Development of Free-Standing Interference Films for Paper and Packaging Applications

M

S

T

DEVELOPMENT OF FREE-STANDING INTERFERENCE

FILMS FOR PAPER AND PACKAGING APPLICATIONS

Johan Holmqvist

Executed at STFI-Packforsk AB, Stockholm - Sweden

2008-03-03

LITH-IFM-EX--08/1920—SE

Linköpings universitet Institutionen för Fysik, Kemi och Biologi 581 83 Linköping

Rapporttyp Report category Licentiatavhandling x Examensarbete C-uppsats D-uppsats Övrig rapport _______________ Språk Language Svenska/Swedish Engelska/English ________________ Titel Title

UTVECKLING AV FRISTÅENDE INTERFERENSTUNNFILMER FÖR PAPPERS- OCH PAKETERINGSTILLÄMPNINGAR

DEVELOPMENT OF FREE-STANDING INTERFERENCE FILMS FOR PAPER AND PACKAGING APPLICATIONS Författare Author Johan Holmqvist Sammanfattning Abstract

The newfound capability of creating moisture sensitive interference multilayered thin films (MLTFs) comprising microfibrillated cellulose and polymers has not previously been possible to implement on surfaces other than silicon wafer strips. Being able to incorporate interference MLTFs on fibre-based materials would introduce the possibility for new applications within authentication, sensing and customer attraction for the paper and packaging industry. By using trichloro (1H, 1H, 2H, 2H-perfluorooctyl) silane we were able to hydrophobically modify silicon substrates, enabling interference MLTF lift-off and thus the creation of free-standing MLTFs of approximately 400 nm thickness. Contact dried MLTFs approximately 250 nm thick, were successfully transferred to copy- and filter paper as well as to cellulose-based dialysis membranes. We can also report on the successful synthesis of interference MLTFs directly on a fibre composite material and on aluminium. Initial tests of a method to quantify the pull-off conditions of the MLTFs from the fluorinated surfaces using the Micro Adhesion Measurement Apparatus showed promising results.

ISRN:

LiTH-IFM-EX-08/1920-SE

_____________________________________________

Serietitel och serienummer ISSN

Title of series, numbering

Nyckelord

Keyword

Layer-by-layer, interference thin film, free-standing, moisture sensor, polyelectrolyte, surface self-2008-03-03

URL för elektronisk version

http://urn.kb.se/resolve?urn=urn:nbn: se:liu:diva-11398

Applied Physics

Department of Physics, Chemistry and Biology Linköping University

E

U

PAPPERS

-

Johan Holmqvist

Examensarbete utfört vid STFI-Packforsk AB, Stockholm - Sverige

2008-03-03

Handledare

Sven Forsberg – STFI-Packforsk AB

Hjalmar Granberg – STFI-Packforsk AB

Lars Wågberg – Kungliga Tekniska Högskolan

Examinator

Kajsa Uvdal – Linköpings Universitet

Opponent

från publiceringsdatum under förutsättning att inga extraordinära omständigheter uppstår. Tillgång till dokumentet innebär tillstånd för var och en att läsa, ladda ner, skriva ut enstaka kopior för enskilt bruk och att använda det oförändrat för ickekommersiell forskning och för undervisning. Överföring av upphovsrätten vid en senare tidpunkt kan inte upphäva detta tillstånd. All annan användning av dokumentet kräver upphovsmannens medgivande. För att garantera äktheten, säkerheten och tillgängligheten finns lösningar av teknisk och administrativ art.

Upphovsmannens ideella rätt innefattar rätt att bli nämnd som upphovsman i den omfattning som god sed kräver vid användning av dokumentet på ovan beskrivna sätt samt skydd mot att dokumentet ändras eller presenteras i sådan form eller i sådant sammanhang som är kränkande för upphovsmannens litterära eller konstnärliga anseende eller egenart. För ytterligare information om Linköping University Electronic Press se förlagets hemsida http://www.ep.liu.se/.

C

The publishers will keep this document online on the Internet – or its possible replacement – for a period of 25 years starting from the date of publication barring exceptional circumstances.

The online availability of the document implies permanent permission for anyone to read, to download, or to print out single copies for his/her own use and to use it unchanged for non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional upon the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility.

According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement. For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its www home page: http://www.ep.liu.se/.

D

EVELOPMENT OF FREE

-

STANDING

INTERFERENCE FILMS FOR PAPER AND

PACKAGING APPLICATIONS

M

S

T

Johan Holmqvist

Cluster:

Dedicated to and in memory of Arne and Astrid Bråten

A

The newfound capability of creating moisture sensitive interference multilayered thin films (MLTFs) comprising microfibrillated cellulose and polymers has not previously been possible to implement on surfaces other than silicon wafer strips. Being able to incorporate interference MLTFs on fibre-based materials would introduce the possibility for new applications within authentication, sensing and customer attraction for the paper and packaging industry. By using trichloro (1H, 1H, 2H, 2H-perfluorooctyl) silane we were able to hydrophobically modify silicon substrates, enabling interference MLTF lift-off and thus the creation of free-standing MLTFs of approximately 400 nm thickness. Contact dried MLTFs approximately 250 nm thick, were successfully transferred to copy- and filter paper as well as to cellulose-based dialysis membranes. We can also report on the successful synthesis of interference MLTFs directly on a fibre composite material and on aluminium. Initial tests of a method to quantify the pull-off conditions of the MLTFs from the fluorinated surfaces using the Micro Adhesion Measurement Apparatus showed promising results.

A

During the entire process of this thesis Hjalmar Granberg and Sven Forsberg of STFI-PACKFORSK AB as well as Lars Wågberg of the Royal Institute of Technology have all continuously supervised and challenged me towards becoming more proficient in my areas of competence. I am truly grateful for having had such a competent and supporting team of supervisors.

Thank you Kajsa Uvdal, Linköpings Institute of Technology, for being the examiner of this thesis, and thank you Ida Hederström for providing the opposition.

Furthermore I would like to thank STFI-PACKFORSK AB for financial support and for trusting me with this task.

Thank you Mikael Ankerfors and colleagues, STFI-Packforsk AB, for providing the microfibrillated cellulose needed, and for the support on this subject.

I would also like to thank everyone at the department of fibre- and polymer technology for sharing numerous tips, laughs and discussions with me, making me feel as a member of the team, starting day one. Lars-Erik, thanks for all of your support regarding just about everything, invaluable!! Erik, thanks for all of your help regarding the MAMA. Oskar, thank you for the help regarding the contact-angle measurements. Christian, your advice regarding the MFC/PEI system has been of great value. Caroline, thanks for being KTH-Caroline ☺.

