kräver nya processer som möjliggör kontroll av materialens struktur på alla
hierarkiska nivåer. I denna avhandling har ny teknologi för kontrollerad
sammanfogning av nanokomponenter baserade på biopolymerer tillämpats för att
uppnå extraordinära egenskaper på makroskopisk nivå. Den valda teknologin är
baserad på strömningsmekanisk sammanfogning, vilket är en lovande metod för att
kontinuerligt framställa makroskopiska strukturerade material från
nanokomponenter.
Avhandlingen består av tre huvudspår; strömningsmekanisk sammanfogning av
nanocellulosa (här syftar vi främst till cellulosananofibriller, CNF) och därtill
associerade grundläggande frågeställningar kring processen kopplat till detta;
framställning av material baserat på kombinationen av nanocellulosa med
silkesproteiner och slutligen framställning av material bestående av amyloidliknande
proteinnanofibriller (PNF). När det gäller sammanfogning av nanocellulosa har
målet vara att förstå de förhållanden under vilka framställningen av makroskopiska
material av CNF resulterar i materialegenskaper som förmår utnyttja de mekaniska
egenskaperna hos nanokomponenterna. Nanocellulosa är det mest förekommande
strukturella bio-baserade elementet på jorden och har en imponerande hög hållfasthet
och styvhet, och naturliga och framställda cellulosamaterial har hittills varit mycket
svagare än nanocellulosa. Genom att strömningsmekaniskt påverka hur
nanocellulosan bildar nanostrukturen hos kontinuerliga fiberr, med målet om
maximal mekanisk prestanda, har vi framställt fiberr som är starkare och styvare än
något annat naturligt eller konstgjort biomaterial (elasticitetsmodul 86 GPa och
draghållfasthet 1.57 GPa). Eftersom den strömningsmekaniska
framställningsprocessen i stor utsträckning är beroende av grundläggande fenomen
som rotationsdiffusion och nätverksgenerering av nanopartiklar, har vi även
tillämpat en ny metod baserad på dubbelbrytning som möjliggör tidsbesparande in-
situ karakterisering av rotationsdiffusion och nätverksdynamik för nanofibrillära
komponenter, inklusive effekterna på anisotropa orienteringsfördelning.
Genteknik möjliggör syntes av silkesproteiner som kan tjäna som grund för nya
biomaterial. Silkesproteiner är emellertid svåra att hanteras och kan bara erhållas i
små mängder från spindlar. Genom att kombinera nanocellulosa med rekombinant
spindelsilkeprotein har vi tillverkat starka, tåliga och bioaktiva nanokompositer. Vi
visar hur små mängder silkesfusionsproteiner kan kombineras med nanocellulosa för
att ge avancerad biofunktionalitet som inte kan uppnås med nanocellulosamaterial.
Slutligen har den strömningsbaserade framställningsprocessen använts att tillverka
material från 100% icke-kristallina byggstenar av vassleprotein. Dessa byggstenar
består av en blandning med p-laktoglobulin som huvudkomponent, vilken
självgenererar amyloidliknande proteinnanofibriller stabiliserade av vätebindningar.
Vi visar hur förhållandena under fibrilleringsprocessen påverkar egenskaper och
morfologin hos proteinnanofibrillerna. Slutligen framställs material av
proteinnanofibriller med olika morfologier, där vi visar hur dessa kan sammanfogas
i starka kontinuerliga fiberr utan tillsats av mjukgörare eller tvärbindande
komponenter.
Nyckelord: cellulosananofibriller, spindelsilkeproteiner, proteinnanofibriller,
fiberr, polariserad optisk mikroskopi, fiberorientering, rotationsdiffusion, töjflöden,
mekaniska egenskaper, biofunktionalitet.
LIST OF PAPERS
This thesis is a summary of the following five appended papers:
I. Multiscale Control of Nanocellulose Assembly: Transferring Remarkable
Nanoscale Fibril Mechanics to Macroscale Fibers.
Nitesh Mittal, Farhan Ansari, Krishne Gowda V., Christophe Brouzet, Pan Chen, Per
Tomas Larsson, Stephan V. Roth, Fredrik Lundell, Lars Wågberg, Nicholas A.
Kotov, and L. Daniel Söderberg.
ACS Nano, 2018, 12 (7), 6378-6388.
II. Size-Dependent Orientational Dynamics of Brownian Nanorods.
Christophe Brouzet, Nitesh Mittal, L. Daniel Söderberg and Fredrik Lundell.
ACS Macro Letters, 2018, 7 (8), 1022-1027.
III. Characterizing the Orientational and Network Dynamics of Polydisperse
Nanofibers at the Nanoscale.
Christophe Brouzet, Nitesh Mittal, Fredrik Lundell and L. Daniel Söderberg.
Macromolecules, 2019. (Accepted)
IV. Ultrastrong and Bioactive Nanostructured Bio-Based Composites.
Nitesh Mittal, Ronnie Jansson, Mona Widhe, Tobias Benselfelt, Karl M. O.
Håkansson, Fredrik Lundell, My Hedhammar, and L. Daniel Söderberg.
ACS Nano, 2017, 11 (5), 5148-5159.
V. Flow-Assisted Assembly of Nanostructured Protein Microfibers.
Ayaka Kamada,ǁ Nitesh Mittal,ǁ L. Daniel Söderberg, Tobias Ingverud, Wiebke
Ohm, Stephan V. Roth, Fredrik Lundell, and Christofer Lendel.
Proceedings of the National Academy of Sciences of the United States of America
(PNAS), 2017, 114, 1232-1237. (ǁCo-first author)
Author's contribution to the appended papers:
- Paper I, IV, V. Principal author: participated in planning and designing, performed
substantial set of experiments and part of manuscript preparation.
- Paper II, III. Co-author: participated in planning and designing, part of
experimental work (sample preparation and characterization) and part of manuscript
preparation.
Other publications not included in the thesis
VI. Ion-Mediated Control of Mechanical Properties of Nanocellulose
Filaments.*
Nitesh Mittal, Farhan Ansari, and L. Daniel Söderberg.
Manuscript in Preparation.
VII. Understanding the Mechanistic Behavior of Highly Charged Cellulose
Nanofibers in Aqueous Systems.
Lihong Geng, Nitesh Mittal, Chengbo Zhan, Farhan Ansari, Priyanka R. Sharma,
Xiangfang Peng, Benjamin S. Hsiao, and L. Daniel Söderberg.
Macromolecules, 2018, 51 (4), 1498-1506.
VIII. Dynamic Characterization of Cellulose Nanofibrils in Sheared and
Extended Semi-Dilute Dispersions.*
Tomas Rosén, Nitesh Mittal, Stephan V. Roth, Peng Zhang, L. Daniel Söderberg and
Fredrik Lundell.
Manuscript in Preparation. (Available on arXiv:1801.07558)
IX. Wettability and Humidification Induced Buried Morphological
Rearrangement at Nanoscale in Thin Films.*
Calvin J. Brett, Nitesh Mittal, Wiebke Ohm, Marc Gensch, Lucas P. Kreuzer, Volker
Körstgens, Martin Månsson, Henrich Frielinghaus, Peter Müller-Buschbaum, L.
Daniel Söderberg, and Stephan V. Roth.
Submitted Manuscript.
X. Printing of Supramolecular 3D Networks for Biosensing and Smart Wound
Dressings.* ǁ
Moien Alizadehgiashi, Daniel Pinto Ramos, Nitesh Mittal, L. Daniel Söderberg, and
Eugenia Kumacheva.
Manuscript in Preparation.
ǁThe study has been carried out at University of Toronto, Canada. The author spent
21 weeks in the group of Prof. Eugenia Kumacheva (Polymers, Interfaces and
Materials Science lab) at the Department of Chemistry as an International Visiting
Graduate Student.
*Titles and authors list may differ in the finally accepted and published version of
manuscripts that are presently under preparation/submitted.
Conference Presentations
I. Flow-Assisted Organization of Nanostructured Bio-Based Materials.
Nitesh Mittal, Fredrik Lundell, Lars Wågberg, My Hedhammar, and Daniel
Söderberg.
255th ACS National Meeting & Exposition, New Orleans, LA, USA, 2018.
II. Hydrodynamically Assembled Nanostructured Bio-Based Materials.
Nitesh Mittal, Fredrik Lundell, and Daniel Söderberg.
MRS Fall Meeting & Exhibition, Boston, MA, USA, 2017.
III. Effect of Cellulose Nanofibril Morphology on the Strength and Stiffness of
Macroscopic Filaments.
Nitesh Mittal, Fredrik Lundell, and Daniel Söderberg.
253rd ACS National Meeting & Exposition, San Francisco, CA, USA, 2017.
IV. Flow Based Approach for the Fabrication of Super-Strong Nanocellulose
Filaments- An Attempt to Utilize its Potential.
Nitesh Mittal, Fredrik Lundell, and Daniel Söderberg.
MRS Fall Meeting & Exhibition, Boston, MA, USA, 2016.
V. Elongational Flow Assists with the Assembly of Protein Nanofibrils.
Nitesh Mittal, Ayaka Kamada, Christofer Lendel, Fredrik Lundell, and Daniel
Söderberg.
APS 69th DFD Meeting, Portland, Oregon, USA, 2016.
VI. Hydrodynamic Alignment and Assembly of Carboxymethylated Cellulose
Nanofibrils: Effect of Concentration.
Nitesh Mittal, Fredrik Lundell, and Daniel Söderberg.
MRS Fall Meeting & Exhibition, Boston, MA, USA, 2015.
VII. Hydrodynamic Alignment and Assembly of Nanofibrillated Cellulose in
the Laminar Extensional Flow: Effects of Solidifying Agents.
Nitesh Mittal, Fredrik Lundell, and Daniel Söderberg.
APS 68th Annual DFD Meeting, Boston, MA, USA, 2015.
Abbreviations and Symbols
CNF Cellulose nanofibril or cellulose nanofiber
CNC Cellulose nanocrystal
PNF Protein nanofibril
RSP Recombinant spider protein or recombinant spider silk protein
TEMPO 2,2,6,6-tetramethylpiperidinyl-1-oxyl
IgG Immunoglobulin G
HDF Human dermal fibroblast
HDMEC Human dermal microvascular endothelial cells
AFM Atomic force microscopy
FE-SEM Field-emission scanning electron microscopy
TEM Transmission electron microscopy
FT-IR Fourier-transform infrared spectroscopy
SAXS Small-angle X-ray scattering
WAXS Wide-angle X-ray scattering
POM Polarized optical microscopy
MD Molecular dynamics
RH Relative humidity
wt Weight
Dr Rotary diffusion coefficient
G’ Storage modulus
fc Orientation index
E Young’s modulus
σ , σu Ultimate tensile strength
ε Strain to failure
Nc Crowding number
CONTENTS
1. Introduction…………………………………………………….... 1
1.1 Biological Materials……………………………………………1
1.1.1 Nanocellulose from Wood……………………….………1
1.1.2 Spider Silk……………………………………………….4
1.1.3 Protein Nanofibrils………………………………………8
1.2 Materials Fabrication Strategies…………………………….... 9
1.2.1 Microchannels In-Situ Structural Dynamics……………13
2. Scope and Objectives……………………………………………14
3. Materials and Methods………………………………………….15
4. Nanostructured CNF Fibers…………………………………… 26
4.1 Assembly and In-Situ Orientation Dynamics of CNFs………26
4.2 Mechanical Properties……………………………………….31
4.3 Deformation Mechanism at the Nanoscale…………………. 34
4.4 Mechanical Properties in a Wider Perspective………………36
5. Fundamentals of the Hydrodynamic Assembly Process….…...39
5.1 Double Flow-Focusing Geometry………………………...…39
5.2 Shear vs Elongation…………………………………...……. 40
5.3 Rotary Diffusion and Size Dependency………………….… 41
5.4 Network Formation at the Nanoscale…………………….… 46
6. Nanostructured CNF/Silk Composites…………………………48
6.1 Fabrication of CNF/Silk Nanocomposites…………………...48
6.2 Mechanical Properties of Anisotropic Fibers…………….… 51
6.3 Bio-Functionalities…………………………………………. 54
7. Nanostructured Protein Microfibers…………………………...59
7.1 Assembly and In-Situ Orientation Dynamics of PNFs………60
7.2 Micro- and Molecular Structure in the Gel State…………….65
7.3 Hierarchical Protein Materials and Comparison with Other
Building Blocks……………………………………………...65
8. Conclusions……………………………………………………... 67
9. Future Work and Outlook………………………………….......70
Acknowledgements…………………………………………………. 73
References…………………………………………………………... 74
Appendix……………………………………………………………. 86
INTRODUCTION
1
1. Introduction
Synthesis of novel materials from biomass has substantial environmental and
economic impacts, particularly in present time of global warming, growing
population and depleting petroleum oil reserves. It is, therefore, the need to discover
material sources that can offer a sustainable path forward.1 In this direction,
industries are making attempts to develop “greener” materials; governments are
encouraging research on bioproducts; scientists in academia are searching for eco-
friendly materials. Till a certain extent, bio-based polymeric materials (also
represented here as biopolymeric materials or bio-based materials) from renewable
resources can contribute positively towards solving this complex problem.2
Natural (or biological) materials possess many inspiring properties such as
complexity, miniaturization, hierarchical organization, hybridization, and
malleability.3 These materials are typically formed through the assembly of
repetitive molecular to nano- and micron-sized building blocks into hierarchical
structures, which are stabilized by non-covalent interactions. Few examples of
biological materials with multiscale organization are silk, cellulose, chitin and
collagen.4-6 The highly ordered arrangement of the building blocks in these
biological materials such as CNFs in plants, chitin nanofibrils in animals, collagen
nanofibrils in ligaments and silk fibroins in spider silk and in silkworms7 provides
them with remarkable mechanical properties and flexibility, while also presenting
biological functions that interact with the surrounding environment.8 These features
make biopolymeric building blocks promising candidates for the development of
structures that are strong, lightweight and biodegradable for automobile, energy,
optics and biomedical applications. Lately, researchers have been seeking ideas of
designing novel materials by mimicking the architecture of natural materials which
is challenging from an engineering perspective. However, rational design
approaches, combining theory and the experimental tools, have aided in designing
bio-based materials through learnings from nature. In this direction, recent initiatives
by forest companies to invest in nanocellulose pilot plants, and by startups to upscale
the production of spider/silkworm silk to make threads for prototype garments are
encouraging for commercialization of bio-based materials.9
INTRODUCTION
2
1.1 Biological Materials
1.1.1 Nanocellulose from wood
Wood is a hierarchical architecture of nature that performs the functions of
providing mechanical support and water and nutrients transport through the tree.
Ibiot has a porous structure where the constituents are arranged at different length
scales to provide the mechanical integrity (Figure 1.1). Cellular architecture of wood
contains porous lumen and relatively dense cell wall.10 The cell wall of wood is made
up of a primary layer (P) and three secondary layers (S1, S2 and S3). These cell walls
are attached to each other by middle lamella (ML). Each layer in the cell wall
consists of semi-crystalline cellulose fibers (~40 wt.%) embedded in a matrix of
amorphous hemicelluloses and lignin (~30 wt.% each). The volume fraction and
orientation of cellulose fibers varies in each layer. For instance, S1 layer has random
orientation of cellulose fibers, whereas, the fibers are highly oriented in S2 layer.
Cellulose fibers can also be obtained from plant fibers, bacteria or tunicates.11 In
cellulose fibers, there is a base fundamental reinforcement unit that strengthen the
cellulose-based structures, known as “nanocellulose”. By extracting the
nanocellulose from the cellulose fibers, the defects associated with the hierarchical
structure can be removed.11
Figure 1.1. Hierarchical structure of wood with components at different length
scales. (Reproduced from Ref.12, Copyright 2013 Elsevier)
INTRODUCTION
3
Nanocelluloses are high aspect ratio nanoparticles (Figure 1.2) that can be
classified broadly in two categories (1) CNCs, also known as whiskers, which are
typically 150-250 nm long, rigid and rod-like with high crystallinity and (2) CNFs,
also known as NFC, MFC nanofibrillated cellulose and microfibrillated cellulose,
respectively, which are relatively longer (up to few microns) and more flexible.11-12
The exact dimension of nanocellulose particles depends on the source as well as on
the chemical pretreatment method. Nanocellulose from wood is typically obtained
from a pulp, which has long cellulose fibers with high cellulose content relative to
native wood. This is because the matrix components are partially separated during
the pulping process. The earliest reports on nanocellulose preparation date back to
early 1950s, where sulfuric acid hydrolysis of cellulose fibers was done to obtain
CNCs.13 Acid hydrolysis dissolves the disordered regions of cellulose fibers leaving
the highly ordered domains intact, as they are resistant to degradation from acids.
Sulfuric acid treatment induces surface charges to CNCs, which provides them the
colloidal stability. CNCs from wood can have width of 3-15 nm.11, 14 Extraction and
characteristic features of CNCs prepared by different methods have been reported
elsewhere.11, 15-16
Figure 1.2. TEM images of nanocelluloses.
CNFs are substantially longer than CNCs (Figure 1.2), and contain highly
ordered as well as less ordered regions (kinks).17 CNFs are typically obtained after
the mechanical disintegration of pulp through a high pressure homogenizer or by
sonication after a mild pre-treatment (enzymatic or chemical). Attempts to obtain
CNFs from pulp fibers were initiated in 1980s, where no pre-treatment on the pulp
fibers was done. The obtained CNFs had a broad size distribution high mechanical
energy was required to liberate the fibrils from the pulp fibers.18-19 Thereafter several
INTRODUCTION
4
pretreatments methods to assist the disintegration process were developed. Chemical
pre-treatment typically induces the charges on the surface of pulp fibers which
facilitates disintegration due to electrostatic repulsion. One widely used chemical
pre-treatment method is TEMPO (2,2,6,6-Tetramethylpiperidin-1-yl)oxyl mediated
oxidation that converts primary hydroxyl groups to carboxylates (negative
charges).20 Other methods include carboxymethylation,21 and cationic
functionalization.22-23 CNFs with surface charge density of up to 1500 µequiv g-1
have been reported with TEMPO mediated oxidation method.20 Due to the relatively
good colloidal stability and ease of processing, TEMPO-oxidized and
carboxymethylated CNFs are used in this thesis.
Molar mass of cellulose is reduced significantly during pretreatment and after
disintegration.20 CNFs of the order of 0.4 µm in length and 2-5 nm in width, may be
obtained when oxidized the pulp to a high surface charge.24 Crystallinity of wood
CNFs is >30%, depending on the type of wood and chemical pre-treatment method.
Availability of abundance of hydroxyl groups provides a functional surface that
plays a key role in determining the properties of CNF based nanocomposites. CNFs
can also form physically entangled network with strong interactions due to the
hydrogen bonds and Van der Waals forces.25
1.1.2 Spider Silk
Spider silk has impressive mechanical properties (strength and elasticity)
despite being spun at ambient temperatures and pressures with water as solvent
(Figure 1.3). The spider makes this fascinating material by processing the main
protein constituents along with auxiliary compounds and by controlling the
folding and crystallization simultaneously, to create a hierarchical structure with
ordered and repetitive arrangement of the building blocks.26 Spider does the
spinning of its fibers using minimal forces as the ‘dope’ (the constituent material
from which silk is spun) is liquid crystalline. Spiders can produce silks with
different characteristics. In materials perspective, spider dragline silk that comes
from the major ampullate gland of spiders, is one of the toughest bio-based
material with strength comparable to synthetic Kevlar and higher than that of
tendons.27 Apart from the remarkable mechanical properties, spider silk has the
ability to control bleeding and promote wound healing.28
INTRODUCTION
5
Figure 1.3. A schematic of the silk production system of a spider. The secondary
structure adopted by the spider proteins (spidroins) along the gland and duct with the
changes in pH, salt concentrations and water contents. (Reproduced from Ref.28,
Copyright 2011 Springer Publishing Group)
Dragline silk is mainly composed of two types of proteins (also refer as
spidroins), major ampullate spidroin (MaSp1 and MaSp2). Architecture of MaSp is
composed of alanine that provides the strength to fibers by the formation of β-sheets
and the Glycine-rich unit, provides elasticity with the presence of amorphous
regions. These repetitive sequences are covered by non-repetitive N-terminal and C-
terminal domains.29-30 The non-repetitive domains assist with the storage and
assembly of spidroins that are stored in the spiders’ glands as liquid crystalline dope.
The conformation of proteins inside the dope is primarily α-helical. Transition of α-
helix to β-sheets occurs inside the spinning duct, where the dope undergoes
conformational changes, in the presence of ions, change in pH, water removal and
the shear forces.31
It is difficult to farm spider because of their territorial and cannibalistic
behavior.32 From an industrial perspective, the amount of silk that can be harvested
directly from the spiders is not sufficient. Recombinant production of spider silk
proteins is considered as an alternate route to overcome this challenge.33 Tobacco,
potatoes, the yeast and E.coli bacteria have been utilized as hosts for synthetic
production of RSPs by the use of genetic engineering.32 However, E.coli is most
widely used host due to ease of genetic manipulation, short reproduction times, and
low-production cost.
INTRODUCTION
6
Figure 1.4. Schematic of IgG-fluorophore binding to Z-silk.34 (Image credit: Ronnie
Jansson)
Functionalization of silk is a valuable route forward to find the applications in
the fields of biotechnology and medicine.35 Several strategies have been applied to
functionalize RSPs to produce bioactive silk-based materials.36-37 The methods of
silk functionalization rely either on non-covalent or covalent interactions (Figure
1.4). Non-covalent functionalization techniques count typically on ionic interactions.
However, physical entrapment of functionalizing molecule is also possible. In
contrast, strong covalent chemical coupling between the material and the
functionalizing molecule is the basis of covalent functionalization. Bonds are stable
over time in the latter case.
INTRODUCTION
7
Figure 1.5. Spider silk proteins fibers. A, B, C are optical microscope images and D
is a SEM image. Scale bars are 2 mm in A and B, 100 µm in C and 10 µm in D.
