MedicineLakex

 

Christiani.eu

Chapter 5
Physicochemical Aspects, Properties,
T. Spychaj, K. Wilpiszewska, S. Spychaj
5.1. Starch as a biorenewable polymer feedstock
Over the last two decades, starch has been extensively studied and applied inthe manufacturing of easily processable materials of technical importance.
Starch is a biorenewable and cheap polymer; the price of starch-based plasticswas 1.25–1.40 euro/kg in 2001 and in some applications it is competitive withconventional petrochemical plastics [1]. In the same year, the world productionof biodegradable starch-based plastics was evaluated as ca. 25 000 tons [1].
Biodegradable polymeric materials are developed with the aim of a gradualreplacement of synthetic polymers based on petrochemical feedstock. Such astrategy will allow for a partial solution of environmental problems associatedwith post-consumer polymer waste.
Starch consists of two polymer components: linear amylose and branched amylopectin, in a proportion depending on the respective botanical origin(Table 5.1) [2]. Native starch is ca. 15–45% crystalline [3]; usually, only amylo-pectin takes part in the formation of the crystalline structure [4].
In order to obtain an amorphous thermoplastic mass from starch, the granular structure has to be disordered (i.e., gelatinized or destructured) [5].
When heated in the presence of water, under high-shear conditions, starchgranules swell, losing crystallinity and birefringence [1].
The glass transition temperature, Tg, and the melting temperature of dry pure starch are higher than its decomposition temperature [1]. The presence T. Spychaj, K. Wilpiszewska, S. Spychaj Table 5.1. Extrapolated values of the glass transition temperature and amylose-to-
amylopectin weight ratio in various types of starch [2]
Amylose/amylopectin (by wt) of a plasticizer lowers the Tg value, i.e., at elevated temperatures starch under-goes gelatinization rather than degradation. For instance, at a high moisturecontent (30 wt% and more), Tg of wheat starch is observed at about 50°C,followed by the crystalline melting transition [6]. The plasticizer molecules couldform hydrogen bonds with the polysaccharide chains of starch (Scheme 5.1),disrupting the inter- and intramolecular hydrogen bonds of native starch [7].
The stronger the hydrogen bonds between starch and the plasticizer, the moredifficult starch recrystallizes during storage [2].
Scheme 5.1. Formation of hydrogen bonds between starch/glycerol and starch/urea
Generally, low molecular weight substances with rather low viscosity are efficient plasticizers; they should also exhibit a high boiling point [2]. The mostcommon starch plasticizers are water and glycerol. Urea and formamide areknown starch plasticizers capable of preventing recrystallization [7]; the resultinghydrogen bonds with starch are even more stable than those formed by glycerol(the most popular starch plasticizer) [8]. However, since urea is a high-meltingsolid with little internal flexibility, urea-plasticized starch becomes rigid andbrittle [8].
Extrusion in the presence of low molecular plasticizers is the simplest method of starch thermoplasticization; the latter seems to be the most common Starch-Urethanes: Physicochemical Aspects, Properties, Application way of starch modification. Numerous reports describing process conditions,various kinds of plasticizers, as well as the properties of the final starch-basedproducts are available in the literature [3,5,7–12]. However, starch thermoplas-ticization by itself is not the subject of interest of this chapter.
The most important problems associated with conventional starch-based materials are their sensitivity to water and brittleness, which can even increaseduring storage due to a phenomenon termed retrogradation [2]. Retrogradation(spontaneous recrystallization) is caused by the tendency of polysaccharide mac-romolecules to form hydrogen bonds [2]. This phenomenon is also responsiblefor another common drawback during processing, e.g., by injection molding,namely shrinkage.
Physical and chemical (or physicochemical) methods of modification are utilized in order to transform starch into a thermoplastic material processableby conventional techniques including melt extrusion, kneading, injectionmolding, or compression molding.
In practice, there are three general ways of starch thermoplasticization: (i) chemical derivatization (substitution of the polysaccharide hydroxyl groupsby reacting with other functional moieties), (ii) extrusion in the presence oflow molecular polar plasticizers or other thermoplastic (usually biodegradable)polymer(s), and (iii) graft copolymerization.
Problems associated with native starch processing by conventional methods typical of thermoplastic polymers and the disadvantages of goods producedby thermoplasticization have led to the conclusion that blending of starch withother biodegradable thermoplastics can be appropriate. Two general ways ofblend production (mainly by extrusion) can be considered, i.e., compoundingand reactive extrusion.