Mom, Dad, Camilla, Tom-Kjetil: Thank you for everything… Helena - I Love You

L

A

DLS – Dynamic Lights Scattering, (apparatus/method) JKR – Johnson, Kendall, Roberts, (adhesion model/MAMA) LbL – Layer by Layer, (method)

MAMA – Micro Adhesion Measurement Apparatus, (apparatus/method) MFC – Microfibrillated Cellulose, (film constituent)

MLTF – Multilayered Thin Film

PDADMAC – poly(diallyl-dimethyl-ammoniumchloride), (film constituent) PDMS – poly(dimethyl siloxane), (MAMA-probe constituent)

PEI – poly(ethyleneimine), (film constituent)

PFOS – trichloro(1H,1H,2H,2H perfluorooctyl)silane, (SAM constituent) PSS – poly(sodium 4-styrenesulfonate), (film constituent)

SAM – self assembled monolayer

T

1 INTRODUCTION... 1

-1.1 Problem Statement ... 2

-1.2 Objectives ... 3

-1.3 Outline and Structure ... 3

-2 THEORY... 5

-2.1 The Layerbylayer Technique ... 5

-2.2 Interference in Thin Films ... 6

-2.3 Dynamic Light Scattering... 8

-2.4 Release Techniques ... 9

-2.4.1 Dissolving a Sacrificial Layer by an Organic Solvent... 9

-2.4.2 The use of Fluorinated Surfaces... 10

-2.4.3 Release through pH Sensitive Disintegration of a Sacrificial Layer... 12

-2.4.4 Dissolving the Substrate: ... 13

-2.4.5 Electrochemical Manipulation of Surface Charge (hypothesis)... 14

-2.4.6 Our Selected Strategy... 15

-2.5 FluoroSilanization... 16

-2.6 Micro Adhesion Measurement Apparatus, (MAMA) ... 19

-2.6.1 The Instrument ... 20 -2.6.2 MAMA Procedure... 21 -3 EXPERIMENTAL... 25 -3.1 Materials... 25 -3.1.1 Miscellaneous:... 25 -3.1.2 Substrates: ... 25 -3.1.3 Polyelectrolytes: ... 25 -3.1.4 Microfibrillated Cellulose:... 26 -3.1.5 Silanisation: ... 26

-3.1.6 Micro Adhesion Measurement Apparatus, (MAMA) ... 26

-3.2 Instruments ... 26

-3.3 Methods & Laborative Setups ... 27

-3.3.1 Preparation of Polyelectrolyte Solutions ... 27

-3.3.2 Preparation of the Microfibrillated Cellulose ... 27

-3.3.3 Preparation of Silicon Wafer Slides ... 28

-3.3.4 Fluorosilanisation ... 28

-3.3.5 Preparation of Spraypainted FibreComposite (Kofesdemonstrator) ... 28

-3.3.6 Preparation of Aluminium strips... 29

-3.3.7 Manual LayerbyLayer Procedure ... 29

-3.3.8 Automated LayerbyLayer using the Nanostrata Stratosequence ... 29

-3.3.9 The Prepared Films ... 30

-3.4 Micro Adhesion Measurement Apparatus (MAMA)... 31

-3.4.1 Preparative ... 31

-4 RESULTS AND DISCUSSION... 33 -4.1 Preparative ... 33 -4.1.1 Material Characterisation ... 33 -4.1.2 Contact Angle Verification of Hydrophobicity of Fluorinated Substrates ... 36 -4.2 Freestanding Films ... 37 -4.3 Transfer to New Carrier Materials ... 41 -4.4 Interference Film Synthesis directly on Aluminium and Fibrebased Substrates .. 44 -4.5 Micro Adhesion Measurement Apparatus (MAMA)... 47 -4.5.1 Successful Release Using MAMA Pulloff... 47 -4.5.2 Evaluation of the MAMA Experiments regarding Pulloff. ... 53 -4.6 Observations... 54 -4.6.1 (Microfibrillated Cellulose | polyEthyleneimine) – Gels ... 54 -4.6.2 The Colour Gradient at the Edge of a MLTF... 54

-5 SUGGESTIONS FOR FURTHER WORK AND APPLICATIONS... 57 -6 CONCLUSIONS... 59 -7 REFERENCES... 61 -8 LIST OF FIGURES AND TABLES... 65 -9 STFI-PACKFORSK DATABASE INFORMATION... 75

-1 I

Recent technological advances in the nano-material field have widely broadened the horizon on the development of interesting new materials. With either novel or improved performance, these new materials can be implemented in the production of new products and new product-applications. Numerous factors influence the success or the failure of a product or feature comprising a novel material or function. Key aspects currently in focus of the industry, when designing new materials are the increased cost-effectiveness in the production, the added value of a product by the application or material and the environmental consequences of the production of the material or product.

Three main fields of current industrial interest and importance, with which this thesis is intimately linked are: customer attraction-, authenticity verification- and sensing-applications. One could argue that a new ability to attract a customer’s attention would be of great value to the advertisement and packaging industries. Further on, not being able to authenticate the contents of a parcel, package or other product, is a growing concern, especially as the global trade of merchandise via the internet is increasing, sometimes making it hard to know and validate that one has bought and received the same product as was ordered. Further adding to this problematic development is product piracy of bootleg copies with forged security devices. The ability to easily maintain a reliable authentication device, that is relatively hard and expensive to make bootleg copies of, could be of large value to numerous industries.

To exemplify possible future applications, one could postulate a carton of milk changing colour upon a customer touching it(customer attraction), or a pharmaceutical container, which contents can be authenticated by a specific interactive identity tag on the container, as the tag is exposed to a specific stimuli(authenticity/sensing application).

In order to meet future market demands, a new technology platform enabling prototype manufacturing of stimuli sensitive, opto-active, nano-scale, interference-films has been developed at STFI-Packforsk AB. The key feature of these films is their stimuli-induced change of colour, which makes the films, in themselves, sensors. The possibility for a film

to work as a sensor is limited to the types of stimuli that can be made to produce a signal, in this case the change of colour.

The research of these films has been focused on, but not limited to, using exhaled breath as stimuli. When exposed to an increased humidity, the films change colour. Film-uptake of water causing the film to swell is one suggested explanation for the colour shift. Other stimuli, such as mechanical pressure and heat etc, remain to be investigated, as do the possibilities to couple adsorption of specific molecules to receptors on the film surface. Antibody-antigen bonding could be one possible solution to the latter.

The observed colour and change thereof, is dependent upon the refractive indices of the chosen film materials and the surrounding media, as well as on the film thickness. These parameters can be controlled in the manufacturing process by carefully selecting the film constituents and by controlling to which extent and thickness, the film is allowed to be synthesised. Due to the fact that the films can function at an incredibly low thickness it is possible to manufacture these films maintaining a low unit-cost, using renewable sources of material in order to maintain sustainability.

1.1 Problem Statement

At present, cut strips of silicon-wafers, silicon being a well studied and readily available model surface, are used as substrate for the film synthesis. The use of silicon as substrate is advantageous in several ways. Silicon-wafers have both flat and well defined surfaces needed in order to synthesize smooth interference-films of uniform colour.

The use of silicon however also has one disadvantage. The hydrophilic nature of the silicon makes adhesion between the used film constituents (also hydrophilic) and the substrate very strong. The currently used film constituents within the frames of our research-project are micro-fibrillated cellulose and polyelectrolytes. In order to implement the suggested sensor application, the thin film needs to be easily transferable between the substrate upon which it was synthesised, and the target product or material. One possibility to circumvent this problem would be synthesising the film directly on the target product.

The problem with which this thesis will wrestle concerns the need for a developed methodology of film transfer, in order to be able to fully implement the capacities of our thin interference films.

1.2 Objectives

To start solving the problem, four areas of main focus were decided upon, as presented below.

Survey of the possibility to evaluate different substrates or methods, regarding film releaseability.