(Reproduced from ref.28, Copyright 2011 Springer Publishing Group)
Recombinant spider silk scaffolds (Figure 1.5) functionalized with cell-
binding motifs have shown positive effects on cell characteristic features such as cell
viability, proliferation and viability.37-38 Cell binding motifs that are present in the
extracellular matrix (ECM) proteins (e.g., laminin, fibronectin, collagen,
vitronectin), mediate interactions between cells and ECM through recognition by
cell-surface integrins.39-40 To mimic the ECM environment, bioactive molecules like
growth factors and cell-binding peptides have been combined with silk building
blocks to create materials that have functional cell scaffolds with cell interacting
properties. Another example of silk for tissue engineering is for the cornea tissue
formation by the use of optically transparent silk films.41 Biodegradability and the
adjustment of drug release profile also make silk building blocks, a suitable carrier
for drug delivery.42 However, to facilitate wider use of the bio-functionalities
observed with the silk-based materials, there is a strong incentive to reduce the high
cost of materials based on RSPs. One potential route to achieve this is to combine
the silk proteins with a compatible building block that can be produced in larger
volumes and is available at lower cost.
INTRODUCTION
8
1.1.3 Protein Nanofibrils
Proteinaceous materials with the core structure of amyloids have recently
gained much attention and efforts are now turning more towards artificial self-
assembly of such structures for diverse set of applications. Utilizing the rich
chemistry and unique properties of proteins is however, not straightforward and thus
the generation of tailored protein materials is challenging due to understanding and
controlling protein self-assembly at multiple length-scales.43 A fruitful research
outcome to solve this challenge is the fibrillar self-assembly of proteins.44 Such near
one-dimensional protein nanostructures that are mostly homomolecular have been
used as load-bearing components in functional coatings, in catalytic scaffolds and
epigenetic information storage.45-47 Hierarchical artificial proteinaceous materials
for a variety of biological functions have also been developed.48-49
Self-assembly of globular proteins under denaturing conditions leads to the
formation of nanofibrils with several nanometers of thickness and micrometers of
length. Dense network of hydrogen bonds is the characteristic core of nanofibrils and
that provides a very high Young’s modulus to these nanofibrillar structures due to
strong intermolecular backbone-backbone interactions between polypeptide chains
within the nanofibrils.50 These PNFs share certain structural characteristics similar
with the pathogenic amyloid aggregations formed in vivo, hence they often called as
“amyloid-like”.51 The high aspect ratio and the surface charges help these nanofibrils
to form entangled networks or gels, which assists with the fabrication of hierarchical
structures.
Figure 1.6. AFM images of PNFs. (Image credit: Christofer Lendel)
INTRODUCTION
9
Proteins obtained from milk, soybean and eggs can self-assemble into
amyloid-like fibrils upon incubation at low pH far from protein’s isoelectric point
and high temperature conditions.52 An often used model protein for self-assembly
into PNFs is Bovine β-lactoglobulin because of its well-characterized structure, and
easiness of availability.51 Morphologies (Figure 1.6) and mechanical properties of
PNFs can be tuned in the presence of salts and also by varying the protein monomer
concentrations and ionic conditions.52 Thus, amyloid-like fibrils assembly is a
general form of nanoscale ordering of polypeptide molecules with impressive
mechanical properties that has a huge potential in the creation of future
proteinaceous functional materials.
1.2 Materials Fabrication Strategies
Multidimensional manufacturing strategies can be applied to mimic the
structure and properties of natural materials by assembling the building blocks. In
this direction, artificial spinning techniques are applied to regulate the orientation of
nanofibrils in 1D, where the in planes organization of fibrils can be achieved through
chemical modification,53 by applying magnetic54 or electric-fields55 and also through
spray-assisted layered assembly.56 In addition, biopolymer nanofibrils can also be
combined with inorganic building blocks such as nanoclays, graphene sheets etc. to
fabricate 2D materials (nanopapers and membranes) to achieve advanced optical and
electrical properties with superior biodegradability.57-58 These materials show
synergistic properties owing to the controlled arrangement of individual
components. The fabrication of 3D materials remains challenging, however,
advances in 3D printing technologies managed to produce biopolymer nanofibril
materials with ordered arrangements.59-61
Wet Spinning: Wet spinning is a widely used technique to control the
organization of nanofibrils in macro-sized fibers. In this process, a liquid jet
(typically a suspension) is extruded through a confined-space or a channel into a
coagulation bath, where the jet solidifies into a fiber after precipitation. The method
of wet spinning has been applied to suspensions or gels of biopolymer nanofibrils,
including nanocellulose,62 chitin63 and silk,64 and spun into oriented fibers with
diameters ranging from micrometers to millimeters and lengths up to several meters.
The alignment of the nanofibrils in the fibers is controlled by the flow conditions
and assembly is induced through the solvent exchange during or after extrusion
through the ducts, where the duct dimensions also play an important role on the
organization of the nanofibrils.62 The interactions between the nanofibrils inside the
duct can be strong as well as weak, depending on the network formed by the
nanofibrils based on the number of contact points, which in typically dependent on
nanofibrils length and concentration in the suspensions or gels.65-66
INTRODUCTION
10
Dry Spinning: Dry spinning technique has also been applied to silk fibroins to
mimic the properties of natural silk by spiders, where the silk dope is converted to a
fiber upon the evaporation of the volatile solvent.67 This technique has been used for
CNFs as well, to generate fibers with oriented nanofibrils.68 Post-processing steps
are typically not required in the dry spinning techniques. One major advantage of
dry spinning technique is that it can generate polymorphic arrangement of
nanofibrils within a single fiber, which is typically not the case with the wet spinning
techniques, where fibrils are oriented along the fiber axis.67 However, dry spinning
requires much higher dope concentrations to spun the fibers to provide a strong
enough network of nanofibrils to avoid the collapse.4
Wet and dry spinning are the efficient techniques to fabricate fibers with aligned
nanofibrils with improved mechanical properties. However, the mechanical
properties of these fibers are much inferior compared to the natural fibers with the
same building blocks. The highest obtained mechanical properties from the dry-spun
CNF fibers is 12.6 GPa of Young’s modulus and 222 MPa of tensile strength.62, 68 A
specific elastic modulus and strength of CNF-based fibers fabricated by wet spinning
were ~22 GPa and ~206 MPa,4 respectively, which are much less than the theoretical
and experimentally determined values characteristic of the individual CNF. Natural
wood pulp fiber with high orientation of nanoscale crystalline planes, free from
damage and natural imperfections, has a highest reported strength of ∼1600 MPa
(stiffness unknown).24, 69 Similarly, the fibers fabricated with the wet spinning of silk
fibroins or proteins are much inferior compared to the mechanical properties of
natural spider silk.70-71 The big gap in mechanical properties4 is due to the lack of
suitable processes and devices that can tailor the nanofibrils organization and thus
improve the mechanical performance.
Microfluidic Spinning: Microfluidic spinning is an advanced way to perform the
wet spinning. It is a promising technique to control the assembly of
nanoparticles/nanofibrils for the fabrication of load-bearing structures such as fibers
with improved mechanical properties (Figure 1.7). An intrinsic advantage of using
microfluidic tools is the ability to control the motion of nanoparticles in suspensions
by optimizing the processing parameters under the influence of flow-fields (shear or
elongation). These tools also provide opportunities to in-situ control the
physicochemical conditions such as pH, salt gradients etc. to induce controlled self-
assembly.4 Although microfluidic tools have demonstrated the ability to produce
fibers with enhanced mechanical properties,24, 34, 66 challenges associated to the
perfect nanostructuring of nanofibrils to realize the full potential of nanoscale
building blocks in macrostructures and also related of the scale-up of production still
need to be addressed.
INTRODUCTION
11
Figure 1.7. Different microfluidic platforms used to fabricate fibers.72 (A) A triple-
orifice spinneret with five inlets connecting to a single outlet . (B) Pulled glass
capillaries based double-coaxial microfluidic device. (C) Glass capillaries integrated
in Polydimethysiloxane (PDMS). (D) A PDMS device fabricated using soft
lithography having rectangular channels. (E) A PDMS device with cylindrical
channels. ECM- extracellular matrix, CAC- collagen-alginate composite, WSC-
water-soluble chitosan, NaA- sodium alginate, PGA- propylene glycol alginate.
(Reproduced from Ref.72 with permission from the Royal Society of Chemistry).
INTRODUCTION
12
Figure 1.8. (A) Schematic of SAXS setup and microfluidic channel. Microfluidic
channel is placed between the source of X-ray beam and the detector. Scattered
photons are recorded by the detector in the form of 2D scattering patterns.
(Reproduced from Ref.73 with permission from the Royal Society of Chemistry) (B)
Quantification of in-situ orientation of nanofibrils using SAXS. (Reproduced from
Ref.74, Copyright 2014 Nature Publishing Group)
INTRODUCTION
13
1.2.1 Microchannels In-Situ Structural Dynamics
Obtaining in-situ structural information of anisotropic nanoparticles under
dynamic flow conditions is challenging. The macroscopic behavior of hierarchical
materials fabricated from such nanoparticles via hydrodynamic approaches, largely
depends on the nanoscale network and interparticle contacts in the microchannels.
These features often associate with changes in the structure of individual phases due
to the flow-fields, temperature, confinement, concentration and channel geometry.
Quantification of in-situ structural dynamics is not an easy task as characterization
techniques which allow spatial and time-resolved measurements are limited. This
largely limits our fundamental understandings on the motion and the behavior of
anisotropic macromolecules.75 Therefore, in-depth characterization of nanoparticle
dynamics and a thorough understanding of the nanoscale physics will pave the way
for flow-based fabrication processes.
Existing characterization techniques such as rheology measurements, and
electron microscopy,76 are predominantly limited to observing static systems (
solid/liquid systems with fixed volume). Other technique such as dynamic light
scattering is hard to couple with microfluidic channels and have problems with the
light absorption by the dispersants and is unable to characterize the polydisperse
samples, which is typically the case with building blocks of biological origin.14
Continuous monitoring of structural changes at the nanoscale in out of equilibrium
microfluidic systems is feasible through SAXS (Figure 1.8). Numerous anisotropic
nanoparticles such as amyloids, CNFs in microfluidic channels have recently been
investigated using SAXS.73-74 While SAXS sources offer high brilliance (1012–14
photons s-1 at the sample) and small beam sizes (100 µm to sub-µm),77 one big
problem is the availability of high intensity X-rays that is primarily limited to
synchrotron sources. New lab-scale methods and tools to characterize nanoparticle
dynamics under flow will thus contribute critically to the development of knowledge
and technology for controlled assembly of colloidal systems.
OBJECTIVES
14
2. Scope and Objectives
Biopolymeric building blocks constitute the basic architectural element of many
biological (or natural) materials that fulfil structural functions. Tightly controlled
and well-ordered arrangements of these building blocks at multiple length scales -
from molecular to nano- and micron level, provides exceptional properties to these
biological materials. However, the complexity of artificially designing new
hierarchical materials by realizing the potential of these natural building blocks at
length scales most suitable to human technologies is a big challenge. The basic
knowledge required to meet this challenge is incomplete in many dimensions, in
particular at the nanoscale. This emphasizes the need for the development of novel
processing strategies and the fundamental understanding thereof, to address multiple
physical and mechanical factors that impose tremendous constraints on designing
nanostructured materials with improved performance.
The scientific objective of this thesis is primarily related to fabrication of
nanostructured biopolymeric materials (mainly fibers) with high mechanical
properties, fundamental understanding of process parameters that affect the materials
characteristics and establishing structure-property relationships in the engineered
materials. By utilizing microfluidic tools, the goal is to improve the understanding
of organization and assembly of nanoscale biopolymeric building blocks in
hierarchical materials. In addition, the possibilities to combine basic knowledge from
different scientific fields such as fluid physics, colloidal chemistry and genetic
engineering to design new high-performance structural and functional biopolymeric
materials are explored. The cases investigated in the present thesis provide
encouraging insights with respect to the general potential of biopolymers.
EXPERIMENTAL SECTION
15
3. Materials and Methods
Detailed information on the chemicals, materials and methods are available in
the appended papers (I to V) and associated supporting information. This chapter
briefly summarizes material preparation and characterization techniques.
Preparation of CNFs
TEMPO-oxidized CNFs were used for the studies reported in paper I to III and
carboxymethylated CNF (commercially available) for the study in paper IV (Figure
3.1). For the CNF in paper IV, 1 wt.% gel was kindly provided by Innventia AB
(now RISE Bioeconomy), Stockholm, Sweden. The CNF was derived after the
defibrillation of a dissolving grade pulp that had been carboxymethylated to a charge
density of 600 ± 50 μequiv g−1, according to a procedure reported before.21
For the studies reported in papers I to III, chemically bleached wood fibers (a mixture
of 60% Norwegian spruce and 40% Scots pine, provided by Domsjo Fabriker AB,
Sweden) were used to prepare CNF suspensions. The relative glucose content of the
pulp was >96%. Pulp fibers were chemically treated with a 2,2,6,6-
tetramethylpiperidinyl-1-oxyl (TEMPO)-mediated oxidation reactions as reported in
detail in paper I. Oxidized pulp fibers were thoroughly washed with DI water.
Suspensions of the chemically pretreated pulp fibers were then passed through a
high-pressure homogenizer. Gel-like CNF suspensions, with a concentration of >5
g L−1 were obtained after homogenization. suspensions were diluted by adding DI
water followed by mixing and sonication. Unfibrillated precipitates were removed
by centrifugation at 5000 rpm for 60 min. The supernatant fluids were then used for
the studies. Dry content was determined using gravimetric analysis.
EXPERIMENTAL SECTION
16
Figure 3.1. TEM images of TEMPO-oxidized CNFs of different surface charge
densities in µequiv g-1 (CNF-380, 550, 820, 980 and 1360) and carboxymethylated
CNFs obtained from Innventia (now RISE Bioeconomy). Scale bars are 500 nm.
Preparation of PNFs
Whey protein isolate (WPI) was obtained from Arla Food Ingredients. A ∼100 g/L
concentration of WPI was dissolved in hydrochloric acid (HCl) at pH 1 and the
mixture was dialyzed with HCl at pH 2 for 24 h at 23°C temperature using a 6-8 kDa
molecular weight membrane. The solution was then diluted to different
EXPERIMENTAL SECTION
17
concentrations to obtain nanofibrils of different morphologies (40 g/L for straight
fibrils and 75–80 g/L for the curvy ones). PNFs were obtained after the incubation
of protein monomers at 90 °C for a period of 3 days. PNF suspensions were then
dialyzed using a 100-kDa molecular weight membrane for 7 days to remove the
unfibrillated proteins. The final concentrations were measured by UV-Vis and
gravimetric analysis of lyophilized materials.
Characterization of cellulose- and protein nanofibrils
Cellulose nanofibrils (CNFs): Length and height of CNFs used in paper I to IV are
measured using TEM and AFM (Figures 3.2 and 3.3). The lengths of TEMPO-
oxidized CNF (paper I to III) and carboxymethylated CNF (paper IV) (∼250 for each
charge density) are measured using TEM (JEOL JEM-1400 TEM) at an accelerating
voltage of 120 kV. The images were acquired using a ruby camera following
“systematic, uniform, random” rule to avoid bias. Droplets of CNF suspensions were
deposited on carbon-coated copper grids that were stained with 2% uranyl acetate
solution. The height of CNFs (∼150 for each charge density) were measured using
AFM (MultiMode 8, Bruker, Santa Barbara, CA, USA). Silicon wafers were
oxidized to form silica layer. The wafers were then washed with ethanol and milli-
Q followed by drying and plasma treatment to make the surface hydrophilic. Wafers
were then dipped in the CNF suspensions and air dried at room temperature.
A protocol based on conductometric titration78 was used to determine the carboxylate
contents of TEMPO-oxidized CNFs (paper I to III). 100 ml suspensions of 0.1 wt.%
CNFs were used, and the pH of solution was adjusted to 2.5 with HCl. 0.01 M
standardized NaOH was used to titrate the suspensions and pH was adjusted to 11,
and the conductivity was monitored with a benchtop meter (FE20 FiveEasy, Mettler-
Toledo).
EXPERIMENTAL SECTION
18
Figure 3.2. Length distribution of TEMPO-oxidized CNFs of different charge
densities.
Figure 3.3. AFM image of TEMPO-oxidized CNFs and width distribution of
different surface charged nanofibrils measured with AFM.
EXPERIMENTAL SECTION
19
Protein nanofibrils (PNFs): Morphology of PNFs were investigated using AFM
(Dimension FastScan Bruker) in tapping mode. Samples were diluted in HCl at pH
2 and droplets of PNF suspensions were placed on freshly cleaved mica surfaces
followed by the air drying. Experiments were also performed in liquid state using
ScanAsyst Liquid+ cantilevers (Bruker) operating in Peak Force mode. Both the
measurements from the dry and the liquid states showed similar differences between
the morphologies of straight and curved PNFs. For further analysis on the fibril’s
morphology, end-to-end lengths and contour lengths were calculated from AFM
images by using Image J software. For the persistence length (see SI or paper V),
~300 fibrils for each morphology were measured. Persistence lengths and fibril
heights measured with AFM were used to get a rough estimation of the relative
elastic moduli of the PNFs using a method reported elsewhere.79-80
Preparation of composite CNF/silk films and fibers
CNF films (paper IV) were casted from a CNF suspension onto the Teflon sheets,
and were air-dried at room temperature. CNF/silk composite films (CNF/Z-silk and
CNF/FN-silk) were also prepared in the same way. Z- and FN-silks were produced
in Escherichia-coli and purified essentially according to a procedure described
before by Hedhammar et al.81-83 CNF and CNF/silk fibers were prepared using a
double flow-focusing geometry. The hydrogel fibers were collected in a water bath,
winded up on a steel frame followed by air drying at room temperature.
Preparation of composite CNF/Z films and coatings
CNF/Z films were casted onto Teflon sheets and air-dried at room temperature in the
same manner as mentioned previously for CNF and CNF/silk. Films of CNF and
CNF/Z-silk were also prepared as controls. The amount of Z molecules in the CNF/Z
films was equimolar to the amount of Z-silk in the CNF/Z-silk films. For the
preparation of CNF/Z coatings, the casted CNF films were immersed (dip-coated) in
a Z protein solution (0.063 mg mL-1) for 15 min at room temperature. Following
removal of the unbound Z solution by thoroughly washing the films using Tris
buffer, films were incubated with 20 mM Tris buffer of pH 8 for 15 min at room
temperature. Controls of CNF/Z-silk and CNF coatings were prepared the same way.
Binding of IgG to CNF/Z-silk
Dried films and fibers of CNF/Z-silk were placed in phosphate-buffered saline of
PBS at pH 7.4, followed by the incubation with rabbit anti-mouse IgG-Alexa Fluor
488 conjugate (0.05 mg mL-1, Thermo Fisher Scientific) containing 0.5% (v/v)
EXPERIMENTAL SECTION
20
Tween 20 for 30 minutes at room temperature. Composite films and fibers were
washed three times with a mixture of PBS and 0.5% Tween 20 buffer. Fluorescence
microscopy was used to check the binding of IgG to the films and fibers. Composites
of CNF and CNF/FN-silk were treated in the same way that were used as controls.
Binding of IgG to CNF/Z
Binding of rabbit anti-mouse IgG-Alexa Fluor 488 conjugate to as prepared CNF/Z
films and coatings was performed the similar way, as previously mentioned for
CNF/Z-silk. For controls, films and coatings of CNF/Z-silk and CNF were used.
Culture of Fibroblasts and Endothelial Cells
Primary human dermal fibroblasts (HDF) were cultured in Dulbecco's Modified
Eagle Medium/Nutrient Mixture F-12 together with fetal bovine serum (5%) and
penicillin−streptomycin (1%). Cells were expanded for a period of 7 days and
viability was checked with trypan blue. Experiments were carried out in passages of
8 to 13. For endothelial cells (EC) culture, the human dermal microvascular
endothelial cells (HDMECs) isolated from dermis were used. Gelatin-coated flasks
were used to expand and culture EC.
Live/Dead Staining
Living and dead cells were visualized with live/dead viability assay based on two-
color fluorescence for real-time identification of living and dead cells with two
probes (intracellular esterase activity and plasma membrane integrity). Enzymatic
conversion is used to determine the intracellular esterase activity in living cells.
Ethidium homodimer-1 enters the damaged membranes of cells and bind to nucleic
acids to undergoes a significant enhancement of fluorescence, resulting in a bright
red fluorescence in the nuclei of dead cells. Phosphate-buffered saline (PBS) was
used to wash the cell matrices.
Filamentous Actin Staining
Staining of filamentous actin in the cells was done using Phalloidin. Before staining,
cell matrices were washed with PBS together with paraformaldehyde (4%), and
permeabilized with Triton X-100 (0.1%). 4′,6-diamidino-2-phenylindole (DAPI)
was used for nuclear staining.
EXPERIMENTAL SECTION
21
Enumeration of Cells Bound to Films
HDMECs were seeded onto CNF films with or without FN-silk proteins and were
cultured for 2, 4, and 6 after washing with PBS buffer. Nuclei and F-actin were
stained, and images were taken using an inverted Nikon Eclipse Ti fluorescence
microscope. Stained nuclei numbers were counted manually in each image and
expressed in terms of number of cells per area (A = 1 mm2). Statistical evaluation
was done using Student’s t-test.
Small Angle X-ray Scattering (SAXS)
SAXS was used to quantify orientation of CNFs and PNFs at different downstream
positions in the channel. Anisotropy of the orientation distribution of nanofibrils was
measured with the anisotropy of 2D scattering patterns. P03 beamline84 at PETRA
III storage ring (DESY, Hamburg) was used to perform the SAXS measurements.
The wavelength (𝜆 ) of the X-ray beam was 0.96 Å with a distance of 6950 mm
between the sample and the detector. The size of the X-ray beam was 20 × 10 µm2
(horizontal × vertical) and to monitor the scattering diffractograms, a single-photon
counting detector (Pilatus 300k by Dectris, Switzerland) with a pixel size of 172 ×
172 µm2 was used.
We used a slightly modified channel geometry for the in-situ fibrils orientation
measurements compared to the one that was used for the fibers fabrication. The
second focusing step from the double flow-focusing geometry was mimicked by a
contraction step, and the width of channel after the contraction was set to 0.5 mm.
Due to the limited access to the synchrotron and high exposure times, experiments
were carried out under dynamic flow conditions without inducing the transition to
gel.