In this chapter, the following biodegradable starch-polyurethane materials are described: (i) products of starch chemical derivatization by the reactionof hydroxyl groups with isocyanates, (ii) materials obtained by starch meltblending with polyurethane(s) alone or with a third biodegradable polymercomponent, (iii) urethane graft copolymers of starch, and (iv) starch-polyureth-ane foams. The last group of materials belongs to crosslinked polymers ratherthan to thermoplastics. Chemical or physicochemical aspects of starchmodification, properties of starch-polyurethane materials (including susceptib-ility to biodegradation), and relevant areas of their application are considered.
Starch-polyurethane materials are of particular importance for the following reasons: (i) susceptibility to biodegradation (Scheme 5.2), (ii) high reactivityof the hydroxyl groups of polysaccharide chains with isocyanate groups and RCOO-O-OOCR > RCOOR1 >> RNHCOOR1 ≈ RNHCOR1 > ROR1 Scheme 5.2. Susceptibility to hydrolysis of different functional groups in the heterochain
polymers [13]
T. Spychaj, K. Wilpiszewska, S. Spychaj no low molecular weight by-product formation, (iii) the reaction of starchurethanization can possibly be catalyzed (e.g., by the addition of dibutyltindilaurate or other catalysts), (iv) hydrogen bond formation between the poly-saccharide chains (i.e., OH groups) and the urethane (or urea, amide, ether,or ester) groups of the second polymer component, thus improving thecompatibilization between starch and the polyurethane components.
The term "biodegradation" is very common, but is often misused [14]. In biodegradation, enzymes of the biosphere essentially participate at least in onestep during the cleavage of the chemical bonds of the material; degradationdoes not necessarily proceed over a short period of time. Therefore, it isimportant to combine the term "biodegradable" with the specification of theparticular environment where biodegradation is expected to occur and theprocess time-scale. Composting is among the most valuable methods to assessthe polymeric material bioassimilation during biodegradability tests.
5.2. Starch-urethane polymers via derivatization of hydroxyl groups
The chemical modification of starch is a method for broadening the range ofits applications. From the chemical point of view, the addition reactions arepreferred for starch derivatization, since no by-products are formed [15]. Starch-urethane polymers can be obtained in a reaction between starch and isocyan-ates, urea or its derivatives, such as alkylureas. However, starch-urea and starch-alkylurea derivatives tend to swell or even dissolve in water [16,17]. Moreover,the substitution efficiency is rather low [17] and the reaction with urea isdifficult to control, because of the possibility of crosslinking and by-productformation [16].
The reaction of starch with isocyanates follows the general pattern of addition reactions (Scheme 5.3). Since moisture present in the reaction systemcauses transformation of the isocyanates into the corresponding substitutedureas, the process is preferably carried out under anhydrous conditions. Elevat-ed temperatures are preferred, although the reaction proceeds even at roomtemperature [18]. Some examples of this interaction are described in [15,18,19].
Early reports refer to the preparation in suspension of starch-urethane derivatives, using organic solvents, such as benzene, toluene, pyridine, R – alkyl or aryl Scheme 5.3. Scheme of reaction between starch and monoisocyanate (for a degree of
substitution, DS = 1)
Starch-Urethanes: Physicochemical Aspects, Properties, Application dimethylsulfoxide (DMSO), N,N-dimethyl formamide, N-methyl pyrrolidoneor morpholine [18,19].
The efficient chemical modification of starch requires a large amount of solvent (Figure 5.1) [15]; thus, the reaction efficiency is reduced at higher starchconcentrations, since less solvated starch hydroxyl groups are available for aneffective addition to isocyanate groups [15].
Starch concentration (mol/cm3) Figure 5.1. Relationship between starch concentration and degree of substitution; solvent
DMSO, residence time 4 min, theoretical DS = 2 [15]
The reaction between starch hydroxyl groups and the isocyanate groups is often catalyzed by dibutyltin dilaurate [20–22], but other catalysts are also used[15,23]. However, the reaction proceeds even in the absence of a catalyst [17,24].
Reported isocyanate modifiers used for starch carbamoylation are both aliphatic, such as hexamethylene diisocyanate (HMDI) [18] and monoisocyan-ates containing 7 to 18 carbon atoms in the aliphatic chain [15,17], andaromatic, such as tolylene 2,4-diisocyanate (TDI) [19], methylene–4,4′-bisphenyldiisocyanate (MDI) [21], phenyl diisocyanate (PI) [18,20], toluene poly(prop-ylene oxide) diisocyanate [24], etc. In the past, aromatic isocyanates were mainlyused for the preparation of starch-urethane derivatives [25,26]. The reactionrate of starch urethanization depends on the nature of the reaction medium,the temperature (Figure 5.2), the kind of isocyanate used (Figure 5.3), thedegree of starch pregelatinization, and the presence of catalysts [25].