Development of a method to produce free-standing interference thin films. (Free-standing either completely or suspended in solution)

Investigation of the possibility to transfer interference thin films between the substrate used for synthesis and fibre-based materials.

Research the feasibility of synthesis of multilayered thin films directly on fibre-based materials.

1.3 Outline and Structure

To get a better understanding of the current situation regarding the research progress in the field of thin film synthesis and the release of such films from their original substrates of synthesis, we decided on carrying out a literature study. The outcome of this study, intended to enlighten us on the possibilities and difficulties accompanied with the release of thin films with thicknesses on the nano-scale, proved valuable as it suggested a plural of previously tested methods. These methods were however not directly applicable to our purpose, mainly due to two reasons. Firstly, the fact that our work focuses on the release and synthesis of interference thin films, dependent upon not exceeding a certain thickness

(section2.2). Secondly that the films whilst thin enough to show interference behaviour

regarding colouring, must also be thick enough to enable intact release or transfer of the film. Thus, we set out to modify and further develop a method for thin film release.

This thesis continues with a description of key theoretical areas needed for explaining the contents of this thesis, after which the results from the above mentioned literature study and the concluding strategies we were able to obtain from them are presented. The experimental section of this thesis is then documented and this is followed by the results section and a discussion thereof. The terminating part of this thesis is devoted to suggestions for future research areas, linked to our results, and this part is in turn followed by a concluding section.

2 T

This section will describe the key theoretical elements used throughout the thesis.

2.1 The Layer-by-layer Technique

Although initially discovered by Iler1 _{in the 1960s, the use of the Layer-by-layer (LbL)}

technique was not widely developed until the re-discovery of the technique by Decher2, 3 _in

the 1990s. Since then numerous of research groups have embraced this technique, and quite a substantial base of knowledge has been established4_.

The principle of LbL depicted by Figure 2-1 is based on the adsorption of polyions or particles to an oppositely charged substrate by alternated immersion of the substrate in solutions of these polyions. By dipping a substrate (often negatively charged silicon) that carries surface charge into a solution of oppositely charged polyions, one initial layer of polyion(A) is adsorbed. This adsorption inverts the surface charge, enabling the adsorption of a layer of oppositely charged polyion(B) to the already adsorbed layer. This again inverts the surface charge, making adsorption of polyion(A) possible yet another time. By cycled immersion, multilayered thin films (MLTFs) can be synthesized1-4_.

Figure 2-1 The LbL-assembly is illustrated. Two types of oppositely charged molecules, A and B,

are alternately adsorbed onto the substrate. By repeating the procedure, MLTFs can be synthesised.

A simple way to describe the pathway used for synthesis of a specific MLTF is labelling as (A|B)x, where the MLTF has undergone x number of dippings in polyion(A) and x number

The LbL-technique is not limited to the use of polyelectrolytes for assembly. Magnetite particles5_{, gold nanoparticles}6_{, clay platelets}7 _{and microfibrillated cellulose (MFC)} 8-11 _are

examples of other materials that have been used in the creation of MLTFs. By carefully selecting the constituents, precise control of the growth-rate of the assembled multilayers is possible, as is the synthesis of smooth MLTFs.

This thesis focuses on the multilayered thin films made from microfibrillated cellulose and oppositely charged, positive poly(ethylene-imine) (PEI) illustrated by Figure 2-2.

Figure 2-2 showing the structural formula of the PEI-molecule. As is

illustrated, the molecule is branched.

Furthermore, LbL-multilayering is not dependent upon the use of oppositely charged particles for the assembly. Hydrogen bonding is one alternative, that has been recently examined12, 13_.

2.2 Interference in Thin Films

Rays of incident reflected at various interfaces of a thin coating undergo interference that is either constructive or destructive. This gives rise to the colouring of thin film coatings given certain conditions.

The observed interference colour is dependent upon the refractive indices of the thin film and the surrounding media. It is also dependent upon the thickness of the thin film and by analogy to this, on the smoothness. The case of a substrate-bound (Figure 2-3) MLTF differs slightly from that of a free-standing (Figure 2-4) one, in that the surrounding media

in the case of a free-standing film, is the same on both sides compared to the different media observed for each side of a substrate bound MLTF.

n₂(MLTF) n₁(air) n₃(substrate) d n₁(air) n₃(substrate) d

Figure 2-3 illustrating a substrate bound MLTF. The surrounding media are air

and substrate, (indicated by their refractive indices (n1)and (n3)). The

MLTF-parameters are the thickness (d) and the refractive index (n2). (inspired by

illustration in14₎ n₁(air) n₁(air) n₂(free-standing MLTF) d n₁(air) n₁(air) n₂(free-standing MLTF) d

Figure 2-4 illustrating the somewhat different properties of a free-standing MLTF. The

medium on both sides of the MLTF is the same and has refractive index (n1). The MLTF has a

thickness of (d) and a refractive index (n2). (inspired by illustration in14)

A deeper look into the interference phenomenon is given by Halliday et al.15_{, where the}

mathematics of interference (not explained here) is described.

By observing the interference colour of our synthesized MLTFs we were able to keep track of their approximate thickness, using a MLTF interference model, previously developed at STFI-Packforsk8_{. Determination of the thickness of a MLTF is enabled given the refractive}

MLTF under daylight illumination. The use of the model is not specifically reported on in this thesis since verification of the interference phenomena is possible by ocular inspection. One important result from the modelling of the interference behaviour of both free-standing and silicon-attached MLTFs is however utilized. By employing the model to systems similar to those in this thesis with respect to refractive indices and surrounding media, interference caused colouring behaviour was predicted (verified by laborative work not included in this thesis) for MLTFs of thicknesses up to approximately 1μm. The reasons for the transparent and colourless behaviour of thicker films include that the interference of the films give rise to relatively few interference peaks when they are thin and additional peaks when they get thicker. When enough peaks are introduced they smear the ocular possibility to detect different colour resulting in the loss of apparent colour (based on the model developed by Anderson8_).

2.3 Dynamic Light Scattering

To characterize the solutions used for the LbL assembly with respect to molecular size and surface charge (estimated by zeta-potential), a Dynamic Light Scattering (DLS) equipment was used. The theory behind these measurements lies beyond the scope of this thesis, however the zeta-potential and DLS will be explained briefly.

The DLS equipment measures the diffusion of light-scattering particles or ions and the particle size is estimated through the obtained hydrodynamic radius.

Macromolecules can become charged when in an aqueous solution through ionization. This ability and the amount of induced charge depend on the functional groups incorporated in the macromolecule. Macromolecular ions attract oppositely charged counter ions present in the solution. The attracted counter ions can be divided into two types; namely the ones that are most attracted by the molecule and follow it through its motion, and the ones loosely attached, that do not stick to the molecule. This in turns gives rise to a slipping plane, defined as the border between these types of ions. At the slipping plane the electric-potential is different from that at the macromolecular surface itself. The

potential at the slipping plane is called zeta-potential, and we have used it to verify that our polyelectrolytes behave as expected in solution regarding charge16_.

2.4 Release Techniques

We decided to conduct a literature-study in order to get an understanding of what had already been researched in the field of free-standing multilayered thin films. The key outcome of this study is presented below in a condensed form. The results enabled us to construct our strategy which is explained in the continuation of this section.

