IN
DEGREE PROJECT TECHNOLOGY, FIRST CYCLE, 15 CREDITS
STOCKHOLM SWEDEN 2020,
Evaluation of joint formation on cellulosic surfaces
FIVAZ ERIKA
KTH ROYAL INSTITUTE OF TECHNOLOGY
SCHOOL OF INDUSTRIAL ENGINEERING AND MANAGEMENT
Abstract
Environmental issues are more and more present in our societies. Pollution engendered by plastic waste have drastically increased these past decades, causing several threats to the ecosystem. Therefore, the need of new biodegradable plastics to replace the actual petroleum-based ones is urgent. Cellulose could be a potential substitute since it is a biopolymer, abundant on Earth. However its properties have to be enhanced to be competitive towards actual plastics. The aim of the project is therefore to get a better understanding of cellulose-cellulose interactions. It focuses on the adhesion between cellulosic surfaces. Contact adhesion measurements have been performed on cellulose beads, with different treatments. All the beads had the same size and same concentration. Some of them were native whereas others were charged (600 µeq/g).
Half of the native beads were surface modified with a starch coating or a Layer by Layer technique using cationic starch and an anionic polyelectrolyte (EXPN64 or FennoBond 85E). The project included preparation of the surface modified beads, pull-off tests, where load and position were recorded as a function of time, as well as measurements of the contact area. It was found that a higher energy was needed to separate charged and surface modified beads, especially the ones modified with EXPN 64, compared to native beads. The project have also shown that the types of beads influenced the contact area and the strength. However a trend was sometimes difficult to find. The data and results obtained in this project could be further re-used to enlarge the study field and investigate the influence of other parameters (size, concentration) on the adhesion of cellulose beads.
Sammanfattning
Miljöfrågor är mer och mer närvarande i våra samhällen. Föroreningar med plastavfall har ökat drastiskt de senaste decennierna och orsakat flera hot mot ekosystemet. Därför är behovet av ny biologiskt nedbrytbar plast för att ersätta de petroleumbaserade brådskande. Cellulosa kan vara en potentiell ersättare eftersom det är en biopolymer. Emellertid måste dess egenskaper förbättras för att vara konkurrenskraftiga gentemot petroleumbaserad plast. Syftet med projektet är därför att bättre förstå cellulosa-cellulosa-interaktioner. Den fokuserar på vidhäftningen mellan cellulosaytor. Kontaktvidhäftningsmätningar har utförts på cellulosapärlor med olika behandlingar. Alla pärlor hade samma storlek och koncentration. Vissa av dem var naturliga medan andra laddades (600 µeq/g). Projektet inkluderade beredning av ytmodifierade pärlor, utdragningstester, där belastning och position registrerades som en funktion av tiden, samt mätningar av kontaktområdet. Det visade sig att högre energi behövdes för att separera laddade och ytmodifierade pärlor, särskilt de modifierade med EXPN 64, jämfört med tonativa pärlor. Projektet har också visat att typerna av pärlor påverkade kontaktområdet och styrkan. Men en trend var ibland svår att hitta. Uppgifterna och resultaten som erhållits i detta projekt kan vidare användas för att förstora studiefältet och undersöka påverkan av andra parametrar (storlek, koncentration) på vidhäftningen av cellulosapärlor.
Author
Erika Fivaz, fivaz@kth.se Materials science, ITM
KTH Royal Institute of Technology
Place for Project
Teknikringen 56 Stockholm, Sweden
Department of Fibre and Polymer Technology
Supervisor
Torbjörn Pettersson
Department of Fibre and Polymer Technology KTH Royal Institute of Technology
Contents
1 Introduction 1
1.1 Motivations . . . 1
1.2 Purpose and goal . . . 1
2 Theoretical aspects about cellulose and regenerated cellulose 3 2.1 Structure and properties of cellulose . . . 3
2.2 Spherical probes from regenerated cellulose . . . 4
2.3 Cellulose modified by ”Layer by Layer” method . . . 7
2.4 Adhesion of cellulose beads . . . 7
3 Experimental 9 3.1 Materials . . . 9
3.2 Drying of cellulose beads . . . 9
3.3 Surface modification . . . 9
3.4 Contact adhesion measurements . . . 10
3.5 Determination of the contact area . . . 10
4 Results 11 4.1 Drying of cellulose beads . . . 11
4.2 Pull-off force . . . 11
4.3 Contact area . . . 13
4.4 Stress-Strain curves . . . 14
4.5 Surface homogeneity . . . 15
5 Discussion 17 5.1 Drying of cellulose beads . . . 17
5.2 Cellulose adhesion . . . 17
5.3 Social, environmental and ethical considerations . . . 19
6 Conclusion and future work 20 6.1 Conclusions . . . 20
6.2 Future work . . . 20
7 Acknowledgements 21
8 References 22
1 Introduction
1.1 Motivations
70% of the Swedish territory is covered by trees. 98% of these trees are part of cultivated forests exploited for paper making and sawmills industry [1]. Wood is composed between 40 and 45% of cellulose , which is a natural polymer that could replace most of the plastics if its properties are enhanced, especially in humid environment [2].