The properties of starch-urethane derivatives are strongly influenced by the degree of substitution as well as by the type and the chain length of theattached aliphatic or aromatic substituent. However, the use of diisocyanates,e.g., HMDI and TDI, leads to starch crosslinking. In this case, the addition ofeven a small amount of modifier changes the product properties.
T. Spychaj, K. Wilpiszewska, S. Spychaj Reaction time (h) Figure 5.2. Effect of temperature and time on the rate of the uncatalyzed reaction of phenyl
isocyanate with corn starch granules in pyridine (a two-fold excess of isocyanate is required
to get the maximum degree of substitution, i.e., DS = 3) [25]
Reaction time (h) Figure 5.3. Uncatalyzed reaction rate of various isocyanates with corn starch granules in
pyridine at 80°C [25]
Starch-Urethanes: Physicochemical Aspects, Properties, Application Starch-urethane derivatives of HMDI that contain at least one urethane group for each polysaccharide unit exhibit limited swelling [17,18]. One of thepotential applications of such products is dusting powder for surgical gloves;the powder will not become sticky even after sterilization at 120°C for 1 h [18].
It was found that the product of the starch and phenyl isocyanate reaction with DS = 0.2 exhibited no toxicity to cultures of Aspergillus oryzae, Penicilliumexpansum and A. Niger, that were grown well on a medium containing thisstarch derivative as a carbon source [26].
Crosslinking of starch chains could be avoided despite the use of a diisocyan- ate by blocking one of the −NCO groups through a reaction with substancescontaining a free hydroxyl, amine, or carboxyl group [19]. Non-symmetricaldiisocyanates are preferred, as in this case one of the isocyanate groups is farmore reactive. For instance, in the case of tolylene 2,4-diisocyanate, the groupin position 4 is about ten times more reactive than that in position 2 [19].
Monoisocyanates with alkyl chains of different length are useful reagents, imparting hydrophobicity and thermoplasticity to starch materials (Tables 5.2and 5.3) [15,27].
Table 5.2. Experimental and theoretical values of DS, yield, T , and melt flow data of starch
carbamates with different alkyl chain lengths [15] Degree of substitution melting, gas formation melting, gas formation melting, gas formation melting, gas formation melting, gas formation melting, gas formation melting, gas formation melting, gas formation The hydrophobicity of the modified starch-urethanes could be determined by contact angle measurements [20,28,29]. For the values of 90° and up(in water), the surface is assumed to be hydrophobic [23]. The contact anglevalue of starch-urethanes containing 2–18 carbon atoms in the alkyl chain(DS = 1.6) was found to be slightly above 100° (Table 5.4) [23].
T. Spychaj, K. Wilpiszewska, S. Spychaj Table 5.3. Experimental and theoretical values of DS and melt flow properties of urethane-
urethane, urethane-urea and urethane-amide starch derivatives [27]
Starch derivative based on Degree of substitution modifier (alcohol, amineor carboxylic acid) Table 5.4. Contact angle values of starch-urethanes with different alkyl chain length [23]
Contact angle (°) The surface of waxy maize starch nanocrystals obtained by hydrolysis of native waxy maize starch granules with sulfuric acid has been modified byphenyl isocyanate [22]. It has been demonstrated by contact angle measure-ments that the modification of this form of starch also led to more hydrophobicparticles (ca. 60°, as compared to 35° for unmodified starch nanocrystals) [22].
The data available lead to the conclusion that the longer the introduced alkyl chain, the better the hydrophobic properties of the starch derivatives.
Generally, starch-urethane derivatives exhibit a very low solubility in organicsolvents [15,18]. Moreover, melting was observed for aliphatic derivatives withhigher DS values (1.0 and above) (Tables 5.2 and 5.3) [15,27,30]. The efficien-cies of modifications by alkyl isocyanates performed in DMSO suspension werefound to be high, up to 97% [15].
Similar starch derivatives were prepared in an apparatus allowing a more intensive mixing than that in a glass flask [15]. Pea starch and undecyl isocyan-ate were reacted in DMSO with dibutyltin dilaurate in a kneader. Despitescaling up the process, the determined DS = 1.8 (theoretical DS = 2.0) was Starch-Urethanes: Physicochemical Aspects, Properties, Application comparable to that of a product obtained in the glass flask. The investigationsof aliphatic starch-urethane derivative formation in a DMSO slurry (in thepresence of dibutyltin dilaurate as a catalyst) demonstrated that the reactiontime ranges are favorable for reactive extrusion (Figure 5.4), which enablescontinuous processing [15]. The extruder plays the role of a stable pump,ensuring constant throughput and efficient mixing even for highly viscous media[31]. Twin-screw extruders were found to be especially effective continuousreactors. The extruder efficiency depends mainly on the design of the screw(s).