Order parameters (S) were quantified based on a similar procedure as described
before.84-86 Scattering patterns with the flow of only DI water were used as
background. An assumption on zero-level intensity is made by considering that no
fibrils is aligned in the perpendicular direction to the flow at the position in the
channel, where highest intensity (orientation) is observed. The intensity at azimuthal
angle (𝜃) = 0 or π is therefore subtracted to all the distributions. The final intensity
distributions were averaged between q range of 0.1 to 0.5 nm−1, for each azimuthal
angle.
S is defined as:
EXPERIMENTAL SECTION
22
𝑆 =<3
2𝑐𝑜𝑠2𝜃 −
1
2> (1)
Expanding the average gives:
𝑆 = ∫ 𝐼(𝜃) (3
2𝑐𝑜𝑠2𝜃 −
1
2) 𝑠𝑖𝑛𝜃 𝑑𝜃
𝜋
0 (2)
which is normalized according to:
∫ 𝐼(𝜃)𝑠𝑖𝑛𝜃 𝑑𝜃 = 1𝜋
0 (3)
where 𝐼(𝜃) is the intensity distribution averaged along examined q-value for each
azimuthal angle.
Fiber Characterization
Wide Angle X-ray Scattering (WAXS): WAXS measurements were performed at PO3
beamline at PETRA III storage ring at DESY (Hambury, Germany). The wavelength
(𝜆 ) of the X-ray beam was 0.96 Å with a distance of 71 mm between the sample and
the detector. The size of the X-ray beam was 6 × 14 µm2 (horizontal × vertical) and
to monitor the scattering diffractograms, a single-photon counting detector (Pilatus
300k by Dectris, Switzerland) with a pixel size of 172 × 172 µm2 was used. Order
parameter (S) and orientation index (fc) were calculated from the intensity
distribution profiles, according to the below equation:
𝑓𝑐 = (180° − 𝐹𝑤ℎ𝑚)/180° (4)
where, Fwhm is the full width at half-maximum of the azimuthal profiles.
Tensile test measurements: Tensile tests were carried out in the Solid Mechanics
Department at KTH on Instron E100 instrument equipped with a load cell of 5N
(Figure 3.4). Temperature and humidity inside the room were controlled based on
the requirements. The dimensions (test-section length and diameter) were measured
by an optical microscope at 20x magnification and few samples were also cross-
checked with SEM for the fiber diameters to validate the results from the optical
microscope and avoid errors. Diameters of individual CNF and PNF fibers were
uniform throughout the cross section; however, the diameters were slightly different
for the different fibers. Typical diameters of the CNF and PNF fibers were around
6.8 ± 0.9 μm and 40 ± 5 μm, respectively. End of the fibers were glued on the papers.
The span length was 9−12 mm, and measurements were performed at a crosshead
speed of 0.5 mm min−1. Cross-section of fibers were assumed to be circular.
EXPERIMENTAL SECTION
23
Figure 3.4. Instron E100 instrument equipped with a 5N load cell used for the tensile
tests of CNF, CNF/silk and PNF fibers. Inset is the zoomed view of test section
where sample is mounted.
Scanning and Transmission Electron Microscopy (SEM & TEM): For the SEM in
papers I, III and IV, a 5 nm platinum−palladium layer was sputtered onto the fibers
surface. The imaging of CNF, CNF-silk and PNF fibers was performed by using a
Hitachi S-4800 Field Emission Scanning Electron Microscope operated at an
acceleration voltage of 1 kV. For information on the bulk structure of CNF fibers in
paper I, samples were polymerized overnight at 60 °C and embedded in epoxy-
embedding medium. Ultrathin sections of 80 nm were cut and placed on 75 mesh
Formvar coated copper grids. The imaging was performed using a JEOL 1400 plus
TEM at 120 kV, equipped with a Ruby camera.
Rheology
We used a parallel plate geometry (DHR-2 rheometer, TA Instruments) to
investigate the viscoelastic properties of CNF, CNF/silk and PNF hydrogels (papers
IV and V). A 25 mm diameter parallel plate setup with a truncation gap of 300 µm
(paper IV) and 500 μm (paper V), respectively was used. For the CNF and CNF/silk
experiments, suspension to gel transition was induced with HCl of pH 2 and for the
EXPERIMENTAL SECTION
24
PNF experiments, we used pH 5.2 acetate buffer (Figure 3.5). Subsequently, 0.25%
strain was applied, and storage and loss moduli were measured for a period of 45
min or more (until a constant value is reached). Fracture stress was also measured at
a constant frequency of 1 Hz.
Figure 3.5. Schematic of the parallel-plate setup used for the rheology of CNF,
CNF/silk or PNF gels. Suspension was placed on the bottom plate of the rheometer
and gelling agent was added along the circumference to form gel.
Rotary Diffusion Measurements
The rotary diffusion constant (Dr), is highly dependent on the length distribution and
concentration of the fibrils, i.e. the physical crosslinking of the fibrils during their
rotation. Doi and Edwards87 developed a model to study the rotary diffusion of
monodisperse rods, which was extended to polydisperse systems by Marrucci and
Grizzuti88-89 (see papers II and III for details).
The diffusion coefficient of polydisperse suspensions in semi-dilute regime is given
by88-89:
𝐷𝑟(𝐿) =𝛽𝑘𝐵𝑇
ƞ𝐿4
1
∫ 𝑐(𝐿′)𝐿′𝑄(𝐿,𝐿′)𝑑𝐿′(∫ 𝑐𝐿
0(𝐿′)𝐿′4𝑄(𝐿,𝐿′)𝑑𝐿′+ 𝐿3 ∫ 𝑐(𝐿′)𝐿′𝑄(𝐿,𝐿′)𝑑𝐿′)′
∞𝐿
∞0
(5)
where β is a numerical factor, kB the Boltzmann constant, T the temperature, η the
viscosity of the solvent and �� the concentration distribution of the rods depending on
L. Q represents the anisotropic corrections depending on the orientation of rods
having a length L and 𝐿′ and is defined as:
𝑄(𝐿, 𝐿′) =4
𝜋 ∫ 𝑑𝑝 𝑑𝑝′𝛹(𝑝, 𝐿)𝛹(𝑝′, 𝐿′)|𝑝ᴧ𝑝′| (6)
where Ψ(𝑝, 𝐿) and Ψ(𝑝′, 𝐿′) being the orientational distributions of the rods of length
L and 𝐿′. Therefore, the diffusion coefficient of a rod of length L depends on its
EXPERIMENTAL SECTION
25
interactions with other rods of different lengths, which may have different
orientation distributions. Hence the relaxation dynamics between the rods and
relative and coupled.
We assume that the anisotropic corrections are negligible, because the system is
always far from full alignment. This means that Q ≈ 1 and this leads to
𝐷𝑟(𝐿) =𝛽𝑘𝐵𝑇𝐿∗
4
ƞ𝐿7 Γ(L) (7)
with 𝐿∗ = (∫ ��(𝐿′)𝐿′𝑑𝐿′∞
0)-1/2
and
𝛤(𝐿) =∫ 𝑐(𝐿′)𝐿′𝑑𝐿′
∞0
∫ 𝑐𝐿
0(𝐿′)𝐿′(𝐿′/𝐿)3𝑑𝐿′+∫ 𝑐(𝐿′)𝐿′𝑑𝐿′)
∞𝐿
(8)
The factor Γ(L) depends on the length L and on the concentration distribution �� of
the nanorods of different lengths, related to the length distribution in case of
polydisperse nanoparticles suspensions (CNFs). It is the correction factor due to
interactions between nanorods of different lengths.
RESULTS & DISCUSSIONS
26
4. Nanostructured CNF Fibers
Excellent strength of nanoscale building blocks in materials is due to their defect-
free molecular structure. It is exceedingly challenging to devise nanostructuring
strategies for translating such high intrinsic properties to macroscopic materials.
Cellulose has theoretically been perceived as strongest and stiffest among the bio-
based materials in its nanoscale crystalline form. However, these superior properties
are yet to be realized for nanocellulose-based macroscopic structures. In this chapter,
our findings on the CNF fibers in pursuit of maximal mechanical performance, by
hydrodynamically controlling the structural ordering of nanofibrils, are reported. We
employ high intensity X-rays and POM to examine the behavior of CNFs in the
microchannel. The use of double flow-focusing channel results in fibers with highly-
ordered and strongly interacting arrangement of CNFs. The E of the fibers reaches
up to 86 GPa and a σu of 1.57 GPa, which are substantially higher than any natural
or synthetically prepared bio-based material. We also compare the mechanical
properties of our fibers with fossil-based synthetic fibers that are typically used.
4.1 Assembly and In-Situ Orientation Dynamics of CNFs
Figure 4.1. Schematic of double flow-focusing channel used for CNFs assembly.
The CNFs suspension is injected in the core flow (light brown color), DI water (blue
color) in the first sheath, and acid at low pH (light green color) in the second sheath
flows. Arrows demonstrate the flow direction. Hydrodynamic and electrostatic
interactions at different positions along the channel are illustrated schematically on
RESULTS & DISCUSSIONS
27
the right. Position 1, poor fibril alignment due to Brownian diffusion and
electrostatic repulsion (illustrated with the dashed arrows) caused by dissociated
carboxyl (−COOH) groups on the fibrils surface. Position 2 hydrodynamically
induced alignment (illustrated by solid, green arrow) occurs during
acceleration/extension. Position 3 further increase in alignment during
acceleration/extension, and in position 4, following the acid addition, Brownian
diffusion is minimized due to the transition of CNF suspension to an immobilized
volume filling arrested state due to protonation of the COO− groups. For illustration,
the relative size of the fibrils has been magnified around 300 times. The use of acid
for transforming the free-flowing CNF suspension to a fibrous colloidal glass
ensures that the electrostatic repulsions are replaced by van der Waals forces in the
protonated carboxyl groups. This complete removal of electrostatic repulsion is not
possible with simple electrolytes.90
Figure 4.2. (A) FE-SEM image of the fiber surface, where the dense fibrillar
network with well-preserved anisotropic arrangement can be seen. (B) FE-SEM
image of the cross-section of the fiber.
RESULTS & DISCUSSIONS
28
Figure 4.3. TEM images of a nanostructured CNF fiber highlighting a uniform
distribution of fibrils inside the bulk structure without any obvious voids or defects.
Scale bars are 100 nm. (A) and (B) correspond to different locations.
Fibers from TEMPO-oxidized CNFs are fabricated by the process of
hydrodynamic alignment and assembly. Fibrils in the suspension are aligned in the
microchannel before “locking” the nanostructure into metastable colloidal glass.91
This goal was accomplished using extensional flow-field-based double flow-
focusing channel (Figure 4.1). In the core flow, suspension of charged CNF fibrils
is used, where fibrils are free to rotate due to electrostatic repulsions and Brownian
motion, only restrained by fibril−fibril physical crosslinking (partial entanglements).
Note that the electrostatic repulsion caused by the dissociated COOH groups on the
surface of CNFs is much higher than the attractive van der Waals forces at neutral
to slightly alkaline pH. The first sheath flow of DI water supports electrostatic
repulsion and prevents transition into the colloidal glassy state in contact with
channel walls; it also aligns the fibrils toward the flow direction. Before the
alignment is diminished by the Brownian diffusion, the second flow of low pH acid
enhances the fibril alignment as observed with the in-situ SAXS experiments
discussed later in this chapter, while reducing the electrostatic repulsion between the
fibrils due to protonation of carboxyl (COO−) groups that allows the supramolecular
interactions between CNFs to self-organize the fibrils into a well-packed state with
maximized CNF−CNF contacts. The continuous threads obtained from the flow-
induced assembly were subsequently held at their ends and air-dried.
Characterization of the assembled structure with FE-SEM (Figure 4.2) sampled in
the longitudinal direction show uniform sized CNF fibers with dense and highly
oriented nanofibrils without obvious packing defects or voids (Figure 4.3).
Micrographs of fiber cross sections sampled in the transversal direction confirm the
dense fibrillar packing and reveal a well-defined layered structure (Figure 4.2B). In
RESULTS & DISCUSSIONS
29
this study (paper I), we used TEMPO-oxidized CNFs with three different surface
charge densities (550, 820 and 1360 µequiv. g-1) named as CNF-550, CNF-820 and
CNF-1360 with mean fibril lengths of 590, 683 and 391 nm, respectively (Figure
3.2).
Figure 4.4. In-situ study of the alignment, dealignment and rotary diffusion of
fibrils. (A) Microscopy image of the channel used for SAXS measurements, placed
between two cross-polarized filters rotated 45° from the vertical axis (white arrows).
RESULTS & DISCUSSIONS
30
White color corresponds to the birefringence signal obtained for the CNF-550
suspension. Numbers represent the positions where in-situ measurements were
carried out. Scale bar is 1 mm. (B) SAXS scattering diffractograms at different
positions along the channel for CNF-550 suspension (top row) and CNF-1360
suspension (bottom row). Curved lines represent the beam stopper. (C) Local order
parameters calculated from the SAXS scattering diffractograms as a function of
downstream position normalized with the channel width (h). (D) Birefringence
signal obtained in a single flow-focusing channel for the CNF-550 suspension. Scale
bar is 1 mm. The black squares along the center line represent the positions where
the Dr values are estimated. (E) Birefringence signal obtained at z/h = 3 as a function
of time when the flow is viciously stopped. The inset shows the initial decay of the
birefringence signal for the CNF-1360 suspension. The dashed line indicates the fit
of the initial decay to measure the Dr. Negative and positive time represents the
condition before and after stopping the flow, respectively. (F) Dr as a function of
downstream position along the center line. The vertical dashed line shows the
position z/h = 3 from where the birefringence decays are plotted in (E). This position
also corresponds to the filled black square at z/h = 3 in (D). The colors in (E) and (F)
correspond to different suspensions: CNF-550 (red) and CNF-1360 (blue).
(Reproduced from Ref.24. Copyright 2018 American Chemical Society)
In-situ monitoring of CNFs behavior in microchannel was essential for
successful adaptation of multiple environments. POM shows the orientation of fibrils
(Figure 4.4A), with gradually increasing birefringence from randomized suspensions
to an oriented state.74 In-situ alignment is quantified using synchrotron-based SAXS
(Figure 4.4B). Local order parameters of CNF suspensions for two different lengths
of the constituting fibrils (CNF-550 with 590 nm and CNF-1360 with 391 nm mean
lengths) are calculated (Figure 4.4C). An order parameter of 1 represents a fully
oriented state of the fibrils (in the direction of fiber preparation), and 0 corresponds
to an isotropic fibril distribution. Initially, the shear from channel walls gave rise to
some ordering (z/h = 1) in the CNF suspension. At the beginning of the focusing step
(z/h = 2), the order parameter decreases due to deceleration of the core flow followed
by a sudden increase after the focusing due to acceleration (3 ≤ z/h < 4).
Subsequently, the order parameter decreases slightly (4 ≤ z/ h < 7) before increasing
again during acceleration in the contraction (7 ≤ z/h < 10). The decrease in order
parameter after the focusing and contraction steps (4 ≤ z/h < 7; 10 ≤ z/h < 15)
indicates nanofibril relaxation toward isotropy, primarily due to Brownian
diffusion.74 Substantial differences between the orientation and disorganization
behavior of CNF-550 and CNF-1360 suspensions are observed: the shorter the fibrils
(CNF-1360), the faster the process of orientation and disorganization (Figure 4.4C).
This effect is due to the diffusivity based on length distributions of the nanofibrils as
RESULTS & DISCUSSIONS
31
given by Dr of CNF-1360, which is twice that to CNF-550 (Figure 4.4D-F) (Refer
papers I and III for details).
4.2 Mechanical Properties
Figure 4.5. Tensile mechanical properties and nanostructure characterization of the
CNF fibers. (A) Stress−strain curves for nanostructured fibers made from CNFs of
different using double flow-focusing channel. Experiments are conducted at 50%
RH. Effect of (B) physical (RH) and (C) chemical (cross-linking with BTCA)
approaches for tuning the tensile mechanical properties of fibers (prepared from
CNF-550 suspension). Plots comparing (D) E and σu of different fibers prepared in
this work. LHC and CL stand for low humidity condition and cross-linking,
respectively. Error bars correspond to the standard deviation obtained from 10
samples for each case. (E) Azimuthal integration of the (2 0 0) scattering plane of
the diffractograms for CNF-550 and CNF-1360 samples. Diffractogram
corresponding to a CNF-550 fiber is shown in the inset. (Reproduced from Ref.24.
Copyright 2018 American Chemical Society)
We establish a relationship between fibril characteristics (length, surface charge)
and tensile mechanical properties, that is important to provide the foundation for
future rational design of materials with targeted performance maxima. For a
thorough understanding on the structure-property relationship, we fabricated fibers
with CNF-550, CNF-820, and CNF-1360 and compared the mechanical properties.
The stress−strain curves in Figure 4.5A show an initial pseudo-elastic linear region,
followed by the deviation from linearity (plastic region). The “knee” in the tensile
RESULTS & DISCUSSIONS
32
test curves represent the elastic−plastic transition and is attributed to yielding
mechanisms related to relative sliding between the fibrils.92 CNF-550 and CNF-820
show similar stress−strain behavior with a E and σu of ∼70 GPa and ∼1200 MPa,
respectively (Figure 4.5A-D). This is probably due to the relatively small difference
in the mean length of the constituent nanofibrils and indicates that strength and
stiffness of the CNF fibers are independent of fibril surface charge. This is further
verified by preparing fibers with CNF-1360 with mean fibril length much lower than
CNF-550 and CNF-820, where we observed a significant decrease in σu (630 MPa).
Moreover, the relatively lower E of CNF-1360 (45 GPa) is also due to the low
nanofibrils orientation, as verified by WAXS (Figure 4.5E). The (2 0 0) reflection of
cellulose I, which generally is used to study the orientation of cellulose crystals, is
here used to track the orientation of CNFs, as crystals are aligned in the fibril
direction.93 The fibril orientation in the CNF fibers was calculated using the
following equation based on azimuthal breadth analysis:78
𝑓𝑐 = (180° − 𝐹𝑊𝐻𝑀)/180° (9)
Full width at half maximum (FWHM) is determined from the azimuthal profile of (2
0 0) plane (Figure 4.5E). The fc for CNF-550 fibers is 0.92 (order parameter = 0.70),
whereas CNF-1360 fibers show fc of 0.83 (order parameter = 0.53). In general, the
stiffest materials have tendency to be the strongest if provided with defect-free
structure and strong interfaces to ensure adequate load transfer and cohesion within
the material.5 The E and σu of our fibers vividly demonstrate the importance of
connectivity and bonding at the fibril−fibril interfaces facilitating multiscale stress
transfer. To further outspread the property range of these materials, we evaluated
different approaches based on physical (varying the ambient condition) and chemical
(covalent cross-linking of the fibrils) strategies. With CNF-550 fibers conditioned at
14% RH for a period of 40 h before testing, CNF-LHC exhibited a E and σu values
as high as 82 ± 4 GPa and 1320 ± 85 MPa, respectively (Figure 4.5D). As one might
partially expect from previous studies on bioinspired nanocomposites, strengthening
of interfibrillar interactions and removal of water molecules from the fibril surface
resulted in marked improvement of fiber mechanics with respect to other CNF
composites but also led to embrittlement and low failure strain. At high RH, water
acts as a plasticizer and allows extended elastic deformations, thus reducing the E
and increasing the ε.94 To reduce the humidity effect and potentially improve the
connectivity between the nanofibrils, the CNF-550 fibers were cross-linked by
1,2,3,4-butane tetracarboxylic acid (BTCA) that was used to neutralize the
suspensions during the flow-based assembly of the CNF fibers. BTCA creates
covalent bridges between CNF fibrils,95 replicating to some degree the cross-links
between cellulose and lignin/hemicellulose. The average strength of cross-linked
fiber (CNF-CL) tested at 50% RH increased to 1430 MPa (highest measured value
RESULTS & DISCUSSIONS
33
of 1570 MPa) with negligible change in modulus (Figure 4.5D). Chemical cross-
linking introduces covalent bonds between the fibrils, which improves the
interfibrillar connectivity and stress transfer.
To the best of our knowledge, even the mechanical properties of fiber fabricated
with CNF-1360 in this study, have previously been unachievable for CNF fibers
prepared with other approaches that rely mostly on the shear forces (Appendix Table
1). Hence, there is a strong and profound benefit of exploiting extensional flow-fields
for the orientation and assembly of CNFs (or elongated particles, in general), to bring
fresh insights in designing future high-performance load-bearing architectures by
getting closer to the theoretical limit of the building blocks. Further, the fc values for
highly oriented nanofibrils in CNF-550 fibers, reported in this study are only slightly
higher than those reported for CNF-based fibers fabricated with other approaches.62,
96 However, up to 6 times higher σu values of our fibers (Appendix Table 1) indicates
that apart from the orientation of nanoscale building blocks, interfaces and
interactions at the nanoscale play a key role in determining the mechanical properties
of macrostructures.
Other important observation is that the highest ever recorded strength of plant
fiber with high orientation of nanoscale crystalline planes, free from damage and
natural imperfections, is ∼1550 MPa (stiffness unknown).69 The plant fiber had
structural integrity provided by nature through centuries of evolution, where
nanocellulose are embedded in soft amorphous matrices of natural binders
(hemicelluloses and lignins) that probably assist with the distribution of the stress
and thus reduce the nanoscale defects. Interestingly, our TEMPO-oxidized CNF
fibers fabricated using double flow-focusing channel reach the σu equal to the highest
ever reported value for a plant fiber found in the nature with E on the order of
∼70−90 GPa.
RESULTS & DISCUSSIONS
34
4.3 Deformation Mechanism at the Nanoscale
Figure 4.6. Stress-strain data from loading-unloading experiments in uniaxial
tension for nanostructured CNF-550 fiber tested at 50 % RH at the same pulling rate
used for the tensile test of the fibers. Beyond the yield point, two cycles were
measured. Yellow and red lines represent the same slopes in different regimes
(before and after yield point).
Figure 4.7. Stress-strain data from loading-unloading experiments in uniaxial
tension for cellulose nanopaper with random-in-plane CNFs network. Reproduced
from Ref.92. (Copyright 2008 American Chemical Society)
RESULTS & DISCUSSIONS
35
Figure 4.8. MD simulation model for the relative sliding between neighboring CNF.