Moreover, the progressive awareness of climate change leads the companies to limit their consumption of petroleum-based plastic items. The combination of these two factors makes Sweden an appropriate country to launch and pursue researches about cellulosic surfaces. It could therefore lead to the apparition of new renewable and sustainable materials, stronger and stiffer than existent one.
Cellulose plastics already exist. However, the most common cellulose-based thermoplastic, cellulose acetate, is produced by esterification of cellulose present in wood pulp [3]. Nevertheless, this latter and all the others esterified cellulose- based materials require the use of chemicals. Indeed, plasticizers and fillers are needed in order to obtain an usable plastic, to make the esterificated product more flexible [4], tougher and more resistant to the moisture. Yet, the main plasticizers of cellulose acetate are highly toxic (phtalates, glycerin) [3], and the interest of using an environmental friendly polymer is therefore critically reduced. Some researchers have already started to work on green and renewable plasticizers and fillers in order to deal with the climate threat. Huan X and coworkers have for instance based their work on carbonized spent coffee grounds [5]. They were able to show that biopolymers filled with environmental friendly carbonized coffee grounds show enhanced properties. An alternative to produce new materials while fighting the climate change is to work with regenerated cellulose. A major commercialised plastics film made out of regenerated cellulose is the so-called ”Cellophane”, which was unfortunately sidelined by similar films based on polypropylene or polyethylene, which are cheaper and way easier to process [6]. Materials from regenerated cellulose are therefore a wide field to study in order to replace petroleum-based plastics by completely ecofriendly bio-based plastics.
1.2 Purpose and goal
This is the main reason why a better understanding of interactions between regenerated cellulose and other surfaces, as well as interactions between two regenerated cellulosic surfaces, are needed. As a lot of studies have already been published about this first issue, this bachelor project aims to focus on adhesion between cellulosic surfaces only.
Cellulose/cellulose interactions were studied through the intermediary of spherical
beads. This is a theoretical model that simplifies the study of adhesion mechanisms between these surfaces, avoiding roughness and inhomogeneity induced by thin film surfaces. Different aspects were evaluated such as the concentration of cellulose used in the formation of the cellulose beads, the way the beads are drying and the force needed to separate the beads. The purpose of the project is therefore to proceed to contact adhesion measurements on different beads in order to compare and understand the effect of the charge and the addition of polyelectrolytes on cellulose. Five types of beads were used in the project: native, charged and surface modified beads with 3 different chemicals.
2 Theoretical aspects about cellulose and regenerated cellulose
2.1 Structure and properties of cellulose
2.1.1 Structure
Cellulose of molecular formula (C6H10O5)nis the most abundant polymer on Earth. Its structure is shown on figure 2.1. It is a biodegradable and biocompatible homopolymer made exclusively out of glucose on its β-d-glucopyranose form. The polymer is linear and mostly syndiotactic. The degree of polymerisation can vary between 300 and 1700 for cellulose coming from wood pulp. It can reach up to 10 000 units for cellulose from cotton [7, 8].
Figure 2.1: Structure of cellulose with the numerous hydroxyl groups [9]
Cellulose is a semi-crystalline polymer with a crystal content close to 50%. This can be explained by the high mobility of the chain and the interactions between the hydroxilic groups [10]. Cellulose can crystallize in four different structures, known as cellulose I,II,III and IV. Cellulose I and II are the most used in the industry.
Cellulose I corresponds to native cellulose found in the nature, whereas cellulose II is obtained by regeneration of cellulose I. Regenerated cellulose possesses a particularly well developed network of hydrogen bonds, between and inside the chains, (cf. fig 2.2) allowing stability to the entire structure [11]. An anti-parallel chain stacking can be observed in cellulose II while cellulose I presents a parallel chain stacking, suggesting an irreversible process from cellulose I to II. Regenerated cellulose is therefore energetically more favourable than native cellulose that can be considered as meta-stable [9–11].
Regenerated cellulose is a quite stiff polymer, with a stiffness close to 160 GPa according to the type of dissolution performed [12]. The amorphous part of cellulose decreases the stiffness of the polymer especially in wet state [10].
2.1.2 Anisotropic behaviour of cellulose
Concerning the behaviour of cellulose towards water, an ambivalence can be noticed.
Indeed, cellophane, which is a regenerated cellulose based plastic, and more generally cellulose is known to be one of the most hydrophilic polymer, with a contact angle close
Figure 2.2: Hydrogen bonds network of regenerated cellulose [7]
to 12° for cellophane [13]. This water affinity also explains the poor stability of cellulosic materials when they are put in contact with water or in humid environments [14].
Yet, the most challenging part when processing regenerated cellulose is its insolubility in water. Cellulose has an amphiphilic character and hydrophobic interactions are fundamental to understand the difficulty of dissolving cellulose as explained in section 2.2.1 [15]. Hydroxyl groups, -OH, are the main characteristic groups present in cellulose, responsible of the hydrogen bonds linking the different chains of the polymer.
As it can be seen on part b) of the figure 2.3, these hydroxyl groups are dominating in the (1¯10) plane of the crystal phase of regenerated cellulose, giving rise to a hydrophilic surface. These groups are situated in the equatorial position of the above- mentioned glucopyranose rings. On the contrary, the orthogonal plane (110) presents a hydrophobic behaviour, due to the absence of hydroxyl groups, only the hydrogen atoms are present in this plane as shown in part a) of the figure 2.3 [13, 16].