The extruder having seven separately heated zones with a temperature profilefrom the feed to the die, e.g., 60/100/120/120/130/140/140°C [15], can be quiteefficient. No higher temperatures were recommended because of the productdiscoloration.
Reaction time (min) Figure 5.4. Time dependence of the starch-urethane formation; reaction of 1-undecyl
isocyanate with wrinkled pea starch in DMSO in the presence of dibutyltin dilaurate catalyst;
theoretical DS = 2 [15]
The application of a reactive extrusion manufacturing process to starch- urethanes is associated with some technical problems. The reaction efficiencieswere found to be low (e.g., a DS of 0.02 was obtained, whereas the theoreticalvalue was 0.5) as a result of the low viscosity of the monoisocyanates usedand the incompatibility of the components, i.e., of ineffective mixing [15].
5.3. Starch/polyurethane blends
The blending of starch granules with synthetic polymers generally results in afiller effect on the polymer properties [32–34]. As the starch content increasedin such polymer blends, there was a decrease in the final material elongationat break and tensile strength [35,36], whereas the modulus was increased [37].
The main reason for the observed decrease in the mechanical properties ofstarch/synthetic polymer blends may be explained by the different properties ofstarch (a hydrophilic polymer) and synthetic (usually hydrophobic) polymers.
T. Spychaj, K. Wilpiszewska, S. Spychaj Polyurethane (PU), a unique polymeric material with interesting physical and chemical properties, has been extensively tailored to meet the highly diver-sified demands of modern technologies, such as coatings, adhesives, fibers, foams,and thermoplastic elastomers [29,38]. The development of waterborne poly-urethane or poly(urethane-urea) formulations has increased dramatically inthe last fifteen years in view of cost reduction and environmental reasons [39].
Tighzert et al. [28,29] developed biodegradable polymeric materials based on thermoplastic starch and a polyurethane aqueous dispersion. PU aqueousdispersions were synthesized from castor or rape seed oil-based polyols.
Isophorone diisocyanate and dimethylol propionic acid were used as the residualsubstrates for the synthesis of aqueous PU dispersions. The effect of the PUcontent on the morphology, miscibility, and physical properties of the resultingmaterials (films [29] or extruded strands [28]) was investigated by scanningelectron microscopy (SEM), differential scanning calorimetry (DSC), dynamicmechanical thermal analysis (DMTA), mechanical property measurements, andwater sensitivity. The results demonstrated that glycerol-plasticized starch canbe mixed with rape seed oil-based waterborne PU on the molecular level whenthe PU content is lower than 20 wt% [29] (for castor oil-based waterborne PU,lower than 15 wt% [28]), whereas phase separation occurs at higher PUcontents. The addition of PU (4–20 wt%) to glycerol-thermoplasticized starchresulted in blends with improved Young's modulus (40–75 MPa), tensilestrength (3.4–5.1 MPa), and elongation at break (116–176%) [28]. Themechanical properties of the respective blends are listed in Table 5.5.
Starch/PU films based on PU with rape seed oil polyol exhibited much higher values of elongation at break (85–480%), toughness (1.8–7.1 MPa), andtensile strength (2.8–4.1 MPa) [29] than neat thermoplastic glycerol-plasticized Table 5.5. Mechanical properties* of thermoplastic starch (TPS)**, polyurethane (PU) and
starch/polyurethane (TPS/PU) blends [28]
*Mechanical data measured after 2 weeks of aging**Glycerol-plasticized starch Starch-Urethanes: Physicochemical Aspects, Properties, Application starch. The surface properties (contact angle) and water absorption of thestarch/PU blends as a function of PU content in the polymeric material areshown in Figures 5.5 and 5.6, respectively. It can be seen that the introductionof PU into the plasticized starch matrix leads to an improvement of the polymersurface and bulk hydrophobicity and a better water resistance of the resultingmaterials [28,29].
Figure 5.5. Contact angle of TPS/PU blends vs. PU content [28]
Transparent sheets made of polymer blends of thermoplastic starch and waterborne polyurethanes were prepared by compression molding and studied byWu and Zhang [40]. These materials were obtained from a polyester-type water-borne polyurethane based on poly(butylene glycol adipate) (Mw = 2150 g/mol),dimethylol propionic acid, tolylene-2,4-diisocyanate, and thermoplastic starch(no data concerning a plasticizer were provided). The results showed that thetensile strength (ca. 30–35 MPa), elongation at break (ca. 5–40%), and waterresistance (317→34 wt% of water) of the TPS/PU sheets were all improved, ascompared to TPS, when the PU content was varied in the range of 5–30 wt%.