Variation of number of hydrogen bonds per unit area for two surfaces in direct
contact. Scale bar is 5 nm. (Image credit: Pan Chen)
High ε values (∼6%) obtained for the uncrosslinked CNF fibers is rather
uncommon for highly oriented cellulose structures. Structural changes and
interfibrillar interactions at the nanoscale are further investigated by cyclic
loading−unloading tests under uniaxial tension in the post-yield regime (Figure 4.6).
The post-yield E may increase due to reorientation of the fibrils as for the case of
random-in-plane CNFs network (Figure 4.7), along the test direction.92 Interestingly,
the fiber integrity is completely recovered upon unloading at post-yield strain values
in the present study. Several micro- and nanocomposites also show a decrease in
modulus due to the damages in the form of cracks at reinforcement−matrix interface.
However, unchanged E in the present case suggests a lack of further change in
orientation (due to the very high orientation of CNFs in the fibers) and the
mechanisms of plastic deformation must involve reformable secondary bonds as in
the case of stick−slip.97 The existence of stick-slip mechanism between the fibrils
was further verified by MD simulation (Figure 4.8), where the number of hydrogen
bonds per unit area remains constant during the relative sliding between the
nanofibrils.98 Plastic deformation of CNF fibers is substantially reduced upon cross-
linking, and the stress− strain curve becomes relatively linear as the secondary
RESULTS & DISCUSSIONS
36
interactions are replaced by covalent bonds (Figure 4.5C). Although interfibrillar
interactions are dominated by the relatively weak hydrogen bonds and van der Waals
forces,25 the high orientation of fibrils amplifies their effects due to collective
synergy of molecular interlocking, leading to stiffening and effective energy
dissipating mechanisms (stick− slip and molecular zip-up).97, 99
4.4 Mechanical Properties in a Wider Perspective
Figure 4.9. Tensile mechanical properties of bio-based and selected synthetic fiber
materials. Overview of specific σu vs specific E for a range of materials. The region
of fibers fabricated in the present work is shown in dark gray. (Reproduced from
Ref.24. Copyright 2018 American Chemical Society)
RESULTS & DISCUSSIONS
37
Figure 4.10. Tensile mechanical properties of bio-based and synthetic fiber
materials including aramids, carbon and ultra-high-molecular-weight polyethylene.
Dragline spider silk is considered as strongest biopolymeric fiber from past
several years. The comparison of the nanostructured CNF fibers reported in this
study, with the properties of dragline silk is also revealing. Notably, the stiffness of
CNF fibers is 8 times higher than dragline spider silk fibers with strengths higher
than most categories of dragline spider silks (Figure 4.9). The specific strength of
our CNF fibers is also higher than the metals, alloys, and silica-based E-glass fibers
(Appendix Table 2). Interestingly, the strongest CNF fibers have tensile strengths
1.2−1.5 times higher than wet-spun carbon nanotubes and graphene fibers100-101,
despite the fact that theoretical strength of crystalline cellulose is orders of
magnitude lower than that of a carbon nanotube and a graphene sheet. This is
possibly one of the limiting factors because the specific strength and stiffness of the
highly oriented CNF fibers are still lower than aramids, ultrahigh molecular weight
polyethylene (UHMWPE), and carbon fibers (Figure 4.10).
Carbon nanotubes as building blocks have much better mechanical properties
than CNFs as they contain 2D network of covalently linked carbon-carbon bonds, so
breaking a carbon nanotube requires high amount of energy to break many such
covalent bonds present together in a dense network. Whereas, UHMWPE consists
mostly of dense crystals (up to 99% crystallinity) where carbon chains are packed in
the most ideal scenario. Such crystals have high modulus and low density that results
in the elevated specific mechanical properties.102 Same is with aramids, where chains
RESULTS & DISCUSSIONS
38
inside the crystals are highly oriented. Nevertheless, the current results on tensile
mechanical properties of CNF fibers highlight the central role of processing
strategies and associated fundamental parameters (e.g., building blocks size,
interactions, and network formation) on realizing the true potential of nanoscale
building blocks.
RESULTS & DISCUSSIONS
39
5. Fundamentals of the Hydrodynamic Assembly
Process
Successful assembly of elongated nanoparticles such as CNFs, PNFs etc. into
high-performance macroscopic materials by using the microfluidic tools, depends
on fundamental phenomena controlling rotational and translational diffusion. The
elevated tensile mechanical properties of CNF fibers reported in chapter 4 and PNF
fibers reported in chapter 7, are results of our improved fundamental understanding
on the behavior of nanoscale building blocks in the microchannels and also due to
the understanding of the parameters associated with the processing technique. It is
rather challenging to understand and quantify the in-situ dynamics of nanoparticles
in the microfluidic channels. In this chapter, our studies on a new experimental
methodology to characterize the orientation and network dynamics of elongated
nanoparticles suspensions in assembly flows using orientation relaxation, are
reported. The relaxation is measured by birefringence and the methodology is
illustrated using nanocelluloses (CNFs and CNCs) as model systems. The advantage
of extensional flow in the organization and assembly of elongated nanoparticles over
the shear flow, are also discussed.
5.1 Double Flow-Focusing Geometry
Microfluidic tools based on co-flow,103-104 step-emulsification,105 T-junction106
or flow-focusing107 geometries have facilitated the development of both fundamental
and applied research in the field of chemistry, materials, biology and physical
sciences.108 In particular, the flow-focusing configuration (Figure 5.2A), where a
core fluid is focused by two sheath flows has been successfully employed for a
variety of applications.109 Flow-focusing relies on an extensional flow and have
demonstrated a great potential for the controlled assembly of elongated nanoparticles
into strong fibers or filaments.24, 34 Extensional flow is crucial for the orientation and
forces elongated nanoparticles to align in the flow direction (discussed in next
section). In the flow-focusing channel, the rate of extension must be larger than the
rotary Brownian diffusion of the nanoparticles to align them along the fiber axis for
the enhancement of uniaxial mechanical properties.74 This rate of extension can be
tuned with the flow-rates of the sheath fluids.
Assembly of elongated nanoparticles in the suspensions to solid fibers inside the
flow-focusing geometry typically relies on a transition to a gelled or an arrested-
state.24, 34, 110 For the CNFs assembly reported in chapter 4, a double flow-focusing
channel has been used (Figure 4.1) as single step flow-focusing channel has
RESULTS & DISCUSSIONS
40
disadvantages of channel clogging and gel build-up at solid interfaces (corners)
where the core and active sheath flow comes in contact.74, 111 This in turn result into
the lower orientation of CNFs inside the fibers. Hence, to obtain an enhanced and
uniform orientation of the nanofibrils throughout the fiber cross-section, we stick to
the multi-step flow-focusing channel (double flow-focusing) for most of the studies
on materials fabrication reported in this thesis. Double flow-focusing channel assists
in enhancing the orientation of nanofibrils inside the fibers (Figure 4.4C) and the use
of DI water into the first-focusing channel, solved the problem of gel build-up at the
channel walls and also prevent shear forces on the nanofibrils from the channel walls.
5.2 Shear vs Elongation
Figure 5.1. Illustration on the arrangement of fibrils while processing through
extensional and shear-based flow systems. Fibrils arrange uniformly in the liquid jet
of core flow of CNF in the extensional flow, which is not the case in shear dominant
systems, where fibrils can flip and rotate while processing. This leads to
inhomogeneous arrangement of fibrils across the cross-section of macroscopic
fibers, which in turn has implications on the mechanical properties.
Shear and extensional flows may give rise to the significant differences in the
orientation of nanofibrils inside the fibers. An extensional flow-field will align fibrils
without intermittent flipping and rotation, as is the case for shear where alignment is
a time-averaged quantity (Figure 5.1).24 The non-uniform arrangement of nanofibrils
from skin to core in the fibers will bring down the number of contact points between
RESULTS & DISCUSSIONS
41
two adjacent fibrils and hence, the structural integrity. This will have an influence
on the mechanical properties of nanostructured materials.
A thorough investigation on the differences in the behavior of CNFs under the
influence of shear and extensional forces by Rosén et al. is further insightful in this
regard.112 Extensional flow is found to be effective in the alignment of CNFs and the
alignment vanishes quickly due to the Brownian effects as soon as acceleration is
stopped, though strong shear is still present inside the microchannels. Rotary
diffusion (discussed in the next section) that is responsible for the de-alignment is
faster under the influence of shear compared to the extensional forces.112 Both shear
and extensional forces tend to align the nanofibrils in the flow direction. However,
in the extensional flows-based systems, the perpendicular distance between fibrils is
closer compared to the shear flow-based systems, that helps to reduce the rotational
diffusion of the nanofibrils.
5.3 Rotary Diffusion and Size Dependency
In this section, a new experimental methodology that allows real-time
assessment of orientation dynamics of elongated nanoparticles under dynamic flow
conditions, is described. The approach is based on a flow-stop methodology
developed by Rosén et al.112 coupled with thorough analysis of the orientation
relaxation.14, 113 Once the flow inside the microchannel reached a steady state, it is
rapidly stopped, and the relaxation of the oriented particle toward isotropy,
depending on the diffusion of the system constituents, is then measured. The
experiments are conducted at different locations in the channel. Combining these
measurements with a model of rotational diffusion,87, 89 this methodology provides
information about the typical size of the aligned nanoparticles before the flow is
stopped. Therefore, the evolution of the length distribution of the aligned
nanoparticles in the flow can readily be quantified. For the demonstration, we used
two types of nanocelluloses: CNCs, being almost monodisperse in length, and CNFs,
with a broad length distribution (polydisperse) (Figure 3.2).
RESULTS & DISCUSSIONS
42
Figure 5.2. (A) Typical birefringence signal for the CNF suspension. The signal is
normalized by its maximum value in the channel. The squares represent the different
locations where the birefringence is averaged in space. (B) Length distributions of
the CNC (dashed dotted line) and CNF (solid line) suspensions. The insets show
TEM images with the scale bars representing 200 nm. (Reproduced from Ref.14.
Copyright 2018 American Chemical Society)
Experiments for rotary diffusion studies are conducted on a single flow-
focusing channel (refer paper II and III for details). The orientation of nanocelluloses
is visualized through birefringence using POM.112 A typical birefringence signal
from CNFs is shown in Figure 5.2A. Once the flow is stopped after reaching a steady
state, the birefringence decay B(t) is observed at multiple locations along the
centerline of the channel marked by the black squares in Figure 5.2A. The length
distributions of the corresponding nanocellulose samples measured using TEM are
plotted in Figure 5.2B.
It is important to model the Brownian effects inside the microfluidic channels
and correlate the return-to-isotropy of the aligned nanoparticles. Doi and Edwards87
developed a theory to study the Brownian motion of monodisperse rods which
depends on the length L, diameter d, and concentration c of the particles. For cL3 ≪
RESULTS & DISCUSSIONS
43
1 (dilute regime), the rods are free to rotate without any interparticle interactions,
while for 1 ≪ cL3 ≪ L/d (semi-dilute regime), the effects of particle interactions on
the particle dynamics become significant due to volume exclusion and network
formation.114-115 With a dry weight concentration of 41 g/L (cL3 ≈ 4.2), the CNCs
suspension belongs to the semi-dilute regime. The rotational diffusion coefficient of
a rod in an isotropic semi-dilute system is
𝐷𝑟 =𝛽
(𝑐𝐿3)2
𝑘𝐵𝑇 ln(𝐿
𝑑)
3𝜋ƞ𝐿3 (10)
where T is the temperature, kB is the Boltzmann constant, η is the solvent viscosity,
and β is a numerical factor. β is expected be of order 1−10, but several experiments
measured its magnitude in the range of 103−104.14, 116 This discrepancy has led the
need to redefine theories, model more precisely the diffusion of the nanorods through
the network, and possibly take into account nanoparticles flexibility. The
birefringence relaxation toward isotropy is predicted to be exponential with a typical
time scale of τ = 1/(6Dr).87 Figure 5.3 shows the birefringence relaxation of the CNC
suspension at different locations in the channel, both in a lin−lin and log−lin plot
(inset). All the curves are normalized by their initial value, B0. The curves almost
collapse, and the inset indicates an exponential decay. This illustrates that
independent of the initial alignment, only one timescale can be associated with the
relaxation of the monodisperse CNCs particle system, in agreement with Doi and
Edwards theory.87
Figure 5.3. Normalized birefringence decays for the CNC suspension at different
positions in the channel: x/h =−1 (blue), 0.75 (light blue), 1.25 (green), 1.75
(magenta), 4.75 (red), and 8 (black). The inset shows the same plot with a
logarithmic scale for the vertical coordinates.
RESULTS & DISCUSSIONS
44
The Brownian diffusion model of Doi and Edwards was primarily applicable
for monodisperse systems and hence, there was a need to extend it to polydisperse
systems, to understand the in-situ dynamics of CNFs or other similar biological
systems, where polydispersity is ubiquitous. Marrucci and Grizzuti proposed a
model for the polydisperse systems.89 In such systems, concentration of
nanoparticles in the suspensions depends on the entanglement length (L*) (see
methods for details).
Figure 5.4. Normalized birefringence decays for CNCs (empty diamonds) and CNFs
(filled circles), at x/h = 1.75. The fit of each curve using the inverse Laplace
transform method is shown as dashed dotted line for CNCs and solid line for CNFs.
The inset shows the normalized birefringence decay for CNF in a log−lin plot.
Figure 5.5. Normalized birefringence decays for CNFs. The symbols are
experimental data and the solid lines are the inverse Laplace transform fit. The colors
are for different positions in the channel: x/h = −1 (blue), 0.75 (light blue), 1.25
(green), 1.75 (magenta), 4.75 (red), and 8 (black). The two insets are schematics of
RESULTS & DISCUSSIONS
45
the orientation of the short (blue) and long (red) fibrils before the flow is stopped, in
the focusing region and far downstream in the channel.
𝐵(𝑡) = ∫ 𝐵0(𝐿)exp (−6𝐷𝑟∞
0𝐿(𝑡)𝑑𝐿 (11)
Here B0(L) is the contribution of the elongated nanoparticles of length L to the total
birefringence signal B0, before the flow is stopped. This leads to a relaxation toward
isotropy with multiple time scales, as can be seen for the CNFs in Figure 5.4 (filled
circles and inset). The difference with the exponential relaxation of the CNCs, also
plotted in Figure 5.4 (empty diamonds). These results highlight a strong coupling
between the orientation relaxation of a nanoparticle system and the length
distribution. This coupling allows us to extract information on the length distribution
of nanoparticles present in the suspension. Indeed, eq. 11 is a Laplace transform of
B0(L) and this quantity can be estimated by inverting B(t), using a method described
by Rogers et al.113 The numerical factor β given in eq. 10 has to be adapted. For the
CNFs, it has been set to β = 1.5 × 103, which is of the same order of magnitude as
reported by other experiments in the earlier studies.14 For the CNCs, β = 0.5 matches
well with the length distribution obtained using TEM in dry state (Figure 5.2B). The
discrepancy between the two values reflects that CNCs and CNFs suspensions form
3D networks differently based on their aspect ratio, and morphology. This analysis
further allows to quantify the evolution of length distribution of the aligned CNFs
along the channel and get information on the orientation dynamics of the nanofibrils
suspension in the flow (Figure 5.5) (refer paper II and III for details).
Above mentioned results allow us to establish a comprehensive understanding
of the orientation dynamics of CNFs in microfluidic channels, specifically under
extensional flow-fields. Shorter CNF nanoparticles are quickly oriented by the flow,
while more time is needed to align long ones. Furthermore, short particles de-align
faster and their motion is therefore difficult to control during the assembly. These
results are further supported by the in-situ SAXS measurements on CNF-550 and
CNF-1360 (Figure 4.4C). This scenario can be also deduced from Figure 5.5, used
as a diagnostic plot: aligned short fibrils in the focusing region exhibit short time
scales in the decay (light blue triangles), while aligned long fibrils further
downstream exhibit longer time scales (red squares and black diamonds). These
findings have strong implications on optimizing the mechanical performance of the
CNF fibers fabricated via flow-based assembly. The typical time for the transition
to gel or an arrested state, achieved with the double flow-focusing geometry in
chapter 4 is estimated to be of the order of a second (paper I),24 that is, longer than
the time found here for the dealignment of the shortest fibrils. Therefore, there is a
possibility that the CNF fibers fabricated using flow-focusing approach in chapter 4,
RESULTS & DISCUSSIONS
46
may composed of long aligned CNFs embedded into an isotropic matrix of shorter
CNFs.
5.4 Network Formation at the Nanoscale
Results in the previous section highlight that the orientational dynamics of CNFs
or other polydisperse nanofiber systems, depends strongly on the length distribution.
Polydispersity has significant influence on the dynamics of nanoscale networks and
in turn the mechanical properties of the hierarchical macroscale materials. A broad
length distribution further challenges our fundamental understanding on the physical
phenomenon at the nanoscale in an entangled network of nanofibers, experiencing
multiple regimes depending on their length and as polydispersity increases the mean
entanglement increases significantly.116 The nanofibers assembly under the
influence of Brownian effects can be controlled by tuning the concentration and
nanofiber dimensions in a way such that entanglement restrains Brownian motion
without impeding flow-induced alignment.24
We used Nc to parametrize the entanglement within the CNFs network in the
suspension. This dimensionless number is defined as the average number of
nanofibers in a sphere with a diameter equal to the length of the fibers.24, 65
𝑁𝑐 = 2
3∅ (
𝐿
𝑑)2 (12)
Here, L is the nanofiber length, d the diameter, and ∅ the volume fraction in the
suspension. The crowding number allows us to define three different concentration
regimes.117 For Nc < 16, the nanofibers are in the dilute regime, with no physical
crosslinking between them. In the semi-concentrated intermediate regime, where 16
< Nc < 60, the nanofibers have significant physical crosslinking but are not
immobilized. For suspensions with Nc > 60, the nanofibers are in a connected
network where free motion is hindered. The two thresholds, Nc = 16 and Nc = 60, are
respectively called connectivity and rigidity thresholds.116 In the literature on
polymer-based rigid rods, the crowding number is defined as the number of
nanofibers in a cube of side L.118 This creates a small discrepancy between the two
definitions, with a factor of one half. However, similar concentration regimes can be
defined in both the cases.
In the case of a network with a nanofiber length distribution, polydispersity
gives rise to a distribution of crowding numbers. Different approaches have been
suggested to estimate this distribution and the mean crowding number <Nc>, for a
RESULTS & DISCUSSIONS
47
log-normal or a general length distribution.116 In all the cases, polydispersity
increases the <Nc>, and therefore the mean entanglement. All the nanofiber
suspensions with concentrations of 0.3 wt.% used for the CNFs fiber fabrication in
chapter 4 exhibit a mean crowding number higher than the percolation threshold
(<Nc> = 16) showing that the interactions between the nanofibers are important and
mechanical interlocking between the nanofibers is present during all stages of flow-
focusing. At the concentration of 0.3 wt.%, most of the nanofiber suspensions are
above the rigidity threshold (<Nc> = 60). This clearly emphasizes that the nanofibers
are highly entangled and constrained in a 3D network. Note that the <Nc> does not
represent the proper entanglement for all nanofibers. Indeed, the entanglement of a
nanofiber originates from the number of contact points with its neighbors. Therefore,
the shorter the nanofibers, the smaller the number of contacts and less the
entanglement. This is further supported by a recent study from Nechyporchuk et
al.119 , where almost five times higher concentration of CNCs compared to CNFs
was required to fabricate nanostructured CNC fibers with the similar approach
demonstrated in chapter 4 for the CNFs and even that high concentration of CNCs
resulted in much weaker mechanical properties of CNC fibers compared to the CNF
fibers. Nevertheless, the entanglement length L∗ defined in the previous section
represents the threshold between small nanofibers in the dilute regime (not
entangled) and the larger ones in the semi-concentrated regime (entangled) (refer
paper III for details).
To maximize nanofiber orientation in the fabrication of materials with enhanced
mechanical properties, it is therefore necessary, to work with long and partially
entangled nanofiber systems, i.e. with broad length distributions and concentrated
suspensions slightly above the rigidity threshold. Indeed, long nanofibers own a de-
alignment timescale larger than the gel transition time scale and large concentrations
lead to high entanglement and thus maximize the flow-induced alignment. Too high
concentrations are however limited by the rigid volume-spanning arrested states,91
where the suspension becomes like a gel due to entanglement. In such gel-like
suspension, flow alignment is completely restricted. Hence, it is highly important to
optimize the nanoparticle systems based on the dimensions and concentrations to
obtain best possible mechanical properties at the macroscale.
RESULTS & DISCUSSIONS
48
6. Nanostructured CNF/Silk Composites
This chapter focuses on the supramolecular assembly of CNF and silk fusion
proteins for the fabrication of strong and bioactive nanocomposites. Cellulose and
silk are two excellent biological materials- silk from spider being tough due to high
extensibility whereas cellulose possess very high strength and stiffness. Along with
the elevated mechanical properties, building blocks of spider silk (spidroins) have
demonstrated potential as biomaterials, which is typically not the case with wood-
based CNFs. Unfortunately, silk proteins cannot be obtained in large quantities from
spiders and process of recombinant production is rather expensive, Therefore, in this
study, we combined small amounts (10%) of functionalized silk proteins with
nature’s most abundant structural component (CNF) to fabricate composites that can
have properties and functionalities of both silk and cellulose. We combined CNFs
with two types of silk fusion proteins: Z-silk and FN-silk.34 In Z-silk, the silk protein
is fused to an affinity domain Z,120 which adds the ability to selectively bind
immunoglobulin (IgG).83 In FN-silk, the material is instead functionalized with a
tripeptide Arg-Gly-Asp (RGD) containing cell-binding motif from the extracellular
matrix (ECM) protein fibronectin.121 The results presented in this chapter are
selected from paper IV. CNF and silk contents are referred in wt.% unless otherwise
stated.