Figure 2.3: a) Hydrophobic surface of cellulose, side view of the rings; b) Hydrophilic surface, front view of the rings. Oxygen are the spotted atoms [13]
2.2 Spherical probes from regenerated cellulose
According to the cellulose face on the outside of the regenerated polymer, the obtained material can be either hydrophobic or hydrophilic [16]. Regeneration of cellulose by
dissolution is nowadays the most green process to derive new cellulose-based materials for various applications, especially since all reactants are reused or can be recycled [6].
2.2.1 Cellulose dissolution
The starting material for the cellulose dissolution is pulp fiber. First of all, the step of washing is particularly important since it enables to remove metallic ions and dissolved unwanted substances present in the wood pulp, such as lignin which is the second major compound of wood cell wall after cellulose [17]. Two steps are needed for the washing.
In the first one the metal ions are released by adding hydrochloric acid to the solution.
In the second step, sodium bicarbonate N aHCO3 enables to dissolve the unwanted colloidal substances.
The major issue when processing cellulose is its insolubility in the most common solvents, due to the lack of flexibility, that is the high persistence length of the chains, and the abundant hydrogen bonds network. The first solvents used to regenerate cellulose were harmful for the environment, thus the idea is to use a renewable solvent able to break the molecular bonds of cellulose without damaging the structure [18].
LiCl/DMAc (Lithium chloride/Dimethylacetamide) is one of them. High molecular weight cellulose can be dissolved without alteration under normal conditions, in a sustainable way. Another advantage of this solvent is that the dissolution is almost total and the polymer in solution keeps its viscosity even after a while, meaning that dissolved cellulose in LiCl/DMAc can be stored and measurements can be performed later on [19]. The solution contains typically 7% wt of LiCl. During the process, a solvent exchange is performed with ethanol. The overall dissolution lasts for approximately one week. As explained in figure 2.4,the H+protons of cellulose hydroxyl groups react with chloride ion splitting the Li-Cl couple and breaking the intermolecular hydrogen bonds formed by cellulose.
When dissolved in LiCl/DMAc and then solidified in a non-solvent as explained in the next section 2.2.2, cellulose looses its crystallinity and becomes almost an amorphous polymer [9].
2.2.2 Formation of cellulose beads
To perform contact adhesion measurements, smoothness of the surfaces is of great importance. To compute the true force acting on a material at atomic and molecular scales, the molecular interactions must be exerted on the entire contact area [9]. This latter aspect could be altered by asperities on the surfaces. That is the reason why, spherical probes, that is cellulose beads in this case, are preferably used. Indeed, spheres limit the risk of surface defects by decreasing the contact area.
Figure 2.4: Mechanism of cellulose dissolution in LiCL/DMAc solvent [6]
Cellulose beads can be shaped, once the dissolved cellulose solution has been prepared.
The process that leads to the final beads is illustrated in Figure 2.5. Previous studies compared the impact of the non-solvent on the quality of the obtained beads. Both the height of dripping and the concentration of the beads have to be taken into account in order to get a smooth surface. The most suitable non solvent for this purpose is ethanol.
Indeed, even though cellulose is not soluble in water, indicating a certain stability of beads in this medium, the surface tension of water is too high (72 mN/m) to enable a smooth sphere to form. Numerous defects appear on the surface due to the excessive dripping height (around 10cm) needed to make the cellulose penetrates the solution.
The idea is to limit the shocks when the drops reach the surface [6]. Ethanol, with a surface tension around 22 mN/m [20], fulfills the different conditions to limit the roughness of the surface and obtain almost perfect spheres with a dripping height close to 1cm, while maintaining a sensible drying time (48h hours at room temperature is considered to be more than enough). Methanol is actually an acceptable non solvent considering the shape of the spheres, yet the drying time can be up to one week at 5 degrees. The efficiency of the experiments is therefore partially altered [21].
As stated above, the concentration of the cellulose is a significant aspect to consider to get exploitable spherical cellulose models. The concentration has indeed a direct influence on the viscosity of the solution. Thus, a too high concentration, higher than 2% wt, creates elliptical samples topped by a tail. In the same way, a concentration lower than 1% wt of cellulose, leads to buckles and wrinkles during the drying step [21].
Moreover, when solidified in the above-mentioned solvents, cellulose beads keep their spherical shape in air and are resistant enough not to break when exposed to a different medium. The proportionality is also conserved when the beads are dried and potentially re-wet. This can be explained by the homogeneous distribution of water and cellulose
inside the spheres [14]. This aspect is crucial to be able to work on these beads and perform further measurements.
Figure 2.5: Schema of the different steps useful for the realisation of cellulose beads [2]
2.3 Cellulose modified by ”Layer by Layer” method
To increase adhesion properties of cellulose, surface modification can be performed.
One common way of modifying a polymer with polyelectrolytes is the Layer by Layer technique, LbL. It consists of an alternation of negatively and positively charged layers coating the material. This method is driven by the adsorption of polyelectrolytes on the surface. According to JL Barrat and JF Joannie, ”polyelectrolytes are polymers bearing ionizable groups, which [...] can dissociate into charged polymers chains” [22].The amount of adsorbed polyelectrolytes as well as the surface that adsorbed (external, micro- or macropores) are dependant on the type of polyelectrolyte, the pH and the ionic strength. Electrostatic interactions between oppositely charged surfaces are the driven forces for the process [10]. Studies have shown that the adhesion properties between two surfaces are increased at the only condition that the two surfaces are brought into contact in the wet state. No adhesion enhancement due to the LbL coating is noticed when the surfaces are dried separately and then brought into contact [23].