In order to facilitate the release of multilayered thin films, synthesized through alternated dipping of substrates in solutions of oppositely charged adsorbates, several different techniques have been used previously. For our purpose, these established techniques offer different possible experimental designs, however all are not directly possible to implement for our cause. Our choice of method, described towards the end of this section, was influenced by some of these techniques, but offers modifications in the experimental design, largely due to the fact that our choice of material is different from those already mentioned. Another reason being that our aim not only includes the successful release of thin films, but also consists of the investigation of the possibility to transfer the thin films from the substrate to a new carrier material. Furthermore, in order to be subjects of the desired interference phenomena, the films must not exceed a certain thickness.

2.4.1 Dissolving a Sacrificial Layer by an Organic Solvent

One way to create a free-standing MLTF is to synthesize the target-film onto a substrate that has been previously coated with a sacrificial layer, i.e. a layer which in its turn is readily dissolvable in a solvent that does not affect the target film, illustrated by Figure 2-5 . After the LbL-deposition of the target film on top of the sacrificial layer, the sample is immersed in a solvent which dissolves the sacrificial layer and thereby renders the target-film released from its substrate, freely suspended in the solvent.

An already well-established application of this idea has been successfully carried out by Mamedov et al 5_{. and Tsukruk et al} 6_{, who both report the use of an assemble-dissolve} technique based on the coating of a substrate with cellulose-acetate, using its solubility in

acetone to create a free-standing film from the target film synthesized on top of the sacrificial layer.

Key features: Dissolvable sacrificial layer. Limitations:

Solvent must not affect or react with target-film.

Sacrificial layer must be readily dissolvable in a solvent that preserves an intact

target-film.

The use of organic solvents such as acetone is not environmentally advantageous.

Advantages: Preparation of very thin (~30nm) free-standing MLTF suspended in

solution is possible.5 Sacrificial layer Target-film Solvent -addition Dissolved sacrificial layer Target-film (free-standing)

Figure 2-5 The sample (left) is submerged into the solvent. The sacrificial layer

then starts to dissolve (right) due to its solubility in the solvent, thus leaving a free-standing target film.

2.4.2 The use of Fluorinated Surfaces.

In order to achieve minimal adhesion of contaminants to a surface, the use of Teflon or other fluorinated materials has been industrially implemented. The frequent use of Teflon coating in the manufacture of frying pans exemplifies this. A fluorinated surface is hydrophobic and this is different from other popular substrates for MLTF-deposition, which mainly consist of glass slides and cut strips of silicon wafers, which are both mainly hydrophilic.

By using Teflon substrates or by fluoro-silanizing a hydrophilic substrate, (often a silicon wafer-strip), one creates a possible approach in the development of synthesis of free-standing MLTFs. The idea is that while the LbL adsorption of the MLTF constituents might not proceed as efficiently as it would with a hydrophilic substrate, the synthesized MLTF will not, when finally adsorbed, adhere to the same extent to the substrate and thus, it will become easier to peel-off mechanically with for instance tweezers.

Successful implementation of this technique in order to achieve free-standing MLTFs has been reported by Lutkenhaus et al. 12 _{as well as by Jaber et al.} 17 _{Both approaches take}

advantage of a Teflon-coated surface. An illustration is provided in Figure 2-6.

Key features: Switch from hydrophilic to hydrophobic substrate to lower the

attractive force between the MLTF and the substrate.

Limitations:

Use of fluorinating agents such as fluoro-silanes needs to be evaluated from a

sustainability perspective. Though the environment needs to be considered, the use of fluoro-silanes to coat silicon-wafers is not largely material consuming, since only substrate-modification is needed.

Successful implementations of this technique have been limited to the release of

relatively thick MLTFs (about 8-9μm) 12, 17_.

Advantages: This technique offers the possibility of dry-release. Except from the

use of fluorinating agents, this procedure demands no further modification to standard LbL protocol.

CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF₃ CF3 CF3 CF3 CF3 CF3 CF3 CF3 Fluorinated surface

MLTF- weakly adhered to the hydrophobic surface

MLTF-peeled off using tweezers Silica surface

Figure 2-6 By introducing CF2 and CF3 groups to the silicon surface, the adhesion between the MLTF and

the substrate is decreased. This makes removal possible, in this figure exemplified by peeling the MLTF of using tweezers. For illustrating purposes the films are illustrated as a A-B-A pattern. The films of his thesis are multilayered i.e. (A|B)20 (not shown here).

2.4.3 Release through pH Sensitive Disintegration of a Sacrificial Layer

Yet another previously shown method to obtain a free-standing MLTF is the use of a pH-sensitive sacrificial layer, depicted by Figure 2-7. Standard LbL procedure describes the alternate use of positively and negatively charged polyelectrolytes in the MLTF synthesis, with the electrostatic interaction between these layers as main contributing factor of these forming4_{. It is however possible to generate MLTFs by other means than that of}

electrostatic interaction of the constituents.

By choosing materials that together have hydrogen bonding capability, one can generate a layered film that is sensitive to the pH-conditions of the surrounding environment. This is due to the fact that the hydrogen-bearing functional group of a hydrogen bonding pair, can be protonated/de-protonated by alteration of the pH of the solution as is for the use of a carboxylic-acid functionality, bonded to an ether functionality13_{. The hydrogen bonding and}

thereby disintegration of the sacrificial layer can thus be controlled since de-protonating the acid functionality will deprive it of its hydrogen bonding capability.

To obtain a free-standing MLTF, one can then use a pH-sensitive film as sacrificial layer, and onto it continue the LBL-synthesis with their constituents, creating a film that is

insensitive to pH-alterations, at least in the same pH interval that is intended for use in disintegrating the sacrificial layer.

Use of such a system has been previously researched by Decher and colleages13_{. Their}

choice of poly(acrylic acid) and poly(ethylene oxide) as constituents of the sacrificial layer, exemplifies the above reasoning regarding pH-sensitivity.

Key feature: pH-responsive sacrificial layer. Assemble at one pH, and

disintegrate the sacrificial layer at another. The target film must be stable and not influenced by the pH alteration.

Limitations: Find a target MLTF that resists disintegration at the target pH. Construct a

sacrificial layer that disintegrates upon reaching a specific pH.

Advantages: This wet-release method should facilitate the release of relatively

thin MLTFs (~200nm reported) 13_. Target-film (free-standing) (A/B)_n (C/D)_m + pH-change

Figure 2-7 As the pH of the surrounding medium is changed, the sacrificial layer

which is held together by hydrogen bonds between constituents A and B dissolves, due to the induced loss of hydrogen bonding capability. The target film is unaffected by the treatment and is left free-standing in the solution.

2.4.4 Dissolving the Substrate:

One method similar to the other mentioned techniques describing sacrificial layers is one using a layer of SiO2 as substrate or sacrificial layer. This technique takes advantage of the

dissolving ability of hydrofluoric-acid (HF) on a SiO2-surface. Previously demonstrated by

Kotov and coworkers7 _{is one technique which features a SiO}

substrate for the LbL synthesis. Upon completion of the MLTF-depositioning, the sample is treated with HF to dissolve the sacrificial layer.

Key features: SiO2 as substrate or sacrificial layer, dissolved by HF.

Limitations: Reactivity of HF towards target film must be prevented. HF needs

careful handling.