6.1 Fabrication of CNF/Silk Nanocomposites
We mixed CNFs and RSPs in a hydrocolloidal suspension, where the stability of
the suspension is controlled by supramolecular interactions.122 A schematic of the
atomic model used for comparison of the molecular sizes of different components,
is shown in Figure 6.1A. We fabricated isotropic films and anisotropic fibers from
the CNF and CNF/silk proteins. For the films, a straightforward preparation process
based on solvent casting is used, where free-standing films with planar fibril
networks are obtained after evaporation of water (Figure 6.1B; Figure 1(b) of paper
IV). Except the simplicity, a key advantage of this technique is the capability to
prepare large-area films with well-defined dimensions, which is not possible using
other film forming methods like dip or spin coating.123 Morphology of the films
surface show a dense web-like fibrillar network, where the CNFs from the wood
pulp fibers are clearly visible. Individual CNFs are wriggled and physically
entangled to each other, indicating strong fibril-fibril interactions (e.g. hydrogen
bonds, Van der Waals interactions92), and a result of fibril flocculation due to the
crowding factor/number density. AFM images reveal the presence of proteins in the
RESULTS & DISCUSSIONS
49
form of small clusters (3-10 nm) homogeneously distributed throughout the dense
CNFs network.
Figure 6.1. (A) Schematic of functionalized RSPs, CNFs and CNF decorated with
RSP. (B) AFM images of CNF and CNF/silk film surfaces, where homogeneous
distribution of proteins throughout the surface can be seen. (C) FE-SEM image of a
CNF fiber prepared with double flow-focusing geometry.
RESULTS & DISCUSSIONS
50
Figure 6.2. (A) Average G’ of CNF and CNF/silk gels obtained from rheology
measurements. (B) FT-IR of CNF and CNF/silk fibers.
The orientation of fibrils in the fabricated films is isotropic in the place. To
fabricate anisotropic fibers, nanofibrils were oriented along the longitudinal
direction. We utilized the same approach of elongational flow-field based double
flow-focusing geometry that was applied for the fabrication of CNF fibers from
TEMPO-oxidized CNFs. For the experiments, we used a core flow of
carboxymethylated CNF or CNF/silk suspensions. The first sheath consists of DI
water (pH ~ 5.6) and the second sheath consists of HCl (pH ~ 2). Hydrogen ions
from the acid neutralizes the carboxyl groups on the fibrils surface to induce self-
assembly. Parallel and bundled arrangement of nanofibrils can be seen in the FE-
SEM image (Figure 6.1C) of CNF fiber. No difference in the surface morphology of
CNF and CNF/silk fibers is observed with FE-SEM.
Rheological measurements (Figure 6.2A) showed that the proteins increase the
gel stiffness of CNF significantly. Inclusion of Z- and FN-silk to CNF, resulted in
30-35% higher average G’ values than the corresponding CNF gels. These results
suggest better crosslinking of CNFs in the presence of proteins is probably due to
the formation of hydrogen bonds between the -NH groups of proteins and hydroxyl
groups of cellulose.124 However, more experimental investigations are required to
get information on the nanoscale interactions between the RSPs and CNFs. The
integration of proteins with CNF was also confirmed in the air dried fibers using FT-
IR. Proteins were not washed away in the water bath, where hydrogel fibers were
collected. Peaks appeared at wavelengths ~1650 cm-1 and ~1530 cm-1 in the
CNF/silk fibers correspond to the amide I and amide II bands of proteins,
respectively, which are related to the secondary structures of proteins and the
formation of β-sheets.125
RESULTS & DISCUSSIONS
51
Another interesting observation is that RSPs can form stable suspension with
charged CNFs that can be processed into hierarchical structures subjected to
hydrodynamic forces in the microchannels, despite the fact such forces are known
to promote uncontrolled aggregation of the silk proteins.126 The capacity of CNF as
a stabilizing agent has also been observed for carbon nanotubes127 and thus
demonstrates a viable route of using it as a common building block and /or as a
stabilizer for the fabrication of high-performance nanocomposites.
6.2 Mechanical Properties of Anisotropic Fibers
Figure 6.3. Mechanical and structural characterization of the CNF and CNF/silk
fibers. (A) Tensile mechanical properties measured at 50% RH. (B) E, (C) σu and
(D) ε of CNF and CNF/silk fibers. (E) WAXS diffractogram of a CNF fiber, where
IDG denotes the intermodular detector gap. (F) Azimuthal integration of the (2 0 0)
scattering plane of the diffractogram shown in Figure 6.3E.
RESULTS & DISCUSSIONS
52
The tensile mechanical properties of CNF and CNF/silk fibers are reported in
terms of E, σu and ε. CNF fibers prepared with carboxymethylated CNF revealed an
average E of 52.8 ± 0.7 GPa, σu of 830 ± 24 MPa, and ε of around 6% with a
toughness of 37 MJ m-3 (Figure 6.3). The highest obtained σu for a CNF fiber is 854.6
MPa. E and σu values obtained with the carboxymethylated CNF are slightly lower
than the values obtained with TEMPO-oxidized CNF fibers prepared with the same
strategy. This could be due to the difference in pulp properties that was used to
liberate the nanofibrils.
The orientation of nanofibrils in the CNF fiber is characterized using WAXS
(Figures 6.3E and F). The (2 0 0) reflection of cellulose I, that in general is used to
study the orientation of crystal planes, is here used to track the orientation of CNFs,
as crystals are aligned in the fibril direction.34 The fibril orientation in the CNF fiber
was calculated using the eq. 9. The value of fc in this study (0.86) is slightly higher
than previously reported values for cellulose fibers formed by the assembly of wood-
based CNF (fc varies from 0.55 to 0.83) and lower than the value of TEMPO-
oxidized CNF-550 fiber reported in chapter 4.24 This could also be a reason for
slightly lower values of mechanical properties obtained in this study. The scatter in
data between E and fc values obtained in this study compared to the other studies
reported earlier, suggest that apart from the degree of orientation, other structural
characteristics of the macroscopic fibers as well as chemical interactions between
the constituent fibrils play an important role in determining the macroscale
mechanical properties. The higher values of strength obtained for the fibers in this
study compared to the other techniques used to orient and assemble CNFs, could
also be related to the smaller dimensions of the current fibers, which in turn decreases
the probability of defects inside the macrostructures.98
We also tested the mechanical properties of CNF/silk fibers to observe any
potential difference in the mechanical properties in the presence of small amounts of
silk fusion proteins. Fibers of CNF/FN-silk showed comparable properties, with E
of 50.2 ± 1.3 GPa, σu of 812.5 ± 25.4 MPa, and ε of 6−8% with toughness of 43 MJ
m−3 (Figure 6.3). Surprisingly, fibers of CNF/Z-silk show higher ε (10%) with
toughness of 55 MJ m−3 and σu on the order of ∼1 GPa (978.8 ± 37.5 MPa) compared
to the pure CNF fibers, whereas the E (53.6 ± 1.8 GPa) is similar. One possible
explanation is that the addition of Z-silk proteins, which harbor a stably folded
helical Z-domain fused to the silk, promotes crosslinking between CNF fibrils and
enhances the overall strength. However, this effect is less evident for the CNF/FN-
silk fibers, in which the silk modules are instead fused to only a short peptide.
Nevertheless, in the seminal work of Xia et al.71 and Lazaris et al.,70 a E of 21 and
12 GPa and σu of 508 and 259 MPa have been reported for artificially spun fibers of
RESULTS & DISCUSSIONS
53
pure silk proteins alone. It is noteworthy that the E and σu of the CNF/silk fibers
designed in this study are 2−4.5 times higher, compared to the pure silk protein fibers
due to the presence of crystalline CNFs, whereas at the same time, the toughness of
artificial silk protein fibers is maintained.70-71 Tensile tests results of fibers are in
good agreement with the rheology measurements on corresponding gels of CNF and
CNF/silk proteins (Figure 6.2A), where increased values of G′ indicate the enhanced
interactions or crosslinking between the fibrils in the presence of silk proteins. The
tensile test results of the air-dried fibers along with FTIR measurements (Figure
6.2B) also indicate that RSPs are not merely present along with the CNFs in the
suspensions, rather they interact and stick well to the surface of CNFs.
Figure 6.4. Schematic representation of a CNF/silk fiber. Green-CNFs; purple-
RSPs.
Friction between the nanofibrils decorated with RSPs and a decrease in the pH
during the self-assembly could lead to the transition of the silk proteins from α-helix
to β-sheets and thereby promote crosslinking due to hydrogen bonds formation.124
Apart from the interfibrillar interactions and the network formation that contribute
to the stiffness of a pure CNF fiber, the presence of silk proteins enhanced the
crosslinking between the CNF (Figure 6.4), which could be a reason for enhanced
mechanical properties. To reveal the molecular origin of enhanced mechanical
properties, a thorough understanding of the conformational changes, intermolecular
interactions and heterogeneous microstructure at the nanoscale in the dried fibers, is
required.
RESULTS & DISCUSSIONS
54
6.3 Bio-Functionalities
Figure 6.5. Functional studies of composite films show presence of bioactivity. (A)
Optical microscopy images of CNF and composite CNF/silk films (top panel). Scale
bar is 4 mm. Images from fluorescence microscopy visualizing the binding of
fluorophore-labeled IgG (green, bottom panel). Scale bar is 1 mm. (B) Signal from
IgG-fluorophore bound to a film casted from a mixture of CNF and the Z domain
alone (CNF/Z), CNF premixed with Z-silk (CNF/Z-silk) or pure CNF, respectively,
visualized by fluorescence microscopy (top panel). Visualization of IgG-fluorophore
bound to a CNF film coated with the Z domain alone (CNF/Z), Z-silk (CNF/Z-silk),
or noncoated (CNF) (bottom panel). Scale bars are 1 mm. (C) HDF cultured on films
of CNF or CNF/FN-silk for 7 days. Top panel: live/dead staining (green, living cells;
bright red, dead cells; dim red, autofluorescence from CNF). Bottom panel: F-actin
staining (green, F-actin; blue, nuclei/ autofluorescence from CNF). Scale bars 500
μm for 2× and 100 μm for 10×. (D) Number of HDMECs (mean + SD) per mm2 on
freestanding films of CNF (open bar) and CNF/FN-silk (gray bar), at different time
points during culture, as evaluated by DAPI staining and manual counting of nuclei.
Statistically significant differences according to Student’s t test: **p < 0.01, ***p <
0.001, and *****p < 0.00001. (E) Representative Alamar blue viability graph of
three independent experiments with HDMEC growing on CNF and CNF/FN-silk
films (n = 3) for 1 week. Statistically significant differences according to Student’s
t test: *p < 0.05. Error bars show standard deviation. (Reproduced from Ref.34.
Copyright 2017 American Chemical Society)
Specific affinity is the ability to selectively sort out one type of molecule from a
complex. It is essential in applications such as purification of antibodies
biomolecules. IgG-binding ability of CNF films in combination with Z-silk (Figure
RESULTS & DISCUSSIONS
55
6.5A and B) was evaluated. Macroscopic appearance of CNF and CNF/silk films are
nearly identical (Figure 6.5A, upper panel), however, CNF/silk films spread more
on the Teflon sheets where the films were casted, which might be due to the changes
in the wetting properties of CNFs in the presence of silk proteins. Fluorophore-
labeled IgG was used to evaluate the IgG-binding ability of the films (Figure 1.4),
which exhibits a clear green signal upon binding to the CNF/Z-silk film (Figure
6.5A, lower-middle panel). The signal is uniform over the entire surface area of the
CNF/Z-silk film, indicating a homogenous distribution of silk proteins in the
composite films. No green signal is observed for either the CNF or from CNF/FN-
silk films used for the comparison (Figure 6.5A, lower-left and right panel,
respectively). This indicates that the addition of Z-silk into the CNF adds a
prominent IgG-binding ability. This also highlights that the Z domains of Z-silk are
exposed outwards in the CNF/Z-silk film and retain the native structure (3D helical
fold) that is required for IgG-binding.
To investigate whether the combination of silk module (4RepCT) in Z-silk is
necessary to achieve IgG-binding ability, a control experiment was performed,
where CNFs were mixed with Z domain alone and casted into the films with silk
module being excluded. While performing the analysis on such films (CNF/Z), a
weak fluorescence signal is detected, indicating significantly low binding of the
labeled IgG (Figure 6.5B, upper-right panel). These results strongly emphasize that
covalent attachment of the Z domain to the silk module is necessary to achieve the
IgG-binding ability. This is probably due to poor adhesion of Z domain alone to the
CNF in the absence of silk module, that acts as an efficient mild glue. It can be
speculated that remaining Z domains will not be able to keep the 3D folded structure
and will be denatured upon direct contact with CNFs. Further investigation on the
role of silk module, is carried out on the CNF films that were coated with a solution
of the Z domain alone and the Z-silk fusion protein, respectively. Again, CNF coated
with the Z domain alone and CNF coated with Z-silk protein, was found to bind a
little IgG like the control film made of pure CNF (Figure 6.5B, lower-right and left
panel, respectively). This concluded that a covalent crosslinking between Z domain
and the silk module is needed, to strongly bind folded Z domains to the composite
materials.
In the recent years, attempts have been made to use wood-based nanocellulose
scaffolds for cell-culture.128 Mostly hydrogels of CNFs have been used for cell
encapsulation in a 3D micronenvironment.129-131 Typically, a good biocompatibility
of CNF with a maintained viability of the encapsulated cells is observed. However,
direct adhesion of cells with cellulose surfaces in a cell-matrix like fashion, is not
shown. This is very likely due to the lack of bioactive motifs in wood-based
RESULTS & DISCUSSIONS
56
cellulose. By glutaraldehyde cross-linking of CNF extracted from trees, Mertaniemi
and co-workers132 could show an improved spreading of cells on the threads after 1
week of culture.
We therefore evaluated the ability of CNF-based materials to support the
adhesion and proliferation of primary cells with the inclusion of a silk protein fused
to a cell binding motif from fibronectin (FN-silk). First, HDFs were seeded onto the
different films and adhesion of cells was traced by viability staining after 1 week
(Figure 6.5C, upper panel). A comprehensive coverage of living fibroblasts was
observed on the CNF/FN-silk films, with limited number of cells found on the films
made of pure CNF, although the present cells were also viable. Fibroblasts cultured
on CNF/FN-silk films exhibited well developed stress fibers and elongated
morphology (Figure 6.5C, lower panel). This verifies that the cell binding motif FN
fused to silk is presented uniformly in the composite films that can be recognized by
cell surface receptors mediating adhesion, e.g. integrins. To challenge the system
further, the films were seeded with HDMECs, a cell type known to be dependent of
ECM proteins for adhesion. The number of cells adhered to the CNF and CNF/silk
films were recorded 2, 3 and 6 days after seeding (Figure 6.5D). Results confirm that
inclusion of the FN-silk protein is mandatory to obtain a material which promotes
cell adhesion. HDMECs were well spread out after one-week culture on the
CNF/FN-silk films (Figure 6.6). In contrast, cells did not show these characteristics
on CNF films. Surface roughness that could be an additional factor towards the
adhesion of cells on any surface is also found to be very similar for the films
fabricated with CNF and CNF/silk proteins. To verify the proliferative nature of the
adhered cells, an Alamar blue viability assay was also carried out during 1 week of
cell culture (Figure 6.5E).
Figure 6.6. Freestanding CNF films with and without FN-silk after 7 days culture
with HDMECs and staining for f-actin (green) and nuclei (blue). Scale bars are 100
µm. (Image credit: Mona Widhe)
RESULTS & DISCUSSIONS
57
Figure 6.7. Affinity properties of the nanostructured fibers. (A) Optical microscopy
images of CNF and CNF/silk fibers. Scale bar is 4 mm. (B) CNF/Z-silk fibers ability
to bind IgG-fluorophore (green) visualized by fluorescence microscopy. Scale bar is
0.5 mm. (Reprinted from Ref.34. Copyright 2017 American Chemical Society)
Figure 6.8. Cell adhesion to fibers. (A) Live/dead staining of HDFs growing on CNF
(left panel) or CNF/FN-silk fibers (right panel) after 2 days (top panel) or 7 days
culture (middle and bottom panels). Green, living cells; bright red, dead cells; dim
red, autofluorescence from CNF. (B) F-actin staining (green) of HDMECs after 2
days (top panel) and 4 days (bottom panel) of culture adhering to CNF (left panel)
or CNF/FN-silk fibers (right panel). Dim blue: autofluorescence from CNF. Scale
bars 250 μm for 4× and 100 μm for 10×. (Reprinted from Ref.34. Copyright 2017
American Chemical Society)
RESULTS & DISCUSSIONS
58
We also evaluated the bio-functionalities in anisotropic fibers. Macroscopic
appearance of air-dried CNF/Z-silk and CNF/FN-silk fibers is like the corresponding
CNF fiber (Figure 6.7A). Regarding the IgG binding ability, a distinct green
fluorescence signal highlighting the binding of IgG, is observed in the CNF/Z-silk
fiber (Figure 6.7B, middle panel), unlike pure CNF or CNF/FN-silk fibers (Figure
6.7B, left and right panel, respectively). This indicates the visibility of appropriately
folded Z domains in the CNF/Z-silk fibers.
When cells were cultured on the fibers, a slightly different behavior was
observed compared to the isotropic films. Cells did not adhere to the CNF films,
however, fibers of CNFs supported some attachment and spreading, though to a
lesser extent than the CNF/FN-silk fibers that contain cell-binding motifs. HDFs
spread well along the CNF/FN-silk fiber, after 2 days in culture media, whereas on
CNF fibers, a few scattered rounded colonies of cells with no signs of uniform
spreading were found at this early point (Figure 6.8A and B, upper panels). However,
after 7 days, the HDFs also found to spread out along the CNF fibers, still in the
form of small clusters indicating rather poor attachment (Figure 6.8A, middle and
lower panels).
Better attachment of HDFs on CNF fibers compared to the CNF films might be
explained partly by the differences in surface morphology; the films have random
orientation of nanofibrils with nano-sized roughness of similar dimensions
throughout, while the CNF fibers have bundled uniaxial arrangement of nanofibrils
(0.2−0.3 µm) that give rise to distinct roughness values within the same material
(Figure 6.1C). It could be due to the enhanced adhesion of ECM proteins secreted to
the culture media, alignment and extension of cells on the oriented structures and
subsequent adhesion and proliferation of cells.133 In this direction, a thorough
investigation of surface micro- and nanoscale topography and correlation with the
cell functions is required. We also observe a high confluence and even distribution
of fibroblasts on the CNF/FN-silk fibers, already from start.
Further, the fibers were seeded with HDMECs, and the adhesion and spreading
of cells were analyzed by staining for F-actin. After 2 days, very few small and
rounded colonies of HDMECs were observed on CNF fibers (Figure 6.8B), while
the cells were observed to be spread out along the CNF/FN-silk fibers. After 4 days,
some HDMECs were observed to spread out also on the pure CNF fibers, though
much less in number compared to the CNF/FN-silk fibers. Thus, in both isotropic
and anisotropic hierarchical CNF structures, the presence of the cell binding motif
FN linked to the silk, promote significantly the cell adhesion to the cellulose-based
materials.
RESULTS & DISCUSSIONS
59
7. Nanostructured Protein Microfibers
Many natural materials are made from proteins. The properties of these
proteinaceous materials are closely linked to the hierarchical assembly of the protein
building blocks. Proteins from egg, soybean etc. can be assembled into PNFs that
have potential to be a building block of the macroscale materials. Recent advances
have revealed the molecular structure of PNFs, the mechanisms associated with
interfibrillar interactions while assembling into macrostructures, remain unexplored.
In addition, previous attempts to fabricate PNF-based materials largely rely on the
use of plasticizers due to weak interfibrillar interactions. In this chapter, our results
on the assembly of PNFs into strong microfibers without adding any plasticizer or
cross-linker by using double flow-focusing channel are summarized. The assembly
of PNFs of two distinct morphologies (straight and curved) is compared. Insights on
the differences in the assembly of PNFs compared to more crystalline biopolymer
building block i.e. CNF, are also provided.
Figure 7.1. Schematic of the double flow-focusing device used for PNFs assembly.
The PNF suspension is injected in the core flow. DI water injected in the first sheath
and acetate buffer (pH 5.2) in the second sheath (top left). The channel ends in a bath
containing acetate buffer (pH 5.2) (top right). Photograph of a hydrogel fiber in the
bath solution after spinning in the double flow-focusing setup (top right). Scale bar
is 1 cm. FE-SEM image of a dried fiber and a zoomed view in the inset (bottom
right). Scale bars are 30 µm and 1 µm, respectively. Fluorescence microscope image
of a fiber stained with thioflavin T (bottom left). Scale bar is 30 µm.
RESULTS & DISCUSSIONS
60
7.1 Assembly and In-Situ Orientation Dynamics of PNFs
The whey protein β-lactoglobulin self-assemble into amyloid-like nanofibrils at
high concentrations of urea,134 in the presence of alcohols,135 and at low pH and
elevated temperature.136 In the latter case, fibrillation occurs after the hydrolysis of
the protein into smaller peptide fragments which then spontaneously assemble into
PNFs. In the present study, WPI is used as a starting raw material due to an advantage
of abundant availability. Details on the assembly of WPI into PNFs of two distinct
morphology are mentioned in paper V. The measured E of straight fibrils is ∼5 GPa,
and for the curved ones is ~0.2−0.4 GPa. The solutions of obtained PNFs were then
extensively dialyzed to remove unfibrillated peptides and proteins. The dialyzed
samples of PNFs undergo a phase transition to a gel (arrested state) at a pH of 5.2
i.e. close to the isoelectric point of β-lactoglobulin.137 However, non-dialyzed
samples resulted in phase separation with opaque aggregates at this pH, highlighting
the importance of proper dialysis as unfibrillated proteins can impede the protein
fibrillar network in the gel state.
Similar to CNFs, the strategy used to fabricate pure protein fibers also rely on
suspension to gel transition during the flow-induced orientation process. We used
the same double flow-focusing device (Figure 7.1) with a core flow of PNF
suspension and two focusing steps of the sheath flows. In the first focusing channel,
DI water was used and in the second channel, we use acetate buffer at pH 5.2. Acetate
buffer neutralizes the charges on the surface of PNFs and induce the self-assembly.