Starch is a possible cationic polyelectrolyte used to coat cellulose. It is a common manner of strengthening paper for instance. Starch is adsorbed by the negatively charged surface once the steric repulsion has been overcome and bridging has occurred.
Steric stabilisation is an extra repulsive force overcoming Van der Waals attraction.
Thus, it can fill the asperities of the surface and enhance the contact area [10].
2.4 Adhesion of cellulose beads
The adhesion between two cellulosic surfaces is due to attractive Van der Waals interactions. For charged surfaces, electrostatic interactions could also create a repulsion. At very short range (few Angstroms), if the contact area is very smooth, some stronger bonds may be responsible for the adhesion, such as hydrogen bonds or even primary covalent bonds [10, 24]. When the adhesion is wet, especially before the beads are dried, capillary forces are additional forces contributing to the adhesion
mechanism. In presence of water, a meniscus forms between the two surfaces as shown in figure 2.6. The pressure, ∆P ,inducing an attractive force, can be described by Laplace equation(cf. Equation 1). The capillary force is the combination of surface tension and viscous force [10, 25].
∆P = 2γ
R (1)
where γ is the surface tension of the liquid and R the radius of the meniscus.
Figure 2.6: Illustration of the water meniscus inducing a capillary force between two spheres [26]
Thus, different forces can be taken into account when talking about adhesion. One common model to describe contact adhesion mechanism as a whole is John Kendall Roberts model (JKR) [24, 27]. It relates the negative load, F, needed to separate two surfaces (generally two spheres)to the adhesion energy, W, through the following equation:
F =−3
2πW R (2)
where R, the equivalent radius of curvature of the system can be calculated as follows for 2 spherical cellulose beads of same diameter:
1 R = 1
R1 + 1
R2 = 2
Rbead (3)
3 Experimental
3.1 Materials
Two types of beads were used to perform the experiments. The beads had been preliminary prepared by dissolution and cellulose regeneration from raw wood pulp.
The beads were dissolved in LiCl/DMAc solution and ethanol was the non-solvent employed to shape the beads. All the experiments conducted in this project, started from beads already prepared (and not from pulp fiber). The two types of beads used were:
1. Native beads, concentration of cellulose 1%wt, diameter 0.6mm
2. Charged beads 600µeq/g, concentration of cellulose 1%wt, diameter 0.6mm The beads were stored in MilliQ water.
Some of the native beads were further modified with cationic starch, charge density 0.2meq/g, supplied by Chemigate (Finland) and two different commercial anionic polyelectrolytes usually used in paper making fabrics: FennoBond 85E, dry content 20,56% as well as EXPN64, dry content 13,96%.
3.2 Drying of cellulose beads
The contact adhesion testing is performed on dry beads. The beads (native and charged) were dried on 3 different surfaces: Polydimethylsiloxane (PDMS), Teflon and glass. The surfaces were first washed with water, ethanol and water. The beads were then paired.
Two different samples were tested: beads dried next to each other and beads dried on top of each other. They were left to dry for approximately one day.
3.3 Surface modification
Half of the native beads were surface modified before being dried. Three different surface modifications were performed on the beads. This was done with the LbL methods as explained in part 2.3. The beads were first coated with starch. For some of them, EXPN64 or Fennobond were added on top of the cationic starch layer. The three different groups of surface modified spheres were native beads 1%wt of cellulose, diameter 0.6mm coated with:
1. Cationic starch (1 layer)
2. Cationic starch (1 layer) + FennoBond (1 layer) 3. Cationic starch (1 layer) + EXPN64 (1 layer)
The solutions used for the LbL coating were prepared according to the following protocols.
Starch solution 1% was cooked by stirring water and starch at 97°C for 30min. The solution has to be stored in fridge and be used in 48h.
EXPN64 and FennoBond 85E 0.1% were simply prepared by mixing the polyelectrolyte with water and adjusting the pH to 6.5 by addition of NaOH in order to make the solution less acid.
The three solutions were then diluted 100 times. The beads were let 10 minutes in the corresponding solutions in order for the polyelectroyte to be adsorbed by the cellulosic surface. Washing had to be performed between each layer.
3.4 Contact adhesion measurements
Contact adhesion testing is performed with a 600g sensor. The paired beads were glued to a glass slide with UV glue (Norland Optical Adhesive 81), the opposing flat was moved towards the upper bead until contact and tare was made. The two attached beads were then pulled apart at a rate of 5µm/sec with a pull off length of 150 µm. The experiment is illustrated on figure 3.1. The entire process is observed with a Dino Lite capture camera. The load and the position were recorded as a function of time. The measures were repeated on three samples for each of the 5 categories of beads (native, charged, 3 different surface modifications).
3.5 Determination of the contact area
After pull off measurements, the contact area between the two beads was determined with ImageJ software. The pictures of the interfaces were taken through the intermediary of Olympus optical binocular microscope, magnification x20.
