Printer Friendly

A thermoplastic elastomer produced by the bacterium Pseudomonas oleovorans.

The existence of elastomers in plants has been known since the 1500s. Natural rubber continues as a main source of elastomeric material even with the advent of many synthetic elastomers due in part to the renewable resource, plants.

Since the 1920s polyesters, generally known as poly([beta]hydroxyalkanoates), (PHAs), have been known to accumulate in various bacteria. The polymer accumulates in intracellular inclusion bodies and was determined to be a reserve carbon and energy source for the bacteria (ref. 1). Researchers found that polymer accumulation was triggered when the environment of the bacteria either lacked, or was limited in, an essential nutrient (such as nitrogen, in the form of ammonium) but had an excess of a carbon food source (such as glucose) (ref. 1). The phenomenon is similar to a bear accumulating fat because of the environmental stress caused by the approach of winter.

The highly studied bacterium Alcaligenes eutrophus, produces a brittle thermoplastic poly([beta]-hydroxybutyrate), (PHB), when grown on glucose. Improvements to the polymer properties were achieved by coaxing the bacteria into producing a copolymer. When grown on a mixture of glucose and propionic acid, a less crystalline, more flexible, tougher thermoplastic, poly([beta]-hydroxybutyrate-co-[beta]-hydroxyvalerate), (PHB/HV) resulted with properties similar to polypropylene (ref. 2).

Because PHAs are naturally produced and degraded by bacteria in response to environmental changes, the polymer is inherently biodegradable. Biodegradability of the polymer after extraction from the bacteria and formation into a desired product has been shown to occur in such environments as soil, aerobic and anaerobic sewage sludge and seawater (ref. 2). PHAs are also considered to be biocompatible (tel. 2). Potentially these unusual polymers could be used for specialty medical applications. Research in the field of bacterial polyesters continues as more bacteria are found which accumulate polymer and different carbon food sources which introduce the production of novel PHAs (ref. 3 and 4).

The bacterium, Pseudornonas oleovorans, has the ability to grow on a variety of long chain carbon sources such as fatty acids, hence the name 'Oleo'-fat, 'vorans'-eater. The PHAs produced contain a long pendant group in each repeat unit. However, even when P. oleovorans is grown on a single carbon source such as sodium octanoate, the metabolism of the bacteria can either add or cleave off two carbons from the food source molecule and produces not a homopolymer but rather a random copolymer (ref. 5). The repeat unit structure and composition of this polymer known as poly([beta]-hydroxyoetanoate), (PHO), is shown in figure 1.

These pendant groups are located on a chiral carbon and are always in the same stereo configuration, which implies the polymer is isotaetic and is therefore capable of crystallizing. The combination of a [T.sub.g] below room temperature (measured at -35[degrees]C) which is due to the long pendant groups and a low degree of crystallinity [estimated at 30% (ref. 6)] due to the various lengths of these pendant groups results in a polymer which exhibits elastomeric behavior at room temperature with the crystalline regions acting as physical crosslinks.

Phase separated morphology is the rule for thermoplastic elastomers, TPEs, and occurs because the majority of TPEs are block copolymers or blends of incompatible polymers (ref. 7). The random copolymer structure of PHO is unique among TPEs yet forms the needed phase separated morphology (crystalline and amorphous regions) required to achieve elastomeric properties in physically crosslinked systems.

In the present study, the physical properties of PHO produced in a fed batch type biosynthesis were evaluated. The equilibrium modulus obtained from stress relaxation studies was used to elucidate the network structure of PHO. Stress-strain properties, hardness and tensile set testing was conducted on PHO film crystallized at room temperature from the melt. Because a large tensile set was observed for PHO, thermal analysis of stretched samples was conducted to determine the cause of the tensile set (ref. 9). In addition, both a short and long term crystallization study was conducted to determine the influence of the crystallization temperature on the rate of crystallization and on the mechanical properties (ref. 10).

For comparative purposes, examples from each of the six different classes of commercially available thermoplastic elastomers (styrenics, urethanes, copolyesters, amides, olefinics, elastomeric alloys), both block copolymers and polymer blends were included in the mechanical evaluation along side PHO. A comparison of the mechanical properties of PHO to commercial TPEs assesses the validity of classifying PHO as a TPE in light of its unique chemical structure compared to all other TPEs.


Biosynthesis of PHO

PHO was produced by P. oleovorans using sodium octanoate as the sole carbon food source in a fed batch type fermentation process (ref. 8). The typical polymer yield for a 24 hour fed batch fermentation in a 12 liter vessel was 20 grams of polymer.

Preparation of films

Films 1 to 1.6 mm thick were prepared from PHO by either melt blending the polymer in a glass casting dish (tensile testing samples) or solvent casting films from a filtered chloroform solution, allowing the solvent to slowly evaporate over 10 days, and then melting the film (crystallization study samples). Crystallization of all the polymer films was controlled by placing the casting dishes containing the melted films in specific temperature baths for certain time periods prior to testing.

The commercially available TPEs included in this study are listed in table 1. Films 1.6 mm thick were formed using a hot press equipped with a vacuum chamber. The molding temperature was chosen according to published literature received from the different companies. The film was removed from the press and allowed to slowly cool to room temperature. A sample of a urethane TPE could not be found in a compression moldable grade. Therefore, a 1.6 mm thick film was formed from a solution processible grade cast from a 2% THF solution.

DSC sample preparation for crystallization studies

Prior to melting the solvent cast films, the required number of 10 mg samples for differential scanning calorimetry (DSC) analysis were punched from the films and encapsulated in DSC pans. These prepared samples were melted along with the films and also placed in the same temperature baths. For the short term crystallization study, samples were removed after 24 hours and a DSC scan obtained. For the long term study, samples were retrieved from the temperature baths periodically and evaluated using the DSC.


Tensile testing was based on the procedures of ASTM D638 - sample geometries: dumbbell Type V, strain rate: two, four and eight min.-l; tings per ASTM D-412 Type 2, strain rate: 1 min.-[1]

Grip slippage was a problem with the dumbbell and strip geometries even when pneumatic grips with sandpaper were used to hold the samples. No extension meter was available, so to measure elongation, a 25 mm gage length was marked on a sample of each material and a ruler held next to the sample during testing. A mark was made on the chart paper when various known elongations of the gage length were reached. A linear correlation was found between the actual and recorded elongation and a correction could then be made to the recorded elongation data for the remaining samples of that material.

The ring geometry was used for samples in the long term crystallized study to circumvent the grip slippage problems observed during initial tensile testing.

Stress relaxation/tensile set testing was based on the procedures of ASTM D-412 - sample geometries: strips 5 mm wide, 40 mm long (small strain); dumbbell per ASTM D-638 type V (large strain). Hardness measurements were based on the procedures of ASTM D-2240

Results and Discussion


The consistency of the polymer produced by fed batch fermentation biosynthesis is reflected in the reproducible composition, molecular weight distribution, thermal transition temperatures and decomposition temperature as shown by the average values and coefficient of variation for these parameters for eight biosynthesis runs shown in table 2. The decomposition temperature did not vary significantly between a nitrogen and air environment indicating thermal decomposition was not oxidative in nature. In summary, P. oleovorans produced a copolymer with consistent composition and molecular weight distribution. Controlling these features resulted in a polymer with consistent thermal transition and decomposition temperatures.

Crystallization rate

The radial growth rate of spherulites is often used to measure crystallization rate. However, PHO does not exhibit spherulitic texture, so in order to determine crystallization rate all samples were melted, then allowed to crystallize