We first examined the assembly of straight fibrils, where the concentration of PNF
in the core flow varied from 0.45% to 1.8% (the concentrations are lower than the
above-mentioned initial concentrations as they refer to the pure PNFs content of the
dialyzed suspensions). Surprisingly, no conditions were found where fibrils could be
assembled into fibers that were strong enough to be picked out from the bath
solution. Weak hydrogel fibers were obtained that were unable to withstand the
surface tension at the liquid–air interface while being pulled out. The curved fibrils,
on the other hand, were able to produce hydrogel fibers that were strong enough to
overcome the liquid-air surface tension (Figure 7.1). Notably, the fiber formation
with curved fibrils was possible at a concentration of 1.8%, where straight fibrils
were not able to produce strong enough hydrogel fibers. The extracted hydrogel
fibers were air-dried (Figure 7.1) and surface morphology was investigated using
FE-SEM. The images reveal a smooth surface with constant cross-section along the
fiber-axis (Figure 7.1). In the zoomed view of the fiber surface, graupel-like units of
the dimensions of a few hundred nanometers were observed. Interestingly,
morphology of the assembled PNFs observed in this study is similar to previous
RESULTS & DISCUSSIONS
61
studies on the assembly of RSPs,71 where colloidal aggregates clustered together to
form weak structures in the solution. However, surface morphology of non-
crystalline PNFs is very different from crystalline CNFs, where we observed
oriented bundles of fibrils on the fiber surface.24 The presence of β-sheet secondary
structures in the fibers was confirmed by fluorescence microscopy (Figure 7.1), by
using the amyloidophilic dye thioflavin T. We further characterized the tensile
mechanical properties of the fabricated fibers, where stress–strain curves reveal a
rather brittle material with E up to 288 MPa and ε of ∼1.5% (paper V). Interestingly,
the measured E values of fibers are in good agreement with the estimated values for
the curved nanofibrils.
RESULTS & DISCUSSIONS
62
Figure 7.2. In-situ study of PNFs in the microchannel. (A) Schematic of the channel
geometry used for the SAXS experiments. White circles show the locations where
in-situ measurements were carried out. (B) SAXS diffractograms at selected
locations in the channel for straight (Upper Row) and curved (Lower Row) fibrils.
(C) Local order parameters calculated from SAXS diffractograms as a function of
downstream position z in the channel. The locations marked with circles in A
correspond to the data points shown in C. (D) Images of the channel placed between
two crossed polarization filters rotated 45° from the vertical axis (white arrows). For
straight fibrils, birefringence due to the alignment is observed after the focusing step.
In contrast, no birefringence is detected for the curved fibrils. (Reproduced from
Ref.66. Copyright 2017 National Academy of Sciences)
RESULTS & DISCUSSIONS
63
To understand the behavior of straight and curved PNFs in the microchannel,
we carried out similar SAXS experiments as reported earlier in chapter 4 for CNFs.
Same channel geometry was used for the SAXS experiments, where the second
focusing step was mimicked by a contraction (Figure 7.2A), as suspension to gel
transition was not required for these experiments. For the PNFs, similar to the CNFs,
an increased alignment of fibrils immediately after the focusing and the contraction
steps was observed (Figure 7.2B and C). Subsequently, relaxation of the aligned
fibrils toward isotropy due to Brownian motion result in decrease in order parameter.
Considerable differences in the behaviors of straight and curved PNFs were
observed. Very low orientation of curved fibrils in the flow direction is observed, as
indicated by the lower values of order parameter (Figure 7.2C). The maximum order
parameter recorded at z = 8 for the straight fibrils is 0.41, which is almost two times
the order parameter of curved fibrils (0.21) at the same channel location. SAXS
results are further confirmed qualitatively by POM (Figure 7.2D), where straight
fibrils displayed birefringence highlighting the ordering of PNFs and no
birefringence was observed for the curved fibrils that further confirms a lower degree
of alignment.
Figure 7.3. Behavior of PNF gels subjected to oscillatory strain in a parallel plate
rheometer. (Reproduced from Ref.66. Copyright 2017 National Academy of
Sciences)
RESULTS & DISCUSSIONS
64
Figure 7.4. Schematic of the interactions of straight and curved PNFs in bulk and in
the flow-focusing channel. (Left) In bulk, the deformation of gels of straight fibrils
under strain is governed by the joint motion (rotation and sliding), whereas curved
fibrils resist the strain due to the higher degree of entanglement. (Right) In the
channel, straight fibrils have higher orientation compared to the curved fibrils in the
flow direction as noticed with SAXS. In contrast, curved fibrils with shorter
persistence length are physically entangled resulting in a strong network between the
fibrils to form a fiber. (Reproduced from Ref.66. Copyright 2017 National Academy
of Sciences)
Figure 7.5. FT-IR spectrum of non-gelled suspension, hydrogel and dried fiber of
curved fibrils.
RESULTS & DISCUSSIONS
65
7.2 Micro- and Molecular Structure in the Gel State
Interfibrillar interactions are investigated for the straight and curvy fibrils by
studying the fracture behavior of the corresponding gel networks under macroscopic
strains using parallel plate rheometer (Figure 3.5). At similar fibril concentrations, a
noticeable difference in the value of G′ was observed (Figure 7.3). Initial alignment
of straight fibrils in a nematic phase could be the reason for the higher G′ values.138
Rheology results also reveal differences in the gel fracture behavior in response to
the increasing oscillatory strains. Gel from straight fibrils exhibits a slow decrease
in G′ before it fractures, whereas for the curved fibrils, a sudden drop in the G′ values
is observed. The differences in the gel fracture behavior is associated with the gel
networks, where the slow deformation for straight fibrils is presumably associated
with rotation and sliding of the nanofibrils, which plastically deforms their network
at high oscillatory strains. On the other hand, network of curved fibrils fractured
without plastic deformation, indicating that there might be an interlocking due to the
entanglements and eventually the entangled network breaks at high strain values.
Rheology results throw insights on the potential differences in the behavior of
morphologically distinct fibrils in the microchannels. When straight fibrils are
aligned by the extensional flow, the possibility to further optimize interfibrillar
network decreases due to higher alignment, resulting in suboptimal physical cross-
linking of nanofibrils. In contrast, when the entangled network of curved fibrils is
extended and thus redistributed under the influence of extensional flow-fields, the
physical crosslinking might be enhanced, and a stronger entangled network is
achieved (Figure 7.4). This is also clear from the FE-SEM image in Figure 7.1, where
dense packing of PNFs can be seen. To get information on the molecular structure
of PNFs in fibers, we acquired FT-IR spectra in amide I and amide II band regions
that correspond to the protein secondary structures.125 The amide I and II bands are
centered at ∼1,625 cm−1 and ∼1,517 cm−1, respectively, which reflects the presence
of β-sheets (Figure 7.5).
7.3 Hierarchical Protein Materials and Comparison with Other
Building Blocks
Typically, the orientation of nanofibrils or nanotubes along the fiber axis results
in strong nanoparticles network that improves the unidirectional mechanical
properties. For instance, the orientation of CNFs and carbon nanotubes inside the
macro-sized fibers resulted in enhanced tensile properties.24, 100 The results presented
in this chapter show a different behavior of PNFs compared to other building blocks.
Although the morphology and dimensions of PNFs resemble largely with CNFs, the
RESULTS & DISCUSSIONS
66
nanoscale structural features and chemical compositions are indeed different. The
differences in the origin of PNFs and CNFs can further elucidate the differences into
the hierarchical structures. CNFs from wood are highly crystalline and are extracted
through chemical pretreatment followed by mechanical decomposition of cellulose
fibers in trees. PNFs, on the other hand, are self-assembled from protein monomers
that do not correspond to any hierarchical architecture already exist in nature. Hence,
PNFs (non-crystalline) and CNFs (crystalline) are two distinct classes of
nanomaterials and have different assembly patterns into macroscopic materials.
Interestingly, our results on the assembly of PNFs into microfibers have some
similarities with the assembly mechanism of RSPs.71 Similar to the PNF microfibers,
spider silk also has a high degree of β-sheet secondary structure and the silk fibers
are formed in a duct inside the spiders, by using the elongational flow followed by
pH alterations.28 However, the nanoscale structural characteristics of amyloid-like
nanofibrils and spider silks are different. In spider silks, β-strands and β-sheets are
parallel to the main fiber axis, in a string-of-bead fashion, in which anti-parallel β-
sheets and amorphous regions co-exist and alternate, providing high strength and
extensibility than amyloids, where β-strands are arranged perpendicular to the fiber
axis while the β-sheets are parallel.139 Other important difference in the assembly of
spider proteins into silk, is the simultaneous formation and control of the
supramolecular structures inside the spider glands. This allows the formation of
entangled structures with optimized intermolecular interactions at the same time
highly ordered β-sheet regions are created. Considering that not even artificially
assembled silk protein fibers reach the mechanical properties on the same level to
the natural spider silk, it is indeed fascinating that we can produce pure protein fibers
from a rather crude raw material and with a simple and potentially scalable
experimental setup. Our insights on the PNF assembly will provide a basis for
rational design of novel materials with tailored properties.
CONCLUSIONS
67
8. Conclusions
The main findings from the studies reported in this thesis are summarized in this
chapter.
Flow-assisted assembly is a promising route for the fabrication of structural
materials with highly ordered arrangement of nanoscale building blocks such as
polymer- and protein-based nanofibrils, nanotubes, and inorganic nanorods.24, 34, 66,
140-143 Mechanical properties of our CNF and PNF fibers further support this
conclusion. The elevated mechanical performance of CNF fibers is a result of
understanding and controlling the key fundamental process parameters (nanoparticle
size, interactions, alignment, diffusion, network formation and assembly), that are
central to obtain a near-perfect nanostructuring. Length of the nanofibrils and
enhancement in interfibrillar interactions by physical and/or chemical crosslinking,
largely controls the mechanical properties of CNF fibers in the dry state. Conditions
such as fibrils length and concentration must be optimized such that there is a
restraint on the Brownian motion of fibrils inside the microchannels by mechanical
interlocking without hindering the movement that in turn will further promote the
flow-induced alignment. Above a certain fibril concentration in the suspensions,
fibril-fibril physical crosslinking not only impedes mobility, but also provide
necessary structural integrity during the assembly. However, too high concentrations
of fibrils in the suspensions results in volume-spanning arrested states due to
entanglements, which restrict the flow-induced alignment.
It is important to understand the effect of flow-fields on the organization of
building blocks, specifically in microfluidics-based design of new materials.
Typically, an extensional flow-field aligns fibrils without intermittent flipping and
rotation, which is the case for shear flow where alignment is a time-averaged
quantity. This enhances the physical crosslinking between the elongated
nanoparticles and thus results in structural integrity and improved mechanical
properties. Diffusion due to Brownian motion inside the microchannels is also a
crucial parameter at the nanoscale. However, the ability to quantify the diffusion
dynamics in microchannels remains challenging as the tools that can provide time-
resolved measurements under dynamic conditions are limited. Using POM, we build
a methodology to trace the rotational Brownian motion of nanocelluloses with the
help of birefringence measurements that further expands the capabilities to probe in-
situ dynamics of flowing suspensions. The results reveal that characteristic length
scales are strongly dependent on the nanoparticles (CNFs) length distributions and
entanglements.24 The knowledge from the diffusion and network dynamics of
CONCLUSIONS
68
nanocelluloses in this thesis can also be used to optimize the mechanical properties
hierarchical materials from other building blocks, apart from nanocelluloses. Our
methodology of dynamic characterization of nanocelluloses should be extendable to
other nanoscale building blocks such as amyloids, chitin, metal nanoparticles etc.
This tool will further help to get a facile and easy access to nanoscale dynamics for
a variety of biophysical systems that are presently limited by the access of
synchrotron-based X-rays and neutron techniques.
The study on CNF/silk composites shows that coupling of genetically engineered
silk proteins and CNFs leads to functional effects in both aqueous dispersed state
and dried state. Charged CNFs enhanced the dispersibility of silk proteins and
thereby the processability, which is not typical otherwise as hydrodynamic forces
tend to agglomerate silk proteins.126 Inclusion of Z-silk proteins gives rise to the
affinity properties in cellulose materials, while the incorporation of FN-silk proteins
in CNFs resulted in ideal matrix for cell culturing that enabled cell attachment,
migration, proliferation and differentiation. The efficient crosslinking between RSPs
and CNFs in the dry state enhanced the interactions between the nanofibrils, thereby
resulted in increased toughness and an elevated strength at break. Nonetheless, based
on the results presented in chapter 6, it can be expected that mechanical properties
and bio-functionalities in bio-based materials can be tuned through the coupling of
proteins and crystalline nanofibrils by interlinking architectures. This study further
shows a functional basis for the modularity found in natural materials and suggests
a biological approach for molecular-level genetic engineering of the soft-matrix in
nanocomposites.
The high order parameter of CNFs in the microchannels resulted in the
enhancement of uniaxial mechanical properties of macroscale fibers.24, 74 However,
the results obtained with the PNFs show a different behavior. Although the
morphology (shape and size) of PNFs might appear like CNFs, the intrinsic
properties are indeed different due to the differences in the origin. CNFs from wood
are extracted by mechanical fibrillation of long cellulose fibers. Crystalline part of
cellulose fibers highlights the intrinsic ability of trees in nature to form crystals by
highly ordered arrangement of the polymer chains. PNFs, on the other hand, are
obtained from the self-assembly of protein monomers that do not correspond to any
highly ordered structures that can lead to the formation of strong nano-sized crystals.
Hence, despite the similarities in size and aspect ratio, both the biopolymeric
building blocks (PNFs and CNFs) are two distinct classes of nanomaterials where
CNFs are crystalline and PNFs are non-crystalline. This is the reason the mechanical
properties of CNF fibers are orders of magnitude higher than PNF fibers. However,
our results with PNFs show some similarities between the microfibers fabricated
with whey proteins reported in this thesis and the assembly of RSPs reported
CONCLUSIONS
69
earlier.144 Like the PNFs, spider silk has β-sheet structures and the natural spider silk
fibers are also formed by a process inside the ampullate glands of spiders which is
similar to the spinning process here, where spidroins (spider proteins) are subjected
to elongational flow and pH alterations.145 However, the orientation of the β-sheets
relative to the fiber axis is perpendicular in the silk fibers to that expected for aligned
PNFs139, which leads to differences in structural characteristics and the mechanical
properties of dragline silk and amyloid-like whey protein nanofibril-based materials.
Furthermore, the molecular- and supramolecular structures that are formed
simultaneously during the spinning process inside the spider glands allows the
formation of entangled structures with much optimized intermolecular interactions.
This argument is further supported by the differences in the mechanical properties
observed between the natural dragline silk and man-made RSPs-based fibers. Our
study with PNFs shows the importance of interfibrillar entanglements over the
orientation to fabricate strong enough free-standing coherent protein fibers.
OUTLOOK
70
9. Future Work and Outlook
Mechanical properties of CNF fibers
Mechanical properties of CNF fibers obtained with our process is significantly
higher compared to other previous attempts made in the field. However, the true
potential of CNF is still unknown and a question of debate. Further developments in
the processing techniques are required to realize the full potential of CNFs in all-
cellulose-based macrostructures. It may be possible that the values reported in this
thesis are already close to the limit, however this still needs to be verified. One
potential way to achieve this is by tuning the length and crystallinity of CNFs. CNFs
from different wood species need to be tested. New ways to in-situ crosslink CNFs
in the microchannels may also contribute positively in this direction.
Role of ions on the mechanical properties of CNF fibers
In this thesis, we mostly used H+ ions from HCl to neutralize the charges of CNFs to
induce the self-assembly of nanofibrils. Apart from H+ ions from the acids, other
monovalent, divalent and trivalent ions such as Na+, Ca+2, Fe+2, Al+3 etc. can also be
used to lock the CNFs in the aligned state in the microchannels. It would be
interesting to see the influence of these ions on the mechanical properties and to
compare the nanoscale network and orientation of nanofibrils in the dried fibers with
WAXS.
Wet strength of CNF fibers
In our studies, we tested only the dry strength of CNF fibers (≤50% RH) as the focus
of our studies was to check the maximum potential of CNFs in all-cellulose
macrostructures. For industrial applications, one major challenge that needs to be
addressed is the sensitivity of CNF towards water. Water acts as a plasticizer for
polysaccharides such as cellulose146 and this drastically decreases the mechanical
properties when exposed to water or moist conditions. Hence, it would be
worthwhile to test the mechanical properties in wet conditions, and potentially
combine CNFs with other building blocks to bring down its sensitivity towards
water.
PNF fibers
Further exploration of processing conditions and/or protein building blocks is
required to fill the parameter space (mechanical entanglements, fibril orientation,
spatial distribution of fibrils, tuning the surface chemistry of PNFs and covalent
OUTLOOK
71
crosslinking) of PNFs assembly in detail. This will provide the basis for design of
novel protein-based materials with improved mechanical properties.
CNF/silk composites
Mechanical properties of CNF/silk composites can be tuned further by changing the
concentration of silk proteins with respect to CNFs. New silk proteins with stronger
ionic interactions and /or new bio-functionalities can also be synthesized and tested
with CNFs. Novel CNF/silk materials with improved mechanical properties or
advanced bio-functionalities can be designed. More on the fundamental side, it
would be interesting to find out the conformation of proteins at the nanoscale in the
dried structures (films and fibers) with respect to CNFs. Techniques like AFM and
neutron scattering may be able to provide more information.
Nanocomposites fabrication
Another direction is to fabricate high-performance nanocomposites with the process
of hydrodynamic orientation and assembly. Other building blocks such as chitosan,
chitin, metal nanoparticles, nanosheets etc. can be tried individually or in
combination and their mechanical properties can be optimized with more
fundamental understanding on their behavior in the suspension state.
Industrial production of fibers
Though we greatly manage to realize the potential of CNFs and PNFs in macro-sized
fibers, it is necessary to maximize the production speed in order to increase
scalability. Parallelization of flow-focusing channels can contribute positively in
scaling up the production. Potential ways to simultaneously dry the hydrogel threads
to be able to roll them up continuously, also need to be determined.
Rotary diffusion measurements using POM
We tried to quantify the in-situ orientational and network dynamics of
nanocelluloses using POM and birefringence measurements. Same technique can
also be used to get fundamental information on the behavior of other building blocks
such as amyloids, chitin etc. in the microchannels to optimize the performance of
nanostructured materials fabricated via hydrodynamic approaches. This tool can also
be used to check batch to batch variations in CNFs and other building blocks that
will further help to adapt these nanomaterials for industrial scale production. With
respect to the nanocellulose suspensions that are used as model systems in this thesis,
we used water as solvent. However, viscosity can be tuned by introducing low or
high molecular weight polymers and their effect on the relaxation dynamics
OUTLOOK
72
(characteristic length scales of diffusion) can also be studied. This information is
relevant for many polymer and particle-based dynamic systems.
Validation with simulations
Experiments performed in this thesis can be the test cases for simulations. It is
important to find out whether the models available today can predict and reproduce
the experimental results presented here. Most of the numerical models consider
elongated nanoparticles (CNFs, PNFs) as nanorods. It is challenging to consider the
periodic kinks or curviness of the fibrils into the numerical models. New models,
therefore, need to be developed to study the behavior of elongated nanoparticles in
the channels based on the differences in the morphology and the information on the
nanoscale physics can in turn be used in the industrial processes dealing with fibers
and suspensions.
ACKNOWLEDGEMENTS
73
Acknowledgements
The Knut and Alice Wallenberg foundation is greatly acknowledged for providing
funding through Wallenberg Wood Science Center.
I would like to express the deepest gratitude to my advisor Daniel for providing me
an opportunity to start this journey 4 years ago and for his continuous guidance,
much patience and support. I could have hardly imagined of meeting such a humble
person and getting a better advisor for my PhD studies. I would like to thank Farhan
for his support and guidance from the very initial days I started at KTH. I would
further like to thank my co-advisor Prof. Fredrik Lundell for the learnings that I have
had from the discussions with him. I thank Prof. Eugenia Kumacheva, University of
Toronto (UofT) for accepting me to her group as an International Visiting Graduate
Student and Prof. Lars Berglund (Director of WWSC) for kindly providing me the
financial assistance through WWSC for an international exchange.
A special thanks to my colleagues at WWSC and all members of WWSC Academy
for the wonderful and unforgettable moments spent together during the winter and
summer schools. It is important highlighting the names of Lilian, Nicola, Tobias B.
and Tahani from WWSC and Daniel and Mahshid from UofT for all the nice
discussions, providing a pleasant working environment and always being very
supportive. I would like to thank my colleagues Christophe, Krishne, Ayaka, Tomas,
and Calvin at the Department of Mechanics for sharing their knowledge and
providing the generous support. Thanks to Prof. Lars Wågberg, Prof. My
Hedhammar, Prof. Christofer Lendel, Prof. Benjamin Hsiao and Prof. Nicholas
Kotov for all the great insights and ideas. Thanks to Prof. Stephan Roth for kindly
providing the beamtime at DESY. Thanks to Prof. Henrik Alfredsson at KTH
Mechanics for always being very kind and helping me with the courses. Thanks to
Irene Linares Arregui from the Solid Mechanics Department at KTH for assistance
with tensile test measurements and helpful discussions.
I am much obliged to Prof. Shervin Bagheri, the reviewer of this work as well as
Lilian and Farhan for their valuable suggestions while proof reading this thesis. Most
importantly, I am forever grateful to my family, especially my parents and brother
for their unconditional love, encouragement, constant motivation and support not
only by numerous sacrifices but also by providing me with the best instead of theirs.
REFERENCES
74
References
1. Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek, G.; Cairney, J.;
Eckert, C. A.; Frederick, W. J.; Hallett, J. P.; Leak, D. J.; Liotta, C. L.; Mielenz,
J. R.; Murphy, R.; Templer, R.; Tschaplinski, T., The Path Forward for Biofuels
and Biomaterials. Science 2006, 311 (5760), 484-489.
2. Buggy, M., Natural fibers, biopolymers, and biocomposites. Edited by Amar K
Mohanty, Manjusri Misra and Lawrence T Drzal. Polymer International 2006,
55 (12), 1462-1462.
3. Sanchez, C.; Arribart, H.; Giraud Guille, M. M., Biomimetism and
bioinspiration as tools for the design of innovative materials and systems.
Nature Materials 2005, 4, 277.
4. Ling, S.; Kaplan, D. L.; Buehler, M. J., Nanofibrils in nature and materials
engineering. Nature Reviews Materials 2018, 3, 18016.