(a) During pull off (b) After separation
Figure 3.1:Illustration of the pull off force process
4 Results
4.1 Drying of cellulose beads
Three surfaces were tested: PDMS, glass and Teflon. The results were not convincing with the glass surface. 75% of the beads separated during the drying process. On the contrary, the results were encouraging on Teflon and PDMS. The first beads dried, ie. native beads, appeared to stick together when dried on top of each other. When they were dried next to each other, the loss (separation of the two beads) was far more important. It seemed that PDMS surface was more sensitive to the orientation of the beads (next or on top of each other) than Teflon. Therefore, all the further beads were dried on Teflon. 77% of the beads adhered to each other when they were dried on Teflon on top of each other. No significant difference was noticed according to the type of beads dried, except for LbL coated beads with starch and EXPN64, for which 50% of the beads separated upon drying.
4.2 Pull-off force
15 successful pull-off force tests were conducted, 3 for each categories of beads. More tests were realised, yet a certain number of them failed due to the non separation of the beads during the tensile test. The samples are labelled in the order they were tested.
For instance, on figure 4.1, the numbering starts at ”4”, meaning that samples 1, 2, 3 did not separate or data have been collected properly. Numerous adjustments were needed to obtain the good software set-up. It appeared that a pull-off length of 150 µm was more suitable than a length of 100 µm. Moreover, the amount of UV-glue put to attach the samples to the glass slides was of great importance. It was noticed that a too small amount of glue conducted to a non separation of the beads. The beads rather separated from the bottom slides in these case.
Load curves were reported for each sample as a function of the position (cf. figures 4.1 to 4.5). All data were conserved for this part of the results, in order to be able to see the diversity of outcomes obtained inside the same group of beads. A difference of load of more than 50g can thus be spotted between the different starch coated beads, as shown on figure 4.3. It can be interesting to notice that the results obtained for native and charged beads (figures 4.1 and 4.2) are more homogeneous than for surface modified beads (figures 4.3, 4.4 and 4.5). For these latter, the load curves can differ a lot from a sample to another.
Samples referred as ”a” on figures 4.2, 4.3 and 4.5, correspond to pairs of beads that have been pulled twice. This means that they separated from the glass slide at first.
Therefore, they could be re-used for a second pull-off test.
Figure 4.1: Comparison of the load curves for the native beads
Figure 4.2: Comparison of the load curves for the 600µeq/g charged beads
Figure 4.3: Comparison of the load curves for the starch coated native beads
Figure 4.4: Comparison of the load curves for the LbL (Starch+EXPN64) native beads
Figure 4.5: Comparison of the load curves for the LbL (Starch+FennoBond 85E) native beads
4.3 Contact area
The contact area was determined by image analysis after the separation of the beads.
As shown in the table 4.1, the contact area is similar for the natives and Layer by Layer beads (≈ 6 · 10−3mm2), with a small standard deviation between all the samples.On the other hand, the contact area of the charged and starch coated beads is more than two times bigger, with a value close to 14· 10−3 mm2. The results are nevertheless more dispersed, respectively 28% and 55% of deviation.
Table 4.1: Table gathering the contact areas for the different samples according to the type of beads
Type of beads Sample number Contact area (10−3mm2) Mean (10−3mm2)
4 7.09
Native 5 6.05 7.01± 0.925
6 7.89
2 14.2
600µeq/g charged 3a 8.29 11.8± 3.12
4 13.1
1 5.12
Starch coated 3a 19.5 13.7± 7.58
4 16.5
1 6.40
LbL: Starch 2a 5.67 5.70± 0.511
+FennoBond 85E 3 5.09
1 6.94
LbL: Starch 2 6.58 5.93± 1.45
+EXPN64 3 4.27
4.4 Stress-Strain curves
From the load and the contact area, it was possible to derive strain-stress curves for each types of beads (cf. figure 4.6).The curves were realised by taking the average curve for each sample and excluding the irrelevant samples. For instance, the data obtained from a pair of beads used twice (indicated as ”a”) or data from non homogeneous surfaces were not taken into account. The strain (cf.Equation 4) and the stress (cf. Equation 5 were calculated with the following formulae:
ϵ = −Positiont+Positiont0
Positiont0 ∗ 100 (4)
σ = Loadt− Loadmax
Contact area ∗ g (5)
Figure 4.6: Comparison of the stress-strain curves between each type of beads In a second phase, the adhesion energy needed to separate two beads was determined.
The results are gathered in the table 4.2. Thus, it can be noticed that native beads chemically modified as well as charged beads require higher energy than native beads without any treatments.
Table 4.2: Compilation of the energy needed to separate two beads calculated from the area under the stress-strain curves of each category of beads
Type of beads and treatments Energy (kJ)
Native 64.7
600µeq/g charged 87.5
Starch coating 77.7
LbL: Starch + FennoBond 85E 72.7 LbL: Starch + EXPN64 112
4.5 Surface homogeneity
It is interesting to notice that not all the coated beads have absorbed the same amount of polyelectrolytes and in the same manner. Therefore, some of them have a rough surface
with a loss of homogeneity. The figures 4.7c and 4.7d show the difference of adsorbance between two starch coated beads. The former is clearly rough and inhomogeneously coated, whereas the latter has only some small asperities and defects. A difference between the modified beads (cf. photos 4.7c, 4.7e, 4.7f) and the non modified ones (cf. figures 4.7a and 4.7b) can also be spotted on the figure 4.7.