Infrared (IR), X-ray diffraction (XRD), DSC and SEM analyses of the samplesindicated that an interaction took place between TPS and PU in the obtainedsheets, resulting in a certain level of miscibility. The sheets also revealed a highercrystallinity than PU and amorphous TPS and a slightly lower lighttransmittance than TPS (decrease from ca. 95% for TPS to 85–88% forTPS/PU blends), suggesting a partial recrystallization of starch. The authors Chapter 8
Natural and Man-Made Cellulose
K. P. Mieck, T. Reußmann, A. Nechwatal
The achievement of better properties by combining different materials has been"practiced" by Mother Nature long before it occurred to humans, as it isdemonstrated by countless examples. For instance, in wood, plants or muscles,stiff fibers are embedded in a softer matrix. The first man-made compositeswere reed-reinforced Nile-mud stones (1450 BC).
For some years, the interest in cellulose natural and man-made fibers for technical use is constantly growing. Apart from ecological and economic reasons,the natural fibers stress-strain behavior suggests their use in composites.
8.2. Opening of cellulose natural fibers and manufacture of cellulose
8.2.1. Natural fibers
Natural fibers can be extracted from a great variety of plants. They can begrouped according to the plant part source (Figure 8.1).
Figure 8.1 shows that in the "Bast fiber" group, flax, hemp, ramie, jute, and kenaf species are most used for plastic reinforcement because of their goodmechanical properties. Nettle fibers are also suitable, but this type has notfound so far industrial application because of the low fiber content of nettleplants. In the "Leaf fiber" group, sisal and pineapple fibers are used in practicalapplications. Cotton and wood fibers are also suitable for the preparation ofcomposites.
K. P. Mieck, T. Reußmann, A. Nechwatal Figure 8.1. Groups of natural fibers
Bast and leaf fibers are produced by plant stalks or leafs. The fibers must be separated as gently as possible from the vegetable glues and woodcomponents. For fiber separation, quite different biological, chemical andmechanical procedures, and combined processes are possible [1]. In recent years,ultrasonic separation, steam explosion decomposition, and the Duralin processhave been reported.
Fiber separation strongly influences the fiber yield and properties (fineness, tenacity, length). The gain of large amounts of clean and wood-free fibers inan inexpensive way is the essential problem in their separation from plants.
The common procedures are the dew retting and mechanical opening processes.
The microwave treatment, the steam explosion separation or the Duralin processcan avoid the risks of conventional retting and result in more valuable, longerfibers. In addition to a better singling, more uniform fiber properties canbe achieved. However, the higher costs of these processes compared to dewretting/mechanical opening have prevented their practical application.
The different vegetable natural fibers show a similar morphological and chemical constitution. They consist of 60–80% cellulose and 15–40% vegetableglue substances (pectins, hemicelluloses, and lignin). The fine structure dependsboth on the type (bast or leaf fiber) and on the conditions during the plantgrowth, the harvest, and the separation processes. Considering the plant growth,the plant species and location, the climate, the soil conditions, and fertilizersare significant factors, while during harvest and separation, the maturity ofthe plant, the weather, the harvesting technique, and the opening processinfluence the fiber properties.
The fine structure and the cellulose content determine both the fiber properties (stress-strain behavior, E-modulus, moisture uptake) and the furtherprocesses leading to textile fiber slivers, technical yarns or fabrics, or non-wovens. During the last decades, extensive studies of the structure and chemicalconstitution of natural fibers have been reported in the literature.
Natural and Man-Made Cellulose Fiber-Reinforced Composites 8.2.2. Cellulose man-made fibers
Fluctuating mechanical properties and odor create problems in the use ofnatural fibers for plastics reinforcement and techniques for their avoidance areknown. Still, the concept of ecological composites containing cellulose man-made fibers is obvious. However, compared to natural fibers, their higherelongation and lower E-modulus are disadvantageous.
The fact that cellulose can be dissolved physically in N-methylmorpholine- N-oxide (NMMNO) has been known since the 1930s. This finding is used on acommercial scale in the new ecological lyocell process [2]. Fibers and filamentyarns produced by the viscose (rayon) process have also been used in composites[3]. However, compared to the new lyocell process, the viscose process createssome problems from the viewpoint of industrial medicine and ecology. Figure 8.2shows the principles of both processes. In the classical viscose process, cellulosemust be alkalized and then derivatized by carbon disulfide; the spinning bathis sulfuric acid. On the other hand, the lyocell process comprises only a littlenumber of steps and the NMMNO solvent is recycled; the pollution level ofthe leaving air and water is very low.