5. Wegst, U. G. K.; Bai, H.; Saiz, E.; Tomsia, A. P.; Ritchie, R. O., Bioinspired
structural materials. Nature Materials 2014, 14, 23.
6. Ling, S.; Chen, W.; Fan, Y.; Zheng, K.; Jin, K.; Yu, H.; Buehler, M. J.; Kaplan,
D. L., Biopolymer nanofibrils: Structure, modeling, preparation, and
applications. Progress in Polymer Science 2018, 85, 1-56.
7. Meyers, M. A.; Chen, P.-Y., Biological Materials Science: Biological
Materials, Bioinspired Materials, and Biomaterials. Cambridge University
Press: Cambridge, 2014.
8. Meyers, M. A.; Chen, P.-Y.; Lin, A. Y.-M.; Seki, Y., Biological materials:
Structure and mechanical properties. Progress in Materials Science 2008, 53
(1), 1-206.
9. Service, R. F., Silken promises. Science 2017, 358 (6361), 293-294.
10. Gibson, L. J., The hierarchical structure and mechanics of plant materials.
Journal of The Royal Society Interface 2012, 9 (76), 2749-2766.
11. Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J., Cellulose
nanomaterials review: structure, properties and nanocomposites. Chemical
Society Reviews 2011, 40 (7), 3941-3994.
12. Dufresne, A., Nanocellulose: a new ageless bionanomaterial. Materials Today
2013, 16 (6), 220-227.
13. Rånby, B. G., Fibrous macromolecular systems. Cellulose and muscle. The
colloidal properties of cellulose micelles. Discussions of the Faraday Society
1951, 11 (0), 158-164.
14. Brouzet, C.; Mittal, N.; Söderberg, L. D.; Lundell, F., Size-Dependent
Orientational Dynamics of Brownian Nanorods. ACS Macro Letters 2018, 7 (8),
1022-1027.
15. Klemm, D.; Kramer, F.; Moritz, S.; Lindström, T.; Ankerfors, M.; Gray, D.;
Dorris, A., Nanocelluloses: A New Family of Nature-Based Materials.
Angewandte Chemie International Edition 2011, 50 (24), 5438-5466.
16. Jonoobi, M.; Oladi, R.; Davoudpour, Y.; Oksman, K.; Dufresne, A.; Hamzeh,
Y.; Davoodi, R., Different preparation methods and properties of nanostructured
REFERENCES
75
cellulose from various natural resources and residues: a review. Cellulose 2015,
22 (2), 935-969.
17. Usov, I.; Nyström, G.; Adamcik, J.; Handschin, S.; Schütz, C.; Fall, A.;
Bergström, L.; Mezzenga, R., Understanding nanocellulose chirality and
structure–properties relationship at the single fibril level. Nature
Communications 2015, 6, 7564.
18. Herrick, F. W.; Casebier, R. L.; Hamilton, J. K.; Sandberg, K. R.,
Microfibrillated cellulose: morphology and accessibility. ; ITT Rayonier Inc.,
Shelton, WA: 1983; p Medium: X; Size: Pages: 797-813.
19. Turbak, A. F.; Snyder, F. W.; Sandberg, K. R., Microfibrillated cellulose, a new
cellulose product: properties, uses, and commercial potential. ; ITT Rayonier
Inc., Shelton, WA: 1983; p Medium: X; Size: Pages: 815-827.
20. Isogai, A.; Saito, T.; Fukuzumi, H., TEMPO-oxidized cellulose nanofibers.
Nanoscale 2011, 3 (1), 71-85.
21. Wågberg, L.; Decher, G.; Norgren, M.; Lindström, T.; Ankerfors, M.; Axnäs,
K., The Build-Up of Polyelectrolyte Multilayers of Microfibrillated Cellulose
and Cationic Polyelectrolytes. Langmuir 2008, 24 (3), 784-795.
22. Hasani, M.; Cranston, E. D.; Westman, G.; Gray, D. G., Cationic surface
functionalization of cellulose nanocrystals. Soft Matter 2008, 4 (11), 2238-2244.
23. Ho, T. T. T.; Zimmermann, T.; Hauert, R.; Caseri, W., Preparation and
characterization of cationic nanofibrillated cellulose from etherification and
high-shear disintegration processes. Cellulose 2011, 18 (6), 1391-1406.
24. Mittal, N.; Ansari, F.; Gowda.V, K.; Brouzet, C.; Chen, P.; Larsson, P. T.; Roth,
S. V.; Lundell, F.; Wågberg, L.; Kotov, N. A.; Söderberg, L. D., Multiscale
Control of Nanocellulose Assembly: Transferring Remarkable Nanoscale Fibril
Mechanics to Macroscale Fibers. ACS Nano 2018, 12 (7), 6378-6388.
25. Nishiyama, Y., Molecular interactions in nanocellulose assembly.
Philosophical Transactions of the Royal Society A: Mathematical, Physical and
Engineering Sciences 2018, 376 (2112), 20170047.
26. Vollrath, F.; Knight, D. P., Liquid crystalline spinning of spider silk. Nature
2001, 410, 541.
27. Hinman, M. B.; Jones, J. A.; Lewis, R. V., Synthetic spider silk: a modular fiber.
Trends in Biotechnology 2000, 18 (9), 374-379.
28. Rising, A.; Widhe, M.; Johansson, J.; Hedhammar, M., Spider silk proteins:
recent advances in recombinant production, structure–function relationships and
biomedical applications. Cellular and Molecular Life Sciences 2011, 68 (2),
169-184.
29. Askarieh, G.; Hedhammar, M.; Nordling, K.; Saenz, A.; Casals, C.; Rising, A.;
Johansson, J.; Knight, S. D., Self-assembly of spider silk proteins is controlled
by a pH-sensitive relay. Nature 2010, 465, 236.
30. Gatesy, J.; Hayashi, C.; Motriuk, D.; Woods, J.; Lewis, R., Extreme Diversity,
Conservation, and Convergence of Spider Silk Fibroin Sequences. Science
2001, 291 (5513), 2603-2605.
31. Römer, L.; Scheibel, T., The elaborate structure of spider silk: structure and
function of a natural high performance fiber. Prion 2008, 2 (4), 154-161.
REFERENCES
76
32. Scheibel, T., Spider silks: recombinant synthesis, assembly, spinning, and
engineering of synthetic proteins. Microbial Cell Factories 2004, 3 (1), 14.
33. Tokareva, O.; Michalczechen-Lacerda, V. A.; Rech, E. L.; Kaplan, D. L.,
Recombinant DNA production of spider silk proteins. Microbial Biotechnology
2013, 6 (6), 651-663.
34. Mittal, N.; Jansson, R.; Widhe, M.; Benselfelt, T.; Håkansson, K. M. O.;
Lundell, F.; Hedhammar, M.; Söderberg, L. D., Ultrastrong and Bioactive
Nanostructured Bio-Based Composites. ACS Nano 2017, 11 (5), 5148-5159.
35. Schacht, K.; Scheibel, T., Processing of recombinant spider silk proteins into
tailor-made materials for biomaterials applications. Current Opinion in
Biotechnology 2014, 29, 62-69.
36. Jansson, R.; Thatikonda, N.; Lindberg, D.; Rising, A.; Johansson, J.; Nygren,
P.-Å.; Hedhammar, M., Recombinant Spider Silk Genetically Functionalized
with Affinity Domains. Biomacromolecules 2014, 15 (5), 1696-1706.
37. Widhe, M.; Johansson, U.; Hillerdahl, C.-O.; Hedhammar, M., Recombinant
spider silk with cell binding motifs for specific adherence of cells. Biomaterials
2013, 34 (33), 8223-8234.
38. Wohlrab, S.; Müller, S.; Schmidt, A.; Neubauer, S.; Kessler, H.; Leal-Egaña,
A.; Scheibel, T., Cell adhesion and proliferation on RGD-modified recombinant
spider silk proteins. Biomaterials 2012, 33 (28), 6650-6659.
39. Chillakuri, C. R.; Jones, C.; Mardon, H. J., Heparin binding domain in
vitronectin is required for oligomerization and thus enhances integrin mediated
cell adhesion and spreading. FEBS Letters 2010, 584 (15), 3287-3291.
40. Chaudhuri, O.; Koshy, S. T.; Branco da Cunha, C.; Shin, J.-W.; Verbeke, C. S.;
Allison, K. H.; Mooney, D. J., Extracellular matrix stiffness and composition
jointly regulate the induction of malignant phenotypes in mammary epithelium.
Nature Materials 2014, 13, 970.
41. Lawrence, B. D.; Marchant, J. K.; Pindrus, M. A.; Omenetto, F. G.; Kaplan, D.
L., Silk film biomaterials for cornea tissue engineering. Biomaterials 2009, 30
(7), 1299-1308.
42. Numata, K.; Kaplan, D. L., Silk-based delivery systems of bioactive molecules.
Advanced Drug Delivery Reviews 2010, 62 (15), 1497-1508.
43. Knowles, T. P. J.; Mezzenga, R., Amyloid Fibrils as Building Blocks for Natural
and Artificial Functional Materials. Advanced Materials 2016, 28 (31), 6546-
6561.
44. Dobson, C. M., Protein folding and misfolding. Nature 2003, 426, 884.
45. Fowler, D. M.; Koulov, A. V.; Balch, W. E.; Kelly, J. W., Functional amyloid
– from bacteria to humans. Trends in Biochemical Sciences 2007, 32 (5), 217-
224.
46. Otzen, D.; Nielsen, P. H., We find them here, we find them there: Functional
bacterial amyloid. Cellular and Molecular Life Sciences 2008, 65 (6), 910-927.
47. Tanaka, M.; Collins, S. R.; Toyama, B. H.; Weissman, J. S., The physical basis
of how prion conformations determine strain phenotypes. Nature 2006, 442,
585.
REFERENCES
77
48. Li, C.; Adamcik, J.; Mezzenga, R., Biodegradable nanocomposites of amyloid
fibrils and graphene with shape-memory and enzyme-sensing properties. Nature
Nanotechnology 2012, 7, 421.
49. Sasso, L.; Suei, S.; Domigan, L.; Healy, J.; Nock, V.; Williams, M. A. K.;
Gerrard, J. A., Versatile multi-functionalization of protein nanofibrils for
biosensor applications. Nanoscale 2014, 6 (3), 1629-1634.
50. Knowles, T. P.; Fitzpatrick, A. W.; Meehan, S.; Mott, H. R.; Vendruscolo, M.;
Dobson, C. M.; Welland, M. E., Role of Intermolecular Forces in Defining
Material Properties of Protein Nanofibrils. Science 2007, 318 (5858), 1900-
1903.
51. Loveday, S. M.; Su, J.; Rao, M. A.; Anema, S. G.; Singh, H., Effect of Calcium
on the Morphology and Functionality of Whey Protein Nanofibrils.
Biomacromolecules 2011, 12 (10), 3780-3788.
52. vandenAkker, C. C.; Engel, M. F. M.; Velikov, K. P.; Bonn, M.; Koenderink,
G. H., Morphology and Persistence Length of Amyloid Fibrils Are Correlated
to Peptide Molecular Structure. Journal of the American Chemical Society 2011,
133 (45), 18030-18033.
53. Habibi, Y., Key advances in the chemical modification of nanocelluloses.
Chemical Society Reviews 2014, 43 (5), 1519-1542.
54. Pullawan, T.; Wilkinson, A. N.; Eichhorn, S. J., Influence of Magnetic Field
Alignment of Cellulose Whiskers on the Mechanics of All-Cellulose
Nanocomposites. Biomacromolecules 2012, 13 (8), 2528-2536.
55. Bordel, D.; Putaux, J.-L.; Heux, L., Orientation of Native Cellulose in an
Electric Field. Langmuir 2006, 22 (11), 4899-4901.
56. Blell, R.; Lin, X.; Lindström, T.; Ankerfors, M.; Pauly, M.; Felix, O.; Decher,
G., Generating in-Plane Orientational Order in Multilayer Films Prepared by
Spray-Assisted Layer-by-Layer Assembly. ACS Nano 2017, 11 (1), 84-94.
57. Jin, H.; Cao, A.; Shi, E.; Seitsonen, J.; Zhang, L.; Ras, R. H. A.; Berglund, L.
A.; Ankerfors, M.; Walther, A.; Ikkala, O., Ionically interacting nanoclay and
nanofibrillated cellulose lead to tough bulk nanocomposites in compression by
forced self-assembly. Journal of Materials Chemistry B 2013, 1 (6), 835-840.
58. Shahzadi, K.; Zhang, X.; Mohsin, I.; Ge, X.; Jiang, Y.; Peng, H.; Liu, H.; Li, H.;
Mu, X., Reduced Graphene Oxide/Alumina, A Good Accelerant for Cellulose-
Based Artificial Nacre with Excellent Mechanical, Barrier, and Conductive
Properties. ACS Nano 2017, 11 (6), 5717-5725.
59. Hausmann, M. K.; Rühs, P. A.; Siqueira, G.; Läuger, J.; Libanori, R.;
Zimmermann, T.; Studart, A. R., Dynamics of Cellulose Nanocrystal Alignment
during 3D Printing. ACS Nano 2018, 12 (7), 6926-6937.
60. Siqueira, G.; Kokkinis, D.; Libanori, R.; Hausmann, M. K.; Gladman, A. S.;
Neels, A.; Tingaut, P.; Zimmermann, T.; Lewis, J. A.; Studart, A. R., Cellulose
Nanocrystal Inks for 3D Printing of Textured Cellular Architectures. Advanced
Functional Materials 2017, 27 (12), 1604619.
61. Alizadehgiashi, M.; Gevorkian, A.; Tebbe, M.; Seo, M.; Prince, E.; Kumacheva,
E., 3D-Printed Microfluidic Devices for Materials Science. Advanced Materials
Technologies 2018, 3 (7), 1800068.
REFERENCES
78
62. Lundahl, M. J.; Klar, V.; Wang, L.; Ago, M.; Rojas, O. J., Spinning of Cellulose
Nanofibrils into Filaments: A Review. Industrial & Engineering Chemistry
Research 2017, 56 (1), 8-19.
63. Yudin, V. E.; Dobrovolskaya, I. P.; Neelov, I. M.; Dresvyanina, E. N.;
Popryadukhin, P. V.; Ivan’kova, E. M.; Elokhovskii, V. Y.; Kasatkin, I. A.;
Okrugin, B. M.; Morganti, P., Wet spinning of fibers made of chitosan and chitin
nanofibrils. Carbohydrate Polymers 2014, 108, 176-182.
64. Yan, J.; Zhou, G.; Knight, D. P.; Shao, Z.; Chen, X., Wet-Spinning of
Regenerated Silk Fiber from Aqueous Silk Fibroin Solution: Discussion of
Spinning Parameters. Biomacromolecules 2010, 11 (1), 1-5.
65. Geng, L.; Mittal, N.; Zhan, C.; Ansari, F.; Sharma, P. R.; Peng, X.; Hsiao, B. S.;
Söderberg, L. D., Understanding the Mechanistic Behavior of Highly Charged
Cellulose Nanofibers in Aqueous Systems. Macromolecules 2018, 51 (4), 1498-
1506.
66. Kamada, A.; Mittal, N.; Söderberg, L. D.; Ingverud, T.; Ohm, W.; Roth, S. V.;
Lundell, F.; Lendel, C., Flow-assisted assembly of nanostructured protein
microfibers. Proceedings of the National Academy of Sciences 2017, 114 (6),
1232-1237.
67. Ling, S.; Qin, Z.; Li, C.; Huang, W.; Kaplan, D. L.; Buehler, M. J., Polymorphic
regenerated silk fibers assembled through bioinspired spinning. Nature
Communications 2017, 8 (1), 1387.
68. Hooshmand, S.; Aitomäki, Y.; Norberg, N.; Mathew, A. P.; Oksman, K., Dry-
Spun Single-Filament Fibers Comprising Solely Cellulose Nanofibers from
Bioresidue. ACS Applied Materials & Interfaces 2015, 7 (23), 13022-13028.
69. El-Hosseiny, F.; Page, D. H., The mechanical properties of single wood pulp
fibers: Theories of strength. Fiber Science and Technology 1975, 8 (1), 21-31.
70. Lazaris, A.; Arcidiacono, S.; Huang, Y.; Zhou, J.-F.; Duguay, F.; Chretien, N.;
Welsh, E. A.; Soares, J. W.; Karatzas, C. N., Spider Silk Fibers Spun from
Soluble Recombinant Silk Produced in Mammalian Cells. Science 2002, 295
(5554), 472-476.
71. Xia, X.-X.; Qian, Z.-G.; Ki, C. S.; Park, Y. H.; Kaplan, D. L.; Lee, S. Y., Native-
sized recombinant spider silk protein produced in metabolically engineered
<em>Escherichia coli</em> results in a strong fiber. Proceedings of the
National Academy of Sciences 2010, 107 (32), 14059-14063.
72. Jun, Y.; Kang, E.; Chae, S.; Lee, S.-H., Microfluidic spinning of micro- and
nano-scale fibers for tissue engineering. Lab on a Chip 2014, 14 (13), 2145-
2160.
73. Lutz-Bueno, V.; Zhao, J.; Mezzenga, R.; Pfohl, T.; Fischer, P.; Liebi, M.,
Scanning-SAXS of microfluidic flows: nanostructural mapping of soft matter.
Lab on a Chip 2016, 16 (20), 4028-4035.
74. Håkansson, K. M. O.; Fall, A. B.; Lundell, F.; Yu, S.; Krywka, C.; Roth, S. V.;
Santoro, G.; Kvick, M.; Prahl Wittberg, L.; Wågberg, L.; Söderberg, L. D.,
Hydrodynamic alignment and assembly of nanofibrils resulting in strong
cellulose filaments. Nature Communications 2014, 5, 4018.
75. Han, Y.; Alsayed, A. M.; Nobili, M.; Zhang, J.; Lubensky, T. C.; Yodh, A. G.,
Brownian Motion of an Ellipsoid. Science 2006, 314 (5799), 626-630.
REFERENCES
79
76. Tanaka, R.; Kuribayashi, T.; Ogawa, Y.; Saito, T.; Isogai, A.; Nishiyama, Y.,
Ensemble evaluation of polydisperse nanocellulose dimensions: rheology,
electron microscopy, X-ray scattering and turbidimetry. Cellulose 2017, 24 (8),
3231-3242.
77. Lopez, C. G.; Watanabe, T.; Martel, A.; Porcar, L.; Cabral, J. T., Microfluidic-
SANS: flow processing of complex fluids. Scientific Reports 2015, 5, 7727.
78. Tang, H.; Butchosa, N.; Zhou, Q., A Transparent, Hazy, and Strong
Macroscopic Ribbon of Oriented Cellulose Nanofibrils Bearing Poly(ethylene
glycol). Advanced Materials 2015, 27 (12), 2070-2076.
79. Adamcik, J.; Jung, J.-M.; Flakowski, J.; De Los Rios, P.; Dietler, G.; Mezzenga,
R., Understanding amyloid aggregation by statistical analysis of atomic force
microscopy images. Nature Nanotechnology 2010, 5, 423.
80. Adamcik, J.; Berquand, A.; Mezzenga, R., Single-step direct measurement of
amyloid fibrils stiffness by peak force quantitative nanomechanical atomic force
microscopy. Applied Physics Letters 2011, 98 (19), 193701.
81. Hedhammar, M.; Rising, A.; Grip, S.; Martinez, A. S.; Nordling, K.; Casals, C.;
Stark, M.; Johansson, J., Structural Properties of Recombinant Nonrepetitive
and Repetitive Parts of Major Ampullate Spidroin 1 from Euprosthenops
australis: Implications for Fiber Formation. Biochemistry 2008, 47 (11), 3407-
3417.
82. Hedhammar, M.; Bramfeldt, H.; Baris, T.; Widhe, M.; Askarieh, G.; Nordling,
K.; Aulock, S. v.; Johansson, J., Sterilized Recombinant Spider Silk Fibers of
Low Pyrogenicity. Biomacromolecules 2010, 11 (4), 953-959.
83. Jansson, R.; Thatikonda, N.; Lindberg, D.; Rising, A.; Johansson, J.; Nygren, P.
Å.; Hedhammar, M., Recombinant Spider Silk Genetically Functionalized with
Affinity Domains. Biomacromolecules 2014, 15 (5), 1696-1706.
84. Buffet, A.; Rothkirch, A.; Döhrmann, R.; Körstgens, V.; Abul Kashem, M. M.;
Perlich, J.; Herzog, G.; Schwartzkopf, M.; Gehrke, R.; Müller-Buschbaum, P.;
Roth, S. V., P03, the microfocus and nanofocus X-ray scattering (MiNaXS)
beamline of the PETRA III storage ring: the microfocus endstation. Journal of
synchrotron radiation 2012, 19 (Pt 4), 647-653.
85. Håkansson, K. M.; Fall, A. B.; Lundell, F.; Yu, S.; Krywka, C.; Roth, S. V.;
Santoro, G.; Kvick, M.; Wittberg, L. P.; Wågberg, L., Hydrodynamic alignment
and assembly of nanofibrils resulting in strong cellulose filaments. Nature
Communications 2014, 5, 4018.
86. van Gurp, M., The use of rotation matrices in the mathematical description of
molecular orientations in polymers. Colloid and Polymer Science 1995, 273 (7),
607-625.
87. Doi, M.; Edwards, S. F., The theory of polymer dynamics. Oxford University
Press: Oxford, 1988.
88. Marrucci, G.; Grizzuti, N., The effect of polydispersity on rotational diffusivity
and shear viscosity of rodlike polymers in concentrated solutions. Journal of
Polymer Science: Polymer Letters Edition 1983, 21 (2), 83-86.
89. Marrucci, G.; Grizzuti, N., Predicted effect of polydispersity on rodlike polymer
behaviour in concentrated solutions. Journal of Non-Newtonian Fluid
Mechanics 1984, 14, 103-119.
REFERENCES
80
90. Fall, A. B.; Lindström, S. B.; Sundman, O.; Ödberg, L.; Wågberg, L., Colloidal
Stability of Aqueous Nanofibrillated Cellulose Dispersions. Langmuir 2011, 27
(18), 11332-11338.