(a) Native bead (b) Charged bead (c) Starch coated bead
(homogeneous)
(d) Starch coated bead (inhomogeneous)
(e) LbL (starch+EXPN64) (f) LbL (starch+FennoBond 85E)
Figure 4.7: Comparison of the homogeneity of the surfaces for the different types of beads.
The scale is similar for each photo.Magnification x20
5 Discussion
5.1 Drying of cellulose beads
As stated in part 4.1, the drying was more efficient when the beads were dried on top of each other. Indeed, the idea of performing the drying step vertically involves a reduction of the area of the cellulose beads directly in contact with the surface. Only one bead touches the surface in this case. When the two beads are in contact with the drying surface (i.e. horizontal drying), for the beads to stay attached by pairs, the affinity between cellulose-cellulose has to be higher than the cellulose-drying surface affinity.
Therefore, the risk of separation is significantly more important. This attraction is driven by capillary forces. A higher percentage of separation was noticed for beads dried on a glass surface. Glass is highly hydrophilic whereas PDMS and Teflon are hydrophobic surfaces [28]. Wet beads would therefore rather stay together than be in contact with these latter surfaces. On the other hand, capillary forces would develop between wet beads and glass surface, explaining the higher separation rate.
5.2 Cellulose adhesion
5.2.1 Load curves
Separation of the beads occurs at the peak of each graph presented in part 4.2. A higher load in absolute value would have been expected for surface modified beads and charged ones since electrostatic interactions emerge from the surface charges [10]. However, as it can be seen on the different load curves, the results are highly dispersed. This diversity of the obtained results makes it hard to deduce a trend for the required load to separate two beads. However, it can be noticed that the results show a lower dispersion for the natives and charged beads (cf. figures 4.1 and 4.2) than surface modified beads.
A possible explanation is the homogeneity of the surfaces. As shown on photographs from figure 4.7, the surface of the non modified beads (4.7a and 4.7b), do not present the same asperities and small defects than the modified beads (4.7e, 4.7f, 4.7c). The modification of the beads may have altered the homogeneity of the surface. However, roughness is an important parameter to take into account when performing contact adhesion measurements, it can highly influence the adhesion between two surfaces.
Homogeneity issues are clearly visible for starch beads on figure 4.3. The microscope image obtained after the pull-off of the samples 1 and 4 (curves blue and grey) is similar to the image 4.7d, where the total lack of homogeneity is visible. The polyelectrolyte did not adsorbed properly, a longer dipping time would probably have improved the result.
This fact potentially explains the really low results got for these samples, compared to the sample 2a, where a smooth surface was observed.
Concerning LbL beads (starch+EXPN64), the sample 3 reaches the peak at a load close to -5g whereas the peak is around -30g for the two other samples (cf. figure 4.4). The difference could be due to drying time. Indeed, 50% of the EXPN beads separated during drying, a high amount compared to other beads. A small number of bonds and interactions may have had time to develop between the two cellulose spheres.
Load curves of LbL (Starch+Fennobond) beads (cf. figure 4.7f) also present one result with a much lower load. The sample 2a as indicated by its name has been pulled twice before separation. It is therefore probable that the first pull-off test broke some bonds and weakened the cellulose-cellulose joint, even if there was no separation. Hence the inferior load required at the second tentative.
Eventually, concentration effect could be mentioned to justify the absence of trend concerning the load curves. As stated by Christopher Carrick in his thesis, the optimal concentration for cellulose beads precipitated in ethanol is 1.5%, this concentration gives the best surface homogeneity. 1% concentrated beads are more prone to swollen than higher concentrated beads, thus they could have a tendency to interpenetrate with themselves. Swollen beads have enlarged dimensions as well as a diminished cohesion, making them more flexible and soft [29].
5.2.2 Stress-strain curves and adhesion energy
Quantifying the adhesion of the cellulose-cellulose joint was easier by normalizing the load with the contact area and the exclusion of some irrelevant data as explained in part 4.4. Native beads were strong and less flexible than most of the surface modified beads. Even if the surface modification did not have the expected effect, i.e. an increase of strain, except for EXPN64 beads, the graph 4.6 shows one important thing: surface modification and charge influence the adhesion and properties of cellulose.
However, the required energy to separate the beads seems to be higher for the modified beads even if they did not show a higher strength. Compared to a simple starch coating, the addition of a layer of FennoBond 85E does not provide higher adhesion to the cellulose. On the contrary, the results are encouraging with a LbL surface modification performed with starch and EXPN64, with an energy 2 times bigger than for native beads (112 kJ vs 64.7 kJ). Moreover these beads had the highest strength (-42MPa) and an average flexibility (0.5% of deformation). Charged beads also present a great adhesion energy (87.5 kJ). However, the stress was sensibly lower in absolute values for these types of beads (around -30MPa).Finally, native beads, which are really strong, lose in adhesion energy due to their lack of deformation during the pull-off.
The absence of correlation between strength and energy, as well as the diversity of combinations for each type of beads concerning the load, the stress and the energy can be partially attributed to the contact area. Indeed, as shown in table 4.1,charged beads and starch coated ones present the higher contact area. The latter may be affected by the elasticity and the surface modification changing the surface tension and therefore having an effect on the adhesion.