Advantage: Relatively thin films have been reported (~50-200nm) 7

LbL-assembly Sample is treated with HF + Deposition of sacrificial -layer

Figure 2-8 The HF assemble-dissolve technique is depicted. Initially LbL assembly onto a sacrificial

layer is performed and this is followed by HF treatment. When treated with HF, the SiO2-sacrificial

layer is removed, rendering the target film free-standing in the surrounding media.

2.4.5 Electrochemical Manipulation of Surface Charge (hypothesis)

It is possible that the use of metal substrates could simplify the release of MLTFs. This hypothesis is based on the use of the metal substrates as electrodes. The metal will have a net negatively charged outermost surface18_{. This could possibly be used to enable}

electrostatic LbL depositioning of alternating positively and negatively charged poly-electrolytes. When the desired film thickness has been reached, a potential is applied which makes the substrate a positively charged electrode. This change of surface charge from negative to positive could induce the desorbtion of intact MLTFs, as the innermost polyelectrolyte layer would be repelled from a surface charge of the same sign (+/-) however no such reports have been found.

Key features: polarisable substrate and the application of surface potential. Limitations: Electrostatic interaction drives the adsorbtion. System must not be

harmed by the applied potential.

Electro-chemical cell Applied potential LbL-assembly

Figure 2-9 Due to the applied potential the surface of the metal-substrate undergoes a

change of polarization, becoming positively charged, and thus repellent of the also positive, electrode-near first layer of the LbL-assembled MLTF.

2.4.6 Our Selected Strategy

Although all of the above mentioned techniques offer relatively easy and already established methods towards creating free-standing MLTFs, we ended up with modifying the ones comprising Teflon substrates, by creating our own Teflon analogues using fluoro-silanizing of silicon substrates to fit our purpose.

The advantages of being able to handle the release of the target-film in a dry environment outweighed the alternatives involving wet-release leading us to choose a modified version of the above mentioned fluorinated surface approach. Although Teflon-coated glass slides are readily available for purchase our desire to be able to monitor the interference behaviour of the MLTFs lead us to develop a method using silicon as a substrate, enabling us to benefit from its contrasting refractive index, compared to glass. Selecting silicon thus simplifies the detection of interference colours in the film by providing a substrate with substantially different refractive index, than the MLTFs.

Being able to peel-off the film using tweezers or similar, in a dry or semi-dry environment (film could be wet although the substrate is not submerged in any solution) would if possible, facilitate the handling, as one would not have to separate the MLTF from the liquid media incorporated with a wet release strategy.

Another advantage of using this method is the fact that the transfer of MLTFs from the substrate to a different carrier material such as paper, through contact drying could be investigated. Contact drying meaning that the MLTF when still connected to the substrate is brought into contact with the target ‘carrier’-material (paper-sheet), under wet or humid conditions, and is then dried together, hypothetically allowing for the transfer of the MLTF to the paper.

As the main scope of this thesis includes the development of release techniques for MLTF intended for applications based on the interference properties of the MLTFs, our work was focused on trying to release as thin MLTFs as possible. As reported by numerous research-groups, creation of relatively large (order of square cm) free-standing MLTFs of 8-9μm thickness is readily possible. As described by section 2.2, these thick films do not show the interference colours of the MLTFs we seek. Consequently, if we are able to create a free-standing segment of MLTF that shows interference colours we will have successfully released MLTFs ten times thinner than previously reported.

Key features:

Dry release of interference MLTFs through weak adhesion between MLTFs and the

hydrophobic fluoro-silanized silicon wafers.

Possible transfer of MLTFs between substrate by using contact drying or other

mechanical manipulation.

Limitations: Mechanically freeing thin films can be difficult in a way that preserves

structure. Our results show that the thin films are fragile (section 4.2).

Advantages:Dry-release should simplify handling, as a MLTF that is suspended in

solution is thought to be harder to handle. 2.5 Fluoro-Silanization

Incorporation onto surfaces of -CF₂- and terminating -CF₃ groups in order to achieve non-stick, highly hydrophobic surfaces is a well known methodology somewhat pioneered by Dupont in the creation and implementation of various Teflon coatings.19 _{The successful}

LbL assembly of MLTFs on Teflon-coated surfaces12, 17 _{inspired us to research the}

Silicon wafers have a native SiO2 outermost layer. The introduction of hydroxyl groups to

such surfaces followed by the immersion of these silicon substrates in fluoro-chloro-silane containing solutions (Figure 2-10), presents a possibility to create CF3-terminated

self-assembled monolayers (SAMs) on silicon.20 _{For the silanization to take place,}

surface-bound water has been reported as a prerequisite.21

The proposed mechanism for the self-assembly of the silane onto the silicon substrate proceeds in four steps21_{. Firstly, the silane molecules physisorb to the outermost adsorbed}

water layer of the silicon substrate. Then the silicon atom of the silane undergoes hydrolysis, Figure 2-11, thus changing its chlorine substituents into hydroxyl groups. The silane molecules then undergo condensation so as to covalently attach to the silicon substrate, Figure 2-12. The final step in the silanisation consists of the in-plane, covalent bonding of the silane molecules amongst themselves, increasing the stability of the SAM, which is illustrated by Figure 2-13.

Cl Cl Si Cl CH₂ CH₂ CF₃ CF₂ CF₂ CF₂ CF₂ CF₂ Figure 2-10 Trichloro (1H,1H,2H,2H

-perfluorooctyl) silane, with its hydrophobic fluorine-containing tail indicated by a blue bar.

Cl Cl Cl Si HO OH OH Si water

Figure 2-11 Surface-bound water enables

hydrolysis of the silane molecules changing their three Cl groups into OH groups.

Si O Si O Si O Si O O H H O H O H O H Si HO O H O H Si + + Si O Si O Si O Si O O O H Si OH HO Si OH +H₂O

Figure 2-12 The silane molecules condense onto the silicon-wafer surface forming the hydrophobic

SAM. The reaction frees water.

Si O Si O Si O Si O O O H Si OH HO Si OH Si O Si O Si O Si O O O H Si O Si OH

+ H

O

Figure 2-13 In-plane stabilizing through covalent bonding between SAM-forming silane molecules.

The ability of the silane molecules to covalently attach to each other is thought to stabilize the formed SAM.

One drawback in using di- or trichloro substituted silane molecules is the fact that the multiple reactive hydroxyl groups of the silicon atom of the silane enables an unwanted agglomeration of the silane molecules and the possible adsorbtion of these aggregates onto the silicon surface. This agglomeration can be avoided by choosing a monochloro substituted silane, as illustrated by Figure 2-14, however this prevents the above mentioned fourth step of the silanisation mechanism (Figure 2-13) resulting in a decreased stability of the SAM due to the lack of in-plane covalent cross-linking. Only one hydroxyl group is hydrolysed for further reaction. A more excessive discussion on this topic is presented by Dutoit et al. 22

R R OH Si R R OH Si R R Si R R O Si OH O H OH Si OH O H OH Si HO Si O O H Si OH O H O Si OH O H O Si a) b)

Figure 2-14 a) illustrating the possible formation of larger aggregates for the trichloro-substituted

silane, which can covalently attach to the surface (unwanted), compared to a monochloro-substituted silane depicted in b) bearing two protective-groups, thus enabling it to either produce a dimer, or attach to the surface. (A produced dimer can not covalently attach to the wafer surface by the same chemistry)

We selected trichloro(1H,1H,2H,2H-perfluorooctyl)silane (PFOS), illustrated by Figure 2-10 as our SAM-forming reagent, based on its tri-substituted nature and its commercial availability.