Figure 8.2. Schematic representation of the rayon and lyocell processes; COD – chemical
oxygen demand
There are different reasons for the replacement of natural fibers by cellulose man-made fibers in composites: (i) the general advantages of cellulosic fibers(eco-image, low density, more hygienic workplaces, lower abrasion of theprocessing equipment, no clinker in waste incineration plants) can be claimed K. P. Mieck, T. Reußmann, A. Nechwatal for both natural and man-made fibers; (ii) both types do basically increasethe composite mechanical performance; (iii) man-made fibers are produced inall outfits that are common to the classical reinforcing fibers, such as glass,aramid, and carbon; (iv) they show very low deviations in mechanical properties(tenacity, modulus, etc.) and in fineness, density, and length distribution withina batch and between different batches; (v) when suitable sizes are used, man-made fibers do not cause any odor problems.
In addition to their lower E-modulus, man-made fibers differ in stress-strain behavior compared to the classical reinforcing glass, aramid, and carbon fibers;starting with a steep slope, the curve goes gradually to the breaking point.
However, both conventional reinforcing fibers and natural fibers show a quasi-linear course. The reason for this is the specific amorphous-crystalline structureof cellulose. Therefore, research was carried out in order to increase the E-modulus of man-made fibers and to adjust their stress-strain curve to a quasi-linear course [4].
8.3. Measurement of the fiber properties
In the literature, there is abundant information concerning the mechanical-physi-cal properties of natural fibers. However, there is a wide scatter in the reportedvalues and, what is more, the test methods are often not described. Therefore,it is difficult to find reliable values concerning the mechanical performance.
Compared to classical material testing procedures, the measurement of the properties of yarn or fiber products that are commonly used as reinforcing com-ponents requires a particular approach. The dependences on a large numberof parameters, e.g., temperature, moisture, testing time and rate, or inner andouter non-uniformity in morphology, complicate the measurements of the fiberproperties; this is also the case of glass, carbon, and aramid fibers. Due totheir specific morphological characteristics, natural fibers pose additionalproblems during testing [5].
8.3.1. The stress-strain test
Two approaches are established for measuring fiber-like and thread-like objects:the samples are subjected to the tensile test either as a single fiber or as afiber bundle. For these tests, a special equipment is necessary, with a suitablemeasuring range and high precision. Well known apparatuses are the Vibrodyn(Lenzing) or Fafegraph (Textechno Dr. Stein) types. In principle, the price forthe high precision of the single fiber test is the long measurement time.
The fiber bundle test has the great advantage of being fast. Under some special conditions, e.g., a pre-stress of all fibers, distributing uniformly thetension force among them, values with sufficient reliability can be reached. Here,both the clamps type and shape, as well as the clamp-fiber interface play animportant role.
The fiber bundle tenacity is influenced by the statistical distribution of the single fiber elongation at break and by the shape of the stress-strain curve.
Natural and Man-Made Cellulose Fiber-Reinforced Composites It is known that the fiber bundle tenacity drops with broader statistical distribu-tions of the single fiber elongation at break and with lower single fiber tenacity.
This drop increases when the stress-strain curve of the single fiber takes a verysteep course in the upper range.
The test method used (single fiber or bundle) depends on the equipment and the gained experience in the respective laboratory. It is noteworthy thatthe specification of the test conditions is required when reporting test results.
The following discussion relates to single fiber tests (at L0 = 10 mm) according to DIN EN ISO 5079 since the interpretation of the stress-straincurves seemed to be important for composite preparation. From a strictlyobjective point of view, the measurements performed on natural fibers are alsobundle tests because mechanically open fibers (also termed technical fibers)were generally tested and they are fiber bundles.
The fineness is tested by means of the gravimetric procedure according to DIN EN ISO 1973, considering the specific features of natural fibers. Inprinciple, the parallelization and combing out are avoided. At different spotsof the whole fiber amount, samples of N > 50 fibers (without conspicuous crosssections) are taken for fineness and tension tests. The cutting length dependson the clamping length. It comes to 30 mm for a clamping length L0 = 10 mm.
Every single fiber used in the tension test is weighed and so the fineness ofevery fiber is calculated. If necessary, the average fineness can be calculatedby the measured single values.
The dependence of the tenacity on the test clamping length is a well known fact. Figure 8.3 shows the fineness-related tenacity as a function of the clampinglength of some conditioned, technical natural fibers.
Figure 8.3. Fiber tenacity vs. clamping length
In the considered range of 1–20 mm, there are some differences between the natural fibers concerning the effect of the clamping length on their tenacity.
These differences are caused by the specific fiber structures that create differentinhomogeneities.