5.3 Social, environmental and ethical considerations
Environmental perspective is an aspect of great importance for this project. All the experiments conducted question the sustainability in plastics making. Pollution produced by crude oil extraction and running out of natural fossil resources are the most evident proofs of the impact of Human on the planet, when producing plastics.
This way of living is not viable on a long term scale. The development of cellulose based plastics as suggested by this project could therefore slow down the climate change and the threat to the environment. Cellulose plastics are indeed made from biopolymers and are entirely degradable. Actual volume plastics such as polypropylene or polyethylene, the most commercialized plastics, degrade on the contrary very slowly causing waste plastic pollution, especially in the oceans. Another benefit of the development of such materials is the manner of processing the raw material. Regeneration of cellulose enables a green processing with very few chemicals. During the polymerization of polyvinyl chloride (PVC, another volume plastics) a toxic gas, hydrogen chloride, is released. The monomer used for the production of PVC is also supposed to be cancergenic. With biomaterials based on cellulose, health of workers and users would not be threaten anymore. Wood, from which cellulose is extracted is also a resource less dependant of the market fluctuations than crude oil.
6 Conclusion and future work
6.1 Conclusions
To conclude, this bachelor project proved the influence of different treatments on the adhesion of cellulosic surfaces. The precise impact of each chemicals was not obvious in all cases, especially concerning the load needed to separate two spheres.
However, the different experiences conducted have shown that the energy required to separate surface modified beads was higher than for native beads. Layer by Layer native coated beads with starch and EXPN64 appear to be the best combination for adhesion.
Charged beads also required higher energy than native beads and some of the modified beads. Native beads were particularly strong but had very little strain, decreasing the adhesion energy, unlike the coating offered by the starch which gave high elasticity to the beads.
Nevertheless, the relevance of the results can be questioned since fluctuations were spotted all along the way. A higher number of samples would have increased the accuracy of the results. In the same way, the adsorption time of polyelectrolytes onto the cellulose surface was probably too low, explaining the inhomogeneity of some surfaces.
The last aspect that could have been improved during the project is the drying time of the modified beads.
6.2 Future work
As a future work, that would be interesting to study the influence of the size and concentration of the beads. In this project all the beads had a 1% wt cellulose content and a diameter of 0,6mm. Changing some of these parameters and trying combination of sizes, concentrations and surface treatments would give an overview of the real benefit on adhesion of each characteristic. Another idea is also to change the charge of the beads and to try the LbL surface modification on charged beads instead of performing it on native ones. The pursuit of this work could give answers to the joint formation between two cellulosic surfaces and help to develop new sustainable and green materials with enhanced properties.
7 Acknowledgements
I would like to acknowledge my supervisor, Torbjörn Pettersson, associate professor at the division of fibre technology, for the time he spent to answer my questions, advise me and discuss my results.
I thank warmly Nadia Asta, the PhD student who explained me in details her work, supported me and helped me during the whole time of my project. I owe her a lot.
The department of Fiber and Technology is acknowledged as well for letting me used their labs and materials.
I also want to acknowledge Anders Eliasson and Anders Tilliander, respectively director of studies and associate professor at the school of industrial engineering and management, in charge of the course MH101x, for the organisation of the course and their quick answers for administrative issues.
8 References
[1] Högselius, Per. AK1213. Class Lecture, Topic: Introduction to Swedish Society.
Division of History of Science, Technology and Environment. Sept. 2019.
[2] Karlsson, Pernilla. “Swelling of Cellulose Fibrillar Matrices and Gels”. PhD thesis. KTH Royal Institute of Technology, 2019.
[3] Mekonnen, Tizazu et al. “Progress in bio-based plastics and plasticizing modifications”. In: Journal of Materials Chemistry A 1.43 (2013), pp. 13379–
13398.
[4] Brydson, J.A. “22 - Cellulose Plastics”. In: Plastics Materials (Seventh Edition).
Ed. by J.A. Brydson. Seventh Edition. Oxford: Butterworth-Heinemann, 1999, pp. 613–634. ISBN: 978-0-7506-4132-6. DOI: https : / / doi . org / 10 . 1016 / B978-075064132-6/50063-2. URL: http://www.sciencedirect.com/science/
article/pii/B9780750641326500632.
[5] Xu, Huan et al. “Coffee Grounds to Multifunctional Quantum Dots: Extreme Nanoenhancers of Polymer Biocomposites”. In: ACS Applied Materials &
Interfaces 9.33 (2017). PMID: 28770986, pp. 27972–27983. DOI: 10 . 1021 / acsami.7b09401. eprint: https://doi.org/10.1021/acsami.7b09401. URL:
https://doi.org/10.1021/acsami.7b09401.
[6] Wang, Sen, Lu, Ang, and Zhang, Lina. “Recent advances in regenerated cellulose materials”. In: Progress in Polymer Science 53 (2016), pp. 169–206. ISSN: 0079- 6700. DOI:https://doi.org/10.1016/j.progpolymsci.2015.07.003. URL:
http://www.sciencedirect.com/science/article/pii/S007967001500074X.