2.6 Micro Adhesion Measurement Apparatus, (MAMA)

Being able to measure the force needed to separate a controlled area of MLTF from its substrate, was desired in order to evaluate different methods of surface pre-treatment, for MLTF-release i.e. quantify the ease with which a MLTF can be lifted off from its substrate. We therefore set out to try and further develop an already existing technique; Micro Adhesion Measurement Apparatus, (MAMA)23_{, in order to possibly answer if our}

modifications to the silicon substrates had in fact facilitated the release of MLTFs.

The MAMA-technique is currently used on measurements of the adhesion between different substrates24_{, often a half-spherically shaped poly-dimethyl siloxane (PDMS)-probe}

and a flat surface (the substrate). Similar instrumental setups are also used, exemplified by the one used by Chaudbury and Whitesides25_{. While the applied load and the contact area}

between the surfaces are being continuously monitored, the two surfaces are firstly brought into contact with each other and are then separated. Generally, the surfaces will adhere to

each other. If they do not, then separation of the surfaces would occur as the point of zero applied load is passed. However, since the surfaces do often adhere, a negative load is then needed to separate the surfaces.

For our purposes, if the MLTF and the probe are bound strong enough, the separation could possibly occur at the interface between the substrate and the MLTF, rather than between the two surfaces that were originally brought together (MLTF and probe). This would mean a transfer of MLTF from substrate to probe. The measurements would thus result in a value of the load needed to pull-off the MLTF per area of MLTF , which is also monitored by the instrument. This value can have two main contributors, namely the adhesive force between the substrate and the MLTF (sought), and the force required to free an internal piece of MLTF from the MLTF surrounding it, i.e. a cohesive breakage. Our work was focused on investigating the possibility to achieve strong adhesion between the probe and the MLTF through electrostatic attraction between a negatively charged outermost surface (MFC outermost layer) of the MLTF, and a positively charged probe (PEI-coated).

2.6.1 The Instrument

The MAMA instrument23_{, as is illustrated by Figure 2-15, consists of an analytical balance}

and a microscopy-coupled camera for measurement, as well as a motorized probe holder. By photographically monitoring the contact area between the surfaces and correlating it to the load at which the MLTF is pulled off from the substrate one can estimate the force per area needed for release and thus by these values compare different surface treatments to each other. The possibility to detect the change in contact area is contributed to the elastic and transparent nature of the PDMS probe. The PDMS probe deforms when pressed against the MLTF, forming a circular contact area that increases with the applied load. The probe is also transparent enough to be able to be photographed and looked through using a microscope. This is used to obtain the measurements of the contact area. To obtain load values the sample is mounted on an analytical balance during the experiment. This enables measurement of the applied and the pulling loads.

Analysis-balance Microscope Camera Computer Analysis-balance Microscope Camera Computer

Figure 2-15 shows (left) the schematics of the MAMA-instrument. The PDMS half

sphere is mounted on the motorized sample holder and is brought into contact with the surface of the sample (between balance and microscope). To the right a successful lift-off is pictured, where a controlled amount of MLTF has been transferred to the PDMS-probe. A white indent in the sample illustrates the corresponding area of the MLTF that has been lifted off.

2.6.2 MAMA Procedure

The method includes a mounting-, loading-, unloading-, pulling- and a pull-off stage, as is illustrated by Figure 2-16.

Pull-off Pulling

Unloading Loading

Figure 2-16 showing a sequence of a MAMA-experiment. The vertical arrows (grey) indicate applied and

withdrawing load. Multiple horizontal arrows indicate a stepwise increment or decrement of the applied load.

One way to follow the experiment is by looking at the ‘load versus time’ or ‘load versus measurement point’ plot depicted in Figure 2-17. Starting at zero, the applied load (grams) increases to a maximum, which is held during a specified time. This is followed by a decrease of applied load, until zero is obtained. This is the point at which the surfaces

would separate, were there no adhesion between them. By applying negative load until breakage i.e. pulling the probe away from the substrate, obtaining a load that corresponds to the adhesive force between the surfaces and their contact area could be made possible.

Load vs. Measurement Point

-8 -6 -4 -2 0 2 4 6 0 5 10 15 20 25 30 35 40 45 Measurement Point(number) Load( g) Serie1 A B C D

Figure 2-17 shows a Load (g) vs. Measurement Point plot for a MAMA-experiment. Four zones are indicated

by red arrows. A-loading, B-maximum load, C-unloading (negative load => pulling), D-maximum pulling load or pull-off load.

The loading and unloading is preformed in parts. The probe holder is controlled by a step-motor, and each increment or decrement (measurement point) of load is defined as a number of steps. Thus, the applied load is not directly controlled for each increment, with respect to absolute value, but is instead controlled by the maximum load allowed (for the experiment), and the number of steps per increment. The number of steps per increment (SPI) can be defined by the user. The time of an entire experiment is not possible to control directly, though a substantial part of the time is due to the time set at maximum load (at least for our measurements). This is due to the fact that the number of unloading increments depends on the attractive force between the probe and the sample.

The possible controlling parameters thus contain maximum load, time at maximum load, SPI-loading, SPI-unloading, SPI-pulling, and the corresponding times separating the increments, one time per phase i.e. loading, unloading and pulling.

Normally, the established Johnson, Kendall and Roberts(JKR) model26 _{is applied to the}

obtained data once the photographs have been measured for diameter in order to get the force per area relationship. However, the explanation of this theory is beyond the scope of this report, as our primary goal is to investigate and assess whether pull-off using the MAMA is a possible and suitable method, fitting our future needs of surface treatment evaluation.

3 E

This section is intended to describe key experimental details of the preformed research.

3.1 Materials

In this part of the section the chemicals and other materials used are listed.

3.1.1 Miscellaneous:

NaCl, analytical-grade from Merck

Ethanol, (EtOH) 96% vol, VWR international, [EC-label: 200-578-6] Water: Milli-q, Millipore - Synergy 185 apparatus, 18,2 MΩ*cm

For different trials of film release: Adhesive tape, filter paper, copy paper, glass slides and flat-ended tweezers were used.

3.1.2 Substrates:

Polished silicon wafers, MEMC Electronic materials, S.p.A., Novara Italy, 150mm/Cz/1-0-0/Boron/p-type, [PUR-0007 Rev.5 38647]

Aluminium foil: standard commercially available.

Kofes, fibre based composite of approximately 40% poly(lactic acid)

STFI-Packforsk AB27, 28

Grafitti-paint, colour: Copper Chrome, Montana Cans

3.1.3 Polyelectrolytes:

poly(Sodium 4-styrenesulfonate)(PSS), Mw: 70.000 Da, Sigma-Aldrich [Cas 25704-18-1]

poly(diallyl-dimethyl-ammoniumchloride)(PDADMAC), Mw: 500.000 Da, reactant grade, CDM Alcofix III.

poly(ethyleneimine)(PEI) Mw: 60.000 Da, Acros Organics [CAS 9002-98-6] Gelatine, Gelatine Porcine Skin, Type A, 300 bloom, Sigma, [CAS 9000-70-8] Carrageenan, Sigma-Aldrich, [CAS 9062-07-1]

3.1.4 Microfibrillated Cellulose:

The microfibrillated cellulose (MFC), Gen 2, batch 16, 2% wt., produced at STFI-Packforsk AB. Major contributing functionality in making the fibrills anionic in aqueous environment are carboxylic acid functionalities according to manufacturer.