K. P. Mieck, T. Reußmann, A. Nechwatal Previous research showed that flax single fibers are more influenced by the clamping length than the technical flax fibers. Furthermore, an essentialdifference was found in the wet-state tenacity vs. clamping length curves of aflax single fiber and the technical flax fiber. The wet tenacity was not higherthan the tenacity in the conditioned state. This result is in agreement withliterature data. Only a twisted yarn structure showed a higher wet tenacity,as compared to the dry tenacity.
The average wet-to-dry tenacity ratio of the single fiber should be ap- proximately 89%, taking into account the confidence interval of the measure-ments. Technical fibers show similar results only for a low clamping length of1 mm. The tenacity drops with increasing clamping length. The strong dropof the wet tenacity of technical fibers as dependent on the clamping length isdue to the fact that the number of single fiber ends between the clampsincreases. The higher numbers of defects inside the technical fiber related tothe softening and the pectin-swelling action of water cause slippage of the singlefiber during tension stress.
Studies of the influence of the clamping length have also shown that the measured elongation does not depend exactly on the clamp construction,surface, pressure, and material properties. In order to mark exactly the fibermaterial, an elongation correction proves necessary, particularly for the clamp-ing lengths L ≦ 10 mm used in the testing of natural fibers.
The elongation inexactness is based on the observation that a certain part of the fiber is pulled out of the clamp and therefore the fiber length changesby ∆Ltot [10]. Under this condition, a clamping defect, ∆Lc, can be calculated:the total length change, ∆Ltot, is plotted against the clamping length, L0,(Figure 8.4) and the extrapolation of the respective straight line to L0 = 0 results in the clamp defect, ∆Lc.
Table 8.1. shows the calculated clamping defects, ∆Lc, for some natural fibers using this procedure, the non-corrected elongation, εn, and the correctedelongation, εc for L0 = 10 mm.
Clamping length, L0 (mm) Figure 8.4. Total length change, ∆Ltot vs. clamping length, L0
Natural and Man-Made Cellulose Fiber-Reinforced Composites Table 8.1. Calculated clamping defects, ∆Lc, for some natural fibers
∆Ltot (mm) ∆Lc (mm) Retted flax 2000, Saneco sliver Retted flax 1999, Mielsdorf sliver Green flax 1999 (mechanically open) Green hemp 1999 (mechanically open) The clamping defects shown in Table 8.1 are valid only for the tested materials. They are generally not valid for all fibers of this type. For an exactmaterial characterization, the clamping defects have to be measured separatelyfor every provenance. As shown below, this correction is especially importantfor modulus tests.
In textile testing, the tensile strength, F (cN), measured in tension tests is usually related to the measurable fiber or thread fineness, Tt (tex), and isgiven as fineness-related tenacity, f. The tension, σ, required for materialcharacterization and further calculations, results from the relationship betweenthe tensile strength and the cross section area, A. Assuming that the crosssection is circular or almost circular, A can be calculated as the quotient ofthe fineness, Tt, and the density, ρ (g/cm3). The tension, σ (N/mm2), can becalculated by Eq. (8.1).
Regarding Eq. (8.1), a certain inexactness should be taken into account because the cross sections of natural fibers are not circular. However, inves-tigations have shown that the correlation between the microscopic measure-ments and the (idealized) cross section areas calcutated by the fiber finenessis very high (r = 0.916). Because of the porosity of natural fibers, the calculatedareas are smaller than those microscopically measured; the respective valueswould be closer when porosity is taken into account. The estimation of theareas from the fiber fineness seems to be justified because only the load-bearingcross section is decisive for the calculation of the tension.
8.3.2. Stress-strain behavior of natural fibers
There are several models for the calculation of the tension and the E-modulus offiber composites. All they assume that the fiber stress-strain behavior is linearuntil the break. However, this assumption is not quite correct. Certainly, the simpleobservation of the stress-strain curves of the classical reinforcing glass, carbon, K. P. Mieck, T. Reußmann, A. Nechwatal and aramid fibers suggests linearity. However, the closer inspection suggests aquasi-linear behavior. Natural fibers do fit this pattern because they show nearlylinear stress-strain curves, as it can be seen in Figure 8.5.
Figure 8.5. Stress-strain behavior of some natural fibers
However, more precise studies using the differentiated tenacity-elongation curves have shown that these fibers also have elongation-dependent moduli,particularly in the initial stage. Thus, there is a principle possibility to calculatethe initial modulus by extrapolation of σ′(ε) for ε→0. However, this extrapola-tion is related to some uncertainties because of some effects at low elongations,e.g., clamping defects, achievement of the required testing rate, etc. Therefore,both the tension-elongation curve and the initial modulus would be fraughtwith large uncertainties. Therefore, our starting point in the calculation of anE-modulus was the general equation defining a nominal (i.e., related to theinitial cross section area) tension – nominal elongation secant modulus accordingto Figure 8.6.