[7] Credou, Julie and Berthelot, Thomas. “Cellulose: from biocompatible to bioactive material”. In: J. Mater. Chem. B 2 (30 2014), pp. 4767–4788. DOI:
10.1039/C4TB00431K. URL: http://dx.doi.org/10.1039/C4TB00431K.
[8] Klemm, Dieter et al. “Cellulose: fascinating biopolymer and sustainable raw material”. In: Angewandte chemie international edition 44.22 (2005), pp. 3358–3393.
[9] Carrick, Christopher. “Macro-, Micro-and Nanospheres from Cellulose: Their Preparation, Characterization and Utilization”. PhD thesis. KTH Royal Institute of Technology, 2014.
[10] Torgnysdotter,
Annsofie. “Fibre/fibre joints: Their characterization and influence on different paper strength properties”. PhD thesis. KTH, 2006.
[11] Stipanovic, Arthur J. and Sarko, Anatole. “Packing Analysis of Carbohydrates and Polysaccharides. 6. Molecular and Crystal Structure of Regenerated Cellulose II”. In: Macromolecules 9.5 (1976), pp. 851–857. DOI: 10 . 1021 / ma60053a027. eprint: https://doi.org/10.1021/ma60053a027. URL: https:
//doi.org/10.1021/ma60053a027.
[12] Eichhorn, SJ and Davies, GR. “Modelling the crystalline deformation of native and regenerated cellulose”. In: Cellulose 13.3 (2006), pp. 291–307.
[13] Yamane, Chihiro et al. “Two different surface properties of regenerated cellulose due to structural anisotropy”. In: Polymer journal 38.8 (2006), pp. 819–826.
[14] Karlsson, Rose-Marie Pernilla et al. “Carbohydrate gel beads as model probes for quantifying non-ionic and ionic contributions behind the swelling of delignified plant fibers”. In: Journal of colloid and interface science 519 (2018), pp. 119–
129.
[15] Medronho, Bruno et al. “Rationalizing cellulose (in) solubility: reviewing basic physicochemical aspects and role of hydrophobic interactions”. In: Cellulose 19.3 (2012), pp. 581–587.
[16] Hubbe, Martin A, Gardner, Douglas J, and Shen, Wei. “Contact angles and wettability of cellulosic surfaces: A review of proposed mechanisms and test strategies”. In: BioResources 10.4 (2015), pp. 8657–8749.
[17] Nilsson, Thomas and Rowell, Roger. “Historical wood – structure and properties”. In: Journal of Cultural Heritage 13.3, Supplement (2012). Wood Science for Conservation, S5–S9. ISSN: 1296-2074. DOI:https://doi.org/10.
1016/j.culher.2012.03.016. URL: http://www.sciencedirect.com/science/
article/pii/S1296207412000659.
[18] Ishii, Daisuke, Tatsumi, Daisuke, and Matsumoto, Takayoshi. “Effect of solvent exchange on the supramolecular structure, the molecular mobility and the dissolution behavior of cellulose in LiCl/DMAc”. In: Carbohydrate Research 343.5 (2008), pp. 919–928.
[19] Nayak, Jyoti N, Chen, Yi, and Kim, Jaehwan. “Removal of impurities from cellulose films after their regeneration from cellulose dissolved in DMAc/LiCl solvent system”. In: Industrial & engineering chemistry research 47.5 (2008), pp. 1702–1706.
[20] Adamson, Arthur W, Gast, Alice Petry, et al. Physical chemistry of surfaces.
Vol. 150. Interscience publishers New York, 1967.
[21] Carrick, Christopher, Pendergraph, Samuel A, and Wågberg, Lars. “Nanometer smooth, macroscopic spherical cellulose probes for contact adhesion measurements”. In: ACS applied materials & interfaces 6.23 (2014), pp. 20928–20935.
[22] Barrat, JL and Joanny, JF. “Theory of polyelectrolyte solutions”. In: Advances in chemical physics 94 (1996), pp. 1–66.
[23] Marais, Andrew, Pendergraph, Samuel, and Wågberg, Lars. “Nanometer-thick hyaluronic acid self-assemblies with strong adhesive properties”. In: ACS applied materials & interfaces 7.28 (2015), pp. 15143–15147.
[24] Rundlöf, Mats et al. Application of the JKR method to the measurement of adhesion to Langmuir-Blodgett cellulose surfaces. 2000.
[25] Mark Rutland, Torbjörn Pettersson. KD2350. Class Lecture, Surfaces, colloids and soft matter. KTH Royal Institue of Technology. 2019.
[26] Uran, Suzana, Šafarič, Riko, and Bratina, Božidar. “Reliable and accurate release of micro-sized objects with a gripper that uses the capillary-force method”. In:
Micromachines 8.6 (2017), p. 182.
[27] Maugis, Daniel. “Adhesion of spheres: the JKR-DMT transition using a Dugdale model”. In: Journal of colloid and interface science 150.1 (1992), pp. 243–269.
[28] McDonald, J Cooper et al. “Fabrication of microfluidic systems in poly (dimethylsiloxane)”. In: ELECTROPHORESIS: An International Journal 21.1 (2000), pp. 27–40.
[29] Katz, JR. “The laws of swelling”. In: Transactions of the Faraday Society 29.140 (1933), pp. 279–297.
TRITA ITM-EX 2020:317
www.kth.se