3.1.5 Silanisation:

n-heptane, puriss, 99%, Riedel-de Haën, [CAS 142-82-5]

Trichloro(1H,1H,2H,2H-perfluorooctyl)silane(PFOS), 97% in heptane, Sigma-Aldrich [CAS 78560-45-9]

3.1.6 Micro Adhesion Measurement Apparatus, (MAMA)

PDMS half-spheres, approximate diameter of 1mm by ocular inspection. Prepared as by Eriksson24_.

3.2 Instruments

The key instruments are listed below. Provided in the list of references are links to the web-pages of the developing companies. These describe the instruments in a comprehensive matter, not possible here.

Plasma-cleaner, Harrick-Plasma29

Contact-angle measurements, KSV Cam 20030

Dynamic Light Scattering instrument, Malvern, Nano-zeta series16, 31

Micro Adhesion Measurement Apparatus (MAMA)23

Sonics Vibra-Cell, rod sonicator, 3mm titanium tip32

3.3 Methods & Laborative Setups

The section describes the methods used throughout the thesis.

3.3.1 Preparation of Polyelectrolyte Solutions

All polyelectrolyte solutions were prepared using Milli-Q. A desired concentration of 1mg/ml was achieved by weighing in an appropriate mass of polyelectrolyte, and then adding a corresponding volume of Milli-Q. Polyelectrolyte solutions were then subjected to characterisation of pH and zeta-potential. Table 3-1 gives an overview of the prepared polyelectrolyte solutions. The polyelectrolyte solution concentration of 1mg/ml were prepared in order to roughly maintain a 0.01M concentration with respect to the repetitive unit of the polyelectrolytes (monomolar concentration), the use of which has been reported on17_{. All solutions were allowed to temperate before use and were thus used at room}

temperature, ~22,5ºC. The pH was set by using 0.1M NaOH and 0.1M HCl respectively.

Table 3-1 The type of polyelectrolyte ion and the pH of the

used solutions. cationic 5.5 Gelatine anionic 9.6 Carrageenan anionic 5.5 PSS cationic 5.5 PDADMAC cationic 10.8 PEI anionic 6.8 MFC poly (+)/(-) pH Prepared Solutions:

3.3.2 Preparation of the Microfibrillated Cellulose

The MFC, produced at STFI-Packforsk was prepared similarly to the method previously described by Wågberg11_{. The 2% wt MFC gel stock was diluted with Milli-Q in the ratio of}

1g of gel per 12ml of Milli-Q. This diluted MFC was then dispersed, ~15ml at a time, in a 20ml glass flask, using a Sonics Vibracell rod-sonicator with a 3mm titanium probe for 10 minutes at 25% of the maximum amplitude setting. The MFC fractions were then pooled and centrifuged at 8.000 g for 2 hours. This was preformed in order to separate larger aggregates of MFC from the desired well-dispersed and well-separated microfibrills. The

resulting supernatant was carefully transferred from the centrifuge tubes into a glass container for storage using an autopipette and the pellet was discarded. The MFC was then characterized with respect to size, zeta-potential and pH.

3.3.3 Preparation of Silicon Wafer Slides

The silicon wafers were cut into slides of two sizes, approximately 7 cm x 1 cm for manual LbL protocol and 7 cm x 2.5 cm for use with the Nanostrata system. These slides were then excessively rinsed in the order of Milli-Q - EtOH - Milli-Q and were then dried under a N2-flux. Following this, the slides were placed in the plasma cleaner for 2 minutes at

medium (10W) effect. The plasma treatment removes surface contaminants, as well as renders the silicon substrates with a clean hydrophilic surface29_.

3.3.4 Fluoro-silanisation

Immediately following the plasma cleaning step, the silicon slides were submerged in a 0.1 M NaOH solution. This was done in order to introduce surface hydroxyl groups. Following a Milli-Q washing step and N2-flux drying the silicon strips were immersed in a 1

mM solution of PFOS in heptane for 20 minutes. The PFOS solution was freshly prepared, adding PFOS to a heptane-containing beaker undergoing stirring. The solution was allowed 3 minutes of stirring, before it was used. This was followed by the rinsing of physisorbed silane molecules by sonicating the wafer strips in a heptane bath for further 20 minutes which in turn was followed by excessive rinsing in heptane followed by Milli-Q. Finally the silicon strips were dried under a N₂-flux.

The PFOS solutions were never reused and were always freshly prepared due to reports of similar chemicals undergoing rapid unwanted agglomeration34_{. Due to the nature of the}

involved chemicals, all handling of PFOS and heptane was carried out under a fume-hood.

3.3.5 Preparation of Spray-painted Fibre-Composite (Kofes-demonstrator)

The Kofes, supplied by STFI-Packforsk, was spray-painted with copper paint by holding the sample vertically, and horizontally spraying it. The sample was allowed to dry overnight, before it was cut into 7 cm x 1 cm strips and was subjected to the LbL procedure. One

larger piece of the Kofes demonstrator, approximately 10 cm x 10 cm was also produced, intended to function as a larger scaled demonstrator of the LbL dipping procedure and the resulting interference colouring behaviour of the coating.

3.3.6 Preparation of Aluminium strips

The aluminium foil was thoroughly washed in the order Milli-Q – EtOH – Milli-Q followed by sample drying with a N2-flux. It was then folded into strips of the approximate

size 7 cm x 1 cm, after which it was treated in the plasma cleaner, at medium effect (10 W) for 10 minutes. Following this the samples underwent the LbL procedure.

3.3.7 Manual Layer-by-Layer Procedure

The manual dipping of the substrates was performed by letting the samples stand upright in 15mL beakers, approximately filled to two-thirds. Care was taken not to dip the entire sample in an effort to try to minimize contamination from the tweezers used to handle the samples. The samples were dipped with the following repetitive cycle. Initially the substrate was dipped for 10 minutes in PEI. This was followed by two consecutive 5 minute rinses in Milli-Q. The samples were then dried under a N2-flux before they were submerged for 20

minutes in MFC. Following this, the samples were again rinsed with Milli-Q by two 5 minute dips. Before entering the PEI solution for the second time, the samples were dried using N2.

For the films containing substrate-near PDADMAC and PSS layers, the above method was altered in the following way regarding the layering of these polyelectrolytes. PEI was substituted with PDADMAC (10min dip) and MFC with PSS (10min dip). When the desired amounts of dipping cycles in these polyelectrolytes had been preformed, additional PEI|MFC layering was preformed as initially described.

3.3.8 Automated Layer-by-Layer using the Nanostrata Stratosequence

When the dipping-robot was utilized, three rinses in Milli-Q were used. This was because of it being the default setup for the robot. Otherwise the general procedure was the same

as for the manual dipping, and is shown by Figure 3-1, however the rinsing times were altered to 3 minutes for each beaker.

Figure 3-1 The figure illustrates the dipping cycle used with the dipping

robot (cycle starts at substrate). The substrate is mounted in a sample holder and is then alternately dipped in positively and negatively charged polyelectrolyte-solutions.