Figure 8.6. Tensile strength vs. length change dependence
Natural and Man-Made Cellulose Fiber-Reinforced Composites If point (F1,L1) is placed at the initial point of the tensile strength vs. length change curve, the force F1 would turn into FV and F2 would turn generally into F; the length L1 would turn into LFV and L2 would turn into LF; the lengthdifference ∆L would turn into ∆LF.Then, an initial secant modulus (Eq. 8.3)can be assumed.
Because FV is already effective on fibers and threads during clamping at ∆L = 0, it follows from F1 = 0, F2 = F, L1 = LFV, and L2 = LF that the modulusvalue is not absolutely exact, but can be calculated from the test results(Eq. 8.4).
The modulus values calculated from Eqs. (8.2) and (8.3) can give a practical measure for the gradient at the relevant point of the σ vs. ε curve, dependingon the selected limits.
This secant modulus was calculated for different natural fibers using Eq.
(8.2) in the ranges of a nominal elongation, ε of 0.1–0.5% and 0.2–0.7%. In asimilar way, the initial secant modulus values were calculated using Eq. (8.4)for F = Fmax and ∆LF = ∆LF and for the same natural fibers. Figure 8.7 shows the dependence of M*NNAS on MNNS/0.1–0.5; this is a linear correlation ofthe form (Eq. 8.5): = 1.1906 M max, ∆LFmax Figure 8.7. M*NNAS, F
K. P. Mieck, T. Reußmann, A. Nechwatal The significant correlation between the different modulus values is charac- terized by a statistical certainty C = R² = 0.8921. Comparison with the modulusMNNS/0.2–0.7 would result in an analogous close correlation. The relation betweenthe modulus in the initial part and the modulus values calculated from thefinal part of the σ vs. ε curve (assuming a strong linearity) was clear for thetested natural fibers.
Taking into account (i) the fundamental restrictions in the definition of an E-modulus, (ii) the quasi-linear trend of the σ vs. ε curve, (iii) the possibil-ity to obtain an average value for the entire elongation dependence, and (iv)the relatively low total elongation of natural fibers, the modulus calculatedusing Eq. (8.4) with F = Fmax and ∆LF = ∆LFmax can be assumed as the E-modulus. Thus, the definition of the E-modulus can be expressed by Eq. (8.6) where En is identified by the subscript n as a non-corrected modulus. Apartfrom the practical definition of an E-modulus, the consideration of theelongation correction seems essential (Eq. 8.7).
The modulus values of Eqs. (8.6) and (8.7) differ in the factor cording to experimental measurements, this factor can reach values in the rangeof 0.3–0.75. Figure 8.8 shows modulus values of natural fibers that are often Figure 8.8. Ec vs. En dependence for various natural fibers
Natural and Man-Made Cellulose Fiber-Reinforced Composites used in composites calculated by Eqs. (8.6) and (8.7). The formation of"clusters" of modulus values for the different natural fiber species can be clearlyseen. Figure 8.8 also gives an impression of the fluctuation range inside a fiberspecies.
8.4.1. Natural fibers
Table 8.2 shows average mechanical parameters of some natural fibers. Thetension, σ, and E-modulus values, which are important for composites, are givenin Figure 8.9 and compared to those of glass. The ranges hinted in Figure 8.9give an idea of the fluctuation of the fiber characteristics, as well. The lines of0.5%, 1%, and 2% mark the connection between σ and ε if Hook's law (σ = E.ε)is valid.
Figure 8.9. Fiber tenacity vs. fiber E-modulus
The tenacity of technical natural fibers is between 0.5 and 0.9 kN/mm2, i.e., they do not reach the strength of glass fibers. The Young's modulus isfound between 10 and 90 kN/mm2; sisal shows the lowest and green hemp —the highest values. Thus, the Young's modulus of some natural fibers iscomparable to that of glass fibers (ca. 73 kN/mm2).
Figure 8.10 shows that the dependence of the tenacity on the fiber diameter is essential for the critical fiber length and the interfacial shear strength. Theaverage fiber diameter of technical natural fibers is in the range of 40–150 µm,i.e., higher by up to one order of magnitude than that of glass fibers.
Separation down to single fibers may result in higher tensile strength values for flax, ramie or similar natural fibers. However, the related additional costand the lower wetting by the matrix make this separation inappropriate.

Source: http://www.christiani.eu/pdf/92899_probe.pdf