91. Nordenström, M.; Fall, A.; Nyström, G.; Wågberg, L., Formation of Colloidal
Nanocellulose Glasses and Gels. Langmuir 2017, 33 (38), 9772-9780.
92. Henriksson, M.; Berglund, L. A.; Isaksson, P.; Lindström, T.; Nishino, T.,
Cellulose Nanopaper Structures of High Toughness. Biomacromolecules 2008,
9 (6), 1579-1585.
93. Håkansson, K. M. O.; Fall, A. B.; Lundell, F.; Yu, S.; Krywka, C.; Roth, S. V.;
Santoro, G.; Kvick, M.; Prahl Wittberg, L.; Wågberg, L.; Söderberg, L. D.,
Hydrodynamic alignment and assembly of nanofibrils resulting in strong
cellulose filaments. Nat Commun 2014, 5.
94. Benítez, A. J.; Torres-Rendon, J.; Poutanen, M.; Walther, A., Humidity and
Multiscale Structure Govern Mechanical Properties and Deformation Modes in
Films of Native Cellulose Nanofibrils. Biomacromolecules 2013, 14 (12), 4497-
4506.
95. Nyström, G.; Marais, A.; Karabulut, E.; Wågberg, L.; Cui, Y.; Hamedi, M. M.,
Self-assembled three-dimensional and compressible interdigitated thin-film
supercapacitors and batteries. Nature Communications 2015, 6, 7259.
96. Benítez, A. J.; Walther, A., Cellulose nanofibril nanopapers and bioinspired
nanocomposites: a review to understand the mechanical property space. Journal
of Materials Chemistry A 2017, 5 (31), 16003-16024.
97. Keckes, J.; Burgert, I.; Frühmann, K.; Müller, M.; Kölln, K.; Hamilton, M.;
Burghammer, M.; Roth, S. V.; Stanzl-Tschegg, S.; Fratzl, P., Cell-wall recovery
after irreversible deformation of wood. Nature Materials 2003, 2, 810.
98. Zhu, H.; Zhu, S.; Jia, Z.; Parvinian, S.; Li, Y.; Vaaland, O.; Hu, L.; Li, T.,
Anomalous scaling law of strength and toughness of cellulose nanopaper.
Proceedings of the National Academy of Sciences 2015, 112 (29), 8971-8976.
99. Keten, S.; Xu, Z.; Ihle, B.; Buehler, M. J., Nanoconfinement controls stiffness,
strength and mechanical toughness of β-sheet crystals in silk. Nature Materials
2010, 9, 359.
100. Behabtu, N.; Young, C. C.; Tsentalovich, D. E.; Kleinerman, O.; Wang, X.; Ma,
A. W. K.; Bengio, E. A.; ter Waarbeek, R. F.; de Jong, J. J.; Hoogerwerf, R. E.;
Fairchild, S. B.; Ferguson, J. B.; Maruyama, B.; Kono, J.; Talmon, Y.; Cohen,
Y.; Otto, M. J.; Pasquali, M., Strong, Light, Multifunctional Fibers of Carbon
Nanotubes with Ultrahigh Conductivity. Science 2013, 339 (6116), 182-186.
101. Xin, G.; Yao, T.; Sun, H.; Scott, S. M.; Shao, D.; Wang, G.; Lian, J., Highly
thermally conductive and mechanically strong graphene fibers. Science 2015,
349 (6252), 1083-1087.
102. Xu, T.; Farris, R. J., Comparative studies of ultra high molecular weight
polyethylene fiber reinforced composites. Polymer Engineering & Science
2007, 47 (10), 1544-1553.
103. Kaufman, G.; Boltyanskiy, R.; Nejati, S.; Thiam, A. R.; Loewenberg, M.;
Dufresne, E. R.; Osuji, C. O., Single-step microfluidic fabrication of soft
monodisperse polyelectrolyte microcapsules by interfacial complexation. Lab
on a Chip 2014, 14 (18), 3494-3497.
REFERENCES
81
104. Kaufman, G.; Mukhopadhyay, S.; Rokhlenko, Y.; Nejati, S.; Boltyanskiy, R.;
Choo, Y.; Loewenberg, M.; Osuji, C. O., Highly stiff yet elastic microcapsules
incorporating cellulose nanofibrils. Soft Matter 2017, 13 (15), 2733-2737.
105. Mittal, N.; Cohen, C.; Bibette, J.; Bremond, N., Dynamics of step-
emulsification: From a single to a collection of emulsion droplet generators.
Physics of Fluids 2014, 26 (8), 082109.
106. Garstecki, P.; Fuerstman, M. J.; Stone, H. A.; Whitesides, G. M., Formation of
droplets and bubbles in a microfluidic T-junction—scaling and mechanism of
break-up. Lab on a Chip 2006, 6 (3), 437-446.
107. Anna, S. L.; Bontoux, N.; Stone, H. A., Formation of dispersions using “flow
focusing” in microchannels. Applied Physics Letters 2003, 82 (3), 364-366.
108. Kazemzadeh, A.; Eriksson, A.; Madou, M.; Russom, A., A micro-dispenser for
long-term storage and controlled release of liquids. Nature Communications
2019, 10 (1), 189.
109. Gowda. V, K.; Brouzet, C.; Lefranc, T.; Söderberg, L. D.; Lundell, F. Effective
interfacial tension in flow-focusing of colloidal dispersions: 3D numerical
simulations and experiments arXiv e-prints [Online], 2019.
https://ui.adsabs.harvard.edu/\#abs/2019arXiv190108939G (accessed January
01, 2019).
110. Kinahan, M. E.; Filippidi, E.; Köster, S.; Hu, X.; Evans, H. M.; Pfohl, T.;
Kaplan, D. L.; Wong, J., Tunable Silk: Using Microfluidics to Fabricate Silk
Fibers with Controllable Properties. Biomacromolecules 2011, 12 (5), 1504-
1511.
111. Håkansson, K. M. O., Online determination of anisotropy during cellulose
nanofibril assembly in a flow focusing device. RSC Advances 2015, 5 (24),
18601-18608.
112. Rosén, T.; Mittal, N.; Roth, S. V.; Zhang, P.; Söderberg, L. D.; Lundell, F.
Dynamic characterization of cellulose nanofibrils in sheared and extended semi-
dilute dispersions arXiv e-prints [Online], 2018.
https://ui.adsabs.harvard.edu/\#abs/2018arXiv180107558R (accessed January
01, 2018).
113. Rogers, S. S.; Venema, P.; Sagis, L. M. C.; van der Linden, E.; Donald, A. M.,
Measuring the Length Distribution of a Fibril System: A Flow Birefringence
Technique Applied to Amyloid Fibrils. Macromolecules 2005, 38 (7), 2948-
2958.
114. Celzard, A.; Fierro, V.; Kerekes, R., Flocculation of cellulose fibers: new
comparison of crowding factor with percolation and effective-medium theories.
Cellulose 2009, 16 (6), 983.
115. Broedersz, C. P.; Storm, C.; MacKintosh, F. C., Nonlinear Elasticity of
Composite Networks of Stiff Biopolymers with Flexible Linkers. Physical
Review Letters 2008, 101 (11), 118103.
116. Brouzet, C.; Mittal, N.; Lundell, F.; Söderberg, D. Characterizing the
Orientational and Network Dynamics of Polydisperse Nanofibers at the
Nanoscale arXiv e-prints [Online], 2018.
https://ui.adsabs.harvard.edu/\#abs/2018arXiv181207663B (accessed
December 01, 2018).
REFERENCES
82
117. Derakhshandeh, B.; Kerekes, R. J.; Hatzikiriakos, S. G.; Bennington, C. P. J.,
Rheology of pulp fiber suspensions: A critical review. Chemical Engineering
Science 2011, 66 (15), 3460-3470.
118. Keep, G. T.; Pecora, R., Reevaluation of the dynamic model for rotational
diffusion of thin, rigid rods in semidilute solution. Macromolecules 1985, 18
(6), 1167-1173.
119. Nechyporchuk, O.; Håkansson, K. M. O.; Gowda.V, K.; Lundell, F.; Hagström,
B.; Köhnke, T., Continuous Assembly of Cellulose Nanofibrils and
Nanocrystals into Strong Macrofibers through Microfluidic Spinning. Advanced
Materials Technologies 2019, 0 (0), 1800557.
120. Nilsson, B.; Moks, T.; Jansson, B.; Abrahmsén, L.; Elmblad, A.; Holmgren, E.;
Henrichson, C.; Jones, T. A.; Uhlén, M., A synthetic IgG-binding domain based
on staphylococcal protein A. Protein Engineering 1987, 1 (2), 107-113.
121. Widhe, M.; Shalaly, N. D.; Hedhammar, M., A fibronectin mimetic motif
improves integrin mediated cell biding to recombinant spider silk matrices.
Biomaterials 2016, 74, 256-266.
122. Malho, J. M.; Arola, S.; Laaksonen, P.; Szilvay, G. R.; Ikkala, O.; Linder, M.
B., Modular Architecture of Protein Binding Units for Designing Properties of
Cellulose Nanomaterials. Angewandte Chemie International Edition 2015, 54
(41), 12025-12028.
123. Aulin, C.; Salazar-Alvarez, G.; Lindstrom, T., High strength, flexible and
transparent nanofibrillated cellulose-nanoclay biohybrid films with tunable
oxygen and water vapor permeability. Nanoscale 2012, 4 (20), 6622-6628.
124. Tian, D.; Li, T.; Zhang, R.; Wu, Q.; Chen, T.; Sun, P.; Ramamoorthy, A.,
Conformations and Intermolecular Interactions in Cellulose/Silk Fibroin Blend
Films: A Solid-State NMR Perspective. The Journal of Physical Chemistry B
2017, 121 (25), 6108-6116.
125. Barth, A., Infrared spectroscopy of proteins. Biochimica et Biophysica Acta
(BBA) - Bioenergetics 2007, 1767 (9), 1073-1101.
126. Foo, C. W. P.; Bini, E.; Hensman, J.; Knight, D. P.; Lewis, R. V.; Kaplan, D.
L., Role of pH and charge on silk protein assembly in insects and spiders.
Applied Physics A 2006, 82 (2), 223-233.
127. Hajian, A.; Lindström, S. B.; Pettersson, T.; Hamedi, M. M.; Wågberg, L.,
Understanding the Dispersive Action of Nanocellulose for Carbon
Nanomaterials. Nano Letters 2017, 17 (3), 1439-1447.
128. Lin, N.; Dufresne, A., Nanocellulose in biomedicine: Current status and future
prospect. European Polymer Journal 2014, 59, 302-325.
129. Torres-Rendon, J. G.; Köpf, M.; Gehlen, D.; Blaeser, A.; Fischer, H.; Laporte,
L. D.; Walther, A., Cellulose Nanofibril Hydrogel Tubes as Sacrificial
Templates for Freestanding Tubular Cell Constructs. Biomacromolecules 2016,
17 (3), 905-913.
130. Lou, Y.-R.; Kanninen, L.; Kuisma, T.; Niklander, J.; Noon, L. A.; Burks, D.;
Urtti, A.; Yliperttula, M., The Use of Nanofibrillar Cellulose Hydrogel As a
Flexible Three-Dimensional Model to Culture Human Pluripotent Stem Cells.
Stem Cells and Development 2014, 23 (4), 380-392.
REFERENCES
83
131. Bhattacharya, M.; Malinen, M. M.; Lauren, P.; Lou, Y.-R.; Kuisma, S. W.;
Kanninen, L.; Lille, M.; Corlu, A.; GuGuen-Guillouzo, C.; Ikkala, O.;
Laukkanen, A.; Urtti, A.; Yliperttula, M., Nanofibrillar cellulose hydrogel
promotes three-dimensional liver cell culture. Journal of Controlled Release
2012, 164 (3), 291-298.
132. Mertaniemi, H.; Escobedo-Lucea, C.; Sanz-Garcia, A.; Gandía, C.; Mäkitie, A.;
Partanen, J.; Ikkala, O.; Yliperttula, M., Human stem cell decorated
nanocellulose threads for biomedical applications. Biomaterials 2016, 82, 208-
220.
133. Bettinger, C. J.; Langer, R.; Borenstein, J. T., Engineering Substrate
Topography at the Micro- and Nanoscale to Control Cell Function. Angewandte
Chemie International Edition 2009, 48 (30), 5406-5415.
134. Hamada, D.; Dobson, C. M., A kinetic study of β-lactoglobulin amyloid fibril
formation promoted by urea. Protein Science 2002, 11 (10), 2417-2426.
135. Gosal, W. S.; Clark, A. H.; Ross-Murphy, S. B., Fibrillar β-Lactoglobulin Gels:
Part 1. Fibril Formation and Structure. Biomacromolecules 2004, 5 (6), 2408-
2419.
136. Veerman, C.; Ruis, H.; Sagis, L. M. C.; van der Linden, E., Effect of
Electrostatic Interactions on the Percolation Concentration of Fibrillar β-
Lactoglobulin Gels. Biomacromolecules 2002, 3 (4), 869-873.
137. Bolisetty, S.; Harnau, L.; Jung, J.-m.; Mezzenga, R., Gelation, Phase Behavior,
and Dynamics of β-Lactoglobulin Amyloid Fibrils at Varying Concentrations
and Ionic Strengths. Biomacromolecules 2012, 13 (10), 3241-3252.
138. Jordens, S.; Isa, L.; Usov, I.; Mezzenga, R., Non-equilibrium nature of two-
dimensional isotropic and nematic coexistence in amyloid fibrils at liquid
interfaces. Nature Communications 2013, 4, 1917.
139. Ling, S.; Li, C.; Adamcik, J.; Shao, Z.; Chen, X.; Mezzenga, R., Modulating
Materials by Orthogonally Oriented β-Strands: Composites of Amyloid and Silk
Fibroin Fibrils. Advanced Materials 2014, 26 (26), 4569-4574.
140. Davis, V. A.; Parra-Vasquez, A. N. G.; Green, M. J.; Rai, P. K.; Behabtu, N.;
Prieto, V.; Booker, R. D.; Schmidt, J.; Kesselman, E.; Zhou, W.; Fan, H.;
Adams, W. W.; Hauge, R. H.; Fischer, J. E.; Cohen, Y.; Talmon, Y.; Smalley,
R. E.; Pasquali, M., True solutions of single-walled carbon nanotubes for
assembly into macroscopic materials. Nature Nanotechnology 2009, 4, 830.
141. Dzenis, Y., Spinning Continuous Fibers for Nanotechnology. Science 2004, 304
(5679), 1917-1919.
142. Vigolo, B.; Pénicaud, A.; Coulon, C.; Sauder, C.; Pailler, R.; Journet, C.;
Bernier, P.; Poulin, P., Macroscopic Fibers and Ribbons of Oriented Carbon
Nanotubes. Science 2000, 290 (5495), 1331-1334.
143. Vigderman, L.; Khanal, B. P.; Zubarev, E. R., Functional Gold Nanorods:
Synthesis, Self-Assembly, and Sensing Applications. Advanced Materials 2012,
24 (36), 4811-4841.
144. Rammensee, S.; Slotta, U.; Scheibel, T.; Bausch, A. R., Assembly mechanism
of recombinant spider silk proteins. Proceedings of the National Academy of
Sciences 2008, 105 (18), 6590-6595.
REFERENCES
84
145. Stark, M.; Grip, S.; Rising, A.; Hedhammar, M.; Engström, W.; Hjälm, G.;
Johansson, J., Macroscopic Fibers Self-Assembled from Recombinant
Miniature Spider Silk Proteins. Biomacromolecules 2007, 8 (5), 1695-1701.
146. Benselfelt, T.; Engström, J.; Wågberg, L., Supramolecular double networks of
cellulose nanofibrils and algal polysaccharides with excellent wet mechanical
properties. Green Chemistry 2018, 20 (11), 2558-2570.
147. Iwamoto, S.; Isogai, A.; Iwata, T., Structure and Mechanical Properties of Wet-
Spun Fibers Made from Natural Cellulose Nanofibers. Biomacromolecules
2011, 12 (3), 831-836.
148. Chen, S.; Schueneman, G.; Pipes, R. B.; Youngblood, J.; Moon, R. J., Effects
of Crystal Orientation on Cellulose Nanocrystals–Cellulose Acetate
Nanocomposite Fibers Prepared by Dry Spinning. Biomacromolecules 2014, 15
(10), 3827-3835.
149. Torres-Rendon, J. G.; Schacher, F. H.; Ifuku, S.; Walther, A., Mechanical
Performance of Macrofibers of Cellulose and Chitin Nanofibrils Aligned by
Wet-Stretching: A Critical Comparison. Biomacromolecules 2014, 15 (7),
2709-2717.
150. Walther, A.; Timonen, J. V.; Díez, I.; Laukkanen, A.; Ikkala, O.,
Multifunctional high‐performance biofibers based on wet‐extrusion of
renewable native cellulose nanofibrils. Advanced Materials 2011, 23 (26),
2924-2928.
151. Lundahl, M. J.; Cunha, A. G.; Rojo, E.; Papageorgiou, A. C.; Rautkari, L.;
Arboleda, J. C.; Rojas, O. J., Strength and Water Interactions of Cellulose I
Filaments Wet-Spun from Cellulose Nanofibril Hydrogels. Scientific Reports
2016, 6, 30695
152. Bledzki, A. K.; Gassan, J., Composites reinforced with cellulose based fibers.
Progress in Polymer Science 1999, 24 (2), 221-274.
153. Li, X.; Tabil, L. G.; Panigrahi, S., Chemical Treatments of Natural Fiber for Use
in Natural Fiber-Reinforced Composites: A Review. Journal of Polymers and
the Environment 2007, 15 (1), 25-33.
154. Xia, X.-X.; Qian, Z.-G.; Ki, C. S.; Park, Y. H.; Kaplan, D. L.; Lee, S. Y., Native-
sized recombinant spider silk protein produced in metabolically engineered
Escherichia coli results in a strong fiber. Proceedings of the National Academy
of Sciences 2010, 107 (32), 14059-14063.
155. Elices, M.; Guinea, G. V.; Plaza, G. R.; Karatzas, C.; Riekel, C.; Agulló-Rueda,
F.; Daza, R.; Pérez-Rigueiro, J., Bioinspired Fibers Follow the Track of Natural
Spider Silk. Macromolecules 2011, 44 (5), 1166-1176.
156. Plaza, G. R.; Corsini, P.; Marsano, E.; Pérez-Rigueiro, J.; Biancotto, L.; Elices,
M.; Riekel, C.; Agulló-Rueda, F.; Gallardo, E.; Calleja, J. M.; Guinea, G. V.,
Old Silks Endowed with New Properties. Macromolecules 2009, 42 (22), 8977-
8982.
157. Guan, J.; Porter, D.; Vollrath, F., Silks cope with stress by tuning their
mechanical properties under load. Polymer 2012, 53 (13), 2717-2726.
158. Inacio, W. P.; Lopes, F. P. D.; Monteiro, S. N., Diameter dependence of tensile
strength by Weibull analysis: Part III sisal fiber. Matéria (Rio de Janeiro) 2010,
15, 124-130.
REFERENCES
85
159. Paul, A.; Thomas, S., Electrical properties of natural-fiber-reinforced low
density polyethylene composites: A comparison with carbon black and glass-
fiber-filled low density polyethylene composites. Journal of Applied Polymer
Science 1997, 63 (2), 247-266.
160. Kalia, S.; Kaith, B. S.; Kaur, I., Pretreatments of natural fibers and their
application as reinforcing material in polymer composites—A review. Polymer
Engineering & Science 2009, 49 (7), 1253-1272.
161. Liu, Y.; Sponner, A.; Porter, D.; Vollrath, F., Proline and Processing of Spider
Silks. Biomacromolecules 2008, 9 (1), 116-121.
162. Cunniff, P. M.; Fossey, S. A.; Auerbach, M. A.; Song, J. W.; Kaplan, D. L.;
Adams, W. W.; Eby, R. K.; Mahoney, D.; Vezie, D. L., Mechanical and thermal
properties of dragline silk from the spider Nephila clavipes. Polymers for
Advanced Technologies 1994, 5 (8), 401-410.
163. Wambua, P.; Ivens, J.; Verpoest, I., Natural fibers: can they replace glass in
fiber reinforced plastics? Composites Science and Technology 2003, 63 (9),
1259-1264.
164. Ashby, M. F.; Gibson, L. J.; Wegst, U.; Olive, R., The mechanical properties of
natural materials. I. Material property charts. Proceedings of the Royal Society
of London. Series A: Mathematical and Physical Sciences 1995, 450 (1938),
123-140.
APPENDIX
86
Appendix
Table 1. Comparison of mechanical properties of our CNF fibers (paper I) with the
previously reported studies on the fabrication of nanocellulose based fibers using
different approaches.
Fiber Materials Strength*
(MPa)
Stiffness*
(GPa)
Year
(Ref.)
Cellulose
nanofibers
Cellulose
nanofibers
306
321
15.3
23.6
2016132
2011147
Cellulose
nanofibers
222 12.6 201568
Cellulose
nanocrystal-
cellulose acetate
220 16 2014148
Nanofibrillated
cellulose
323 37.7 2014149
Cellulose
nanofibrils
275 22.5 2011150
Cellulose
nanofibrils
576 18.8 201485
Cellulose
nanofibrils
Cellulose
nanofibrils
297
850
21
53.5
2016151
201734
Cellulose
nanofibrils
1570 86 Present
work
(paper I)
*Strength and stiffness reported here corresponds to the highest values obtained in
the reference studies. Terminologies used for CNFs are the ones reported in the
source articles.
APPENDIX
87
Table 2. Source references for data points used in Figure 4.9 and 4.10.
Fiber Reference
Coir Bledzki et al.152
Cotton Li et al.153
RSP Xia et al154; Elices et al.155; Plaza et al.156
Nylon Guan et al.157
Sisal Inacio et al.158
Jute Paul et al159; Bledzki et al.152
Hemp Li et al.153
Ramie Kalia et al.160
Dragline Liu et al.161; Cunniff et al.162
E-Glass Li et al.153; Wambua et al.163
S-Glass Li et al.153
Metals &
Alloys
http://www.grantadesign.com/download/pdf/teaching_resource_b
ooks/2-Materials-Charts-2010.pdf
*Additional data points are adapted from Ashby and Wegst.164