3.3.9 The Prepared Films

Approximately one hundred samples were prepared in total. Due to the often destructive nature of the release attempts i.e. manipulation with razors, knives adhesive tape. For clarity, the selected samples featured and discussed in the results section of this thesis are presented in Table 3-2

Table 3-2 The samples featured in this thesis.

Experiment Build-up Colour (substrate) Colour (released) Thickness (modeled) MLTF free-standing (PEI|MFC)22 green violet 400nm

MLTF copy-paper (PDADMAC|PSS)4-(PEI|MFC)10 orange transparent whitish 250nm

Kofes strips (Carrageenan|Gelatin)10 transparent(glossy) -

-Kofes demonstrator (Carrageenan|Gelatin)10 transparent(glossy) -

-MAMA (PDADMAC|PSS)4-(PEI|MFC)10 orange transparent whitish 250nm

3.4 Micro Adhesion Measurement Apparatus (MAMA)

The experimental part related to the MAMA is described below.

3.4.1 Preparative

The probe holder (glass surface) was thoroughly cleaned with Milli-Q – EtOH – Milli-q and was then dried with a N2-flux. The PDMS half-speres were mounted on a clean glass

slide and were then rinsed in the order of Milli-Q - EtOH- Milli-Q, after which they were plasma-treated for 1 minute on medium (10 W) effect. This was followed by 10 minutes of incubation in PEI after which the probes were thoroughly rinsed with Milli-Q before being dried under an N2-flux and transferred to the glass surface of the probe holder. The

substrates were MLTFs prepared on fluorinated surfaces according to the above mentioned protocol. One difference was that the MLTFs in these experiments consisted of two parts, the substrate-near being (PDADMAC|PSS)4 and the surface-near being (PEI|MFC)10. The

release of (PDADMAC|PSS)-MLTFs demonstrates a more general method than would the release of a pure (PEI|MFC) MLTF, due to the fact that MFC is not generally available for purchase, whereas both PDADMAC and PSS are commercially available.

3.4.2 Experimental

We focused our investigation on the pull-off of a MLTF with an outermost negatively charged MFC surface from a fluorinated silicon surface. The PDMS probe was coated with positively charged PEI to complement and bond to the negatively charged MFC,

outermost on the MLTF. The planned and performed experiments are shown by Table 3-3. Due to numerous mishaps and the quite rigorous and time-consuming experimental preparations, only a couple of measurements were preformed. These however proved successful regarding MLTF-release, as is described further in the results & discussion section of the thesis.

The sample was mounted on the balance which was then tared (zero-load). The probe was then pressed against the sample resulting in an increasing load on the balance. Through the transparent probe, made from polydimethylsiloxane (PDMS), the camera registered the contact area between the PDMS probe at the measure points and the sample making it possible to correlate an area and a load to one another. When the maximum load had been applied during the desired time, the unloading started. The unloading continued until the MLTF had been transferred or until the surfaces were separated. Both dry and wet measurements were carried out. What differentiated the two types from each other was that a droplet of Milli-Q(wet) was applied onto the substrate prior to the PDMS probe and the substrate being brought together, in the wet technique. This wet technique was applied to increase the electrostatic bonding capability of the surfaces. Once the maximum load had been applied, the excess water was blown away using a N2-flux.

Table 3-3 showing the experimental setup that was planned and preformed using the Micro Adhesion

Measurement Apparatus (MAMA).

1g 2g 5g 10g ┌ steps ┐ ┌ steps ┐ ┌ steps ┐ ┌ steps ┐

1h 20* 100 20 100 20 100 20*** 100

2h 20 100* 20 100 20** 100 20 100 described in results&discussion

* probe pulled off from holder (no glue used) ** probe broke (pulled in two pieces)

*** caused contact area to take up entire frame (no measurent possible)

unmarked cancelled due to time shortage

Max-Delay

Maximum Load

(All experiments were performed with 5s between each increment/decrement of load (SPI), and with MLTF-buildup [(PDADMAC|PSS)₄-(PEI|MFC)₁₀]

4 R

In this section the results are presented and discussed.

4.1 Preparative

This section features the preparative characterisations.

4.1.1 Material Characterisation

Provided below is information regarding particle size that was obtained for PEI and MFC using dynamic light scattering, illustrated by Figure 4-1 and Figure 4-2, respectively.

Figure 4-1 The figure gives an estimation of the diameter of the PEI molecules. Three

over-layered curves are shown indicating an approximate diameter of ~5 nm. As the figure indicates, larger aggregates do exist, but are relatively few with respect to the total volume of light-scattering substance.

Figure 4-2 The figure indicates a value for MFC that would normally correspond to

particle/aggregate diameter (~10 nm). In this case however, because the fibrills are assumed to have a somewhat cylindrical geometry, this value is thought to correspond to the diameter of the cylinder (as is discussed further in the continuation of this section).

The solutions of PEI and MFC were further characterised with respect to zeta-potential, as is shown by Table 4-1, Figure 4-3 and Figure 4-4. These data were also retrieved using the dynamic light scattering apparatus.

Table 4-1 The zeta-potential of MFC as well as that of PEI, as shown in the column headed ZP.

Figure 4-3 The distribution of the dynamic-light-scattering measurement data obtained for PEI

regarding zeta-potential. Three over-layered peaks at ~25mV.

Figure 4-4 Three over-layered curves showing the distribution of the zeta-potential of MFC, derived

from the dynamic light scattering measurements, (~-115 mV).

The results from the DLS measurements of the zeta-potentials confirm that the preparative work of this thesis is in accordance with the previously reported research8, 10, 11 _{on MLTF}

synthesis with PEI and MFC as positively and negatively charged constituents respectively. Furthermore the estimated size of the constituents correspond to those reported by Axnäs and Wågberg10, 11_{. The fact that the real sizes of the microfibrills differs from those}

obtained by the DLS experiments in this and other reports is discussed further by Axnäs 10_.

In short it can be explained by the fact that the Brownian motion of particles utilized in DLS measurements is not likely to proceed in the lengthwise orientation of the MFC.

Whereas the DLS apparatus assumes a globular particle conformation, the microfibrills, as shown by atomic-force microscopy and transmission electron microscopy10_{, are roughly}

cylindrically shaped with approximate measurements of 10 nm in diameter and up to 1 μm in length. The DLS values thus correspond well with the diameter of the microfibrills.

4.1.2 Contact Angle Verification of Hydrophobicity of Fluorinated Substrates To verify the success of the fluorinating step, contact-angle measurements were carried out. The fact that spontaneous de-wetting occurred when rinsing of the fluorinated surfaces in Milli-Q was preformed gave a hint of success in the change of surface behaviour from hydrophilic to hydrophobic. However, we wanted to quantify this by determining the contact angle.

Measuring at three separate points on a surface, we reached contact angles of 110.5 degrees, for Milli-Q against the fluorinated surface. As a reference the contact angle of Milli-Q on a silicon wafer was measured to about 15 degrees. This value correlates well with the previously reported contact angles of self-assembled monolayers of PFOS analogues on silicon. Achieving contact angles of up to around 130-140 degrees have been reported, although such high contact angles rather suggest the multilayering of molecules than ordering and further packing of molecules within one monolayer20_.

Photographs representing reference and substrate measurements are provided in Figure 4-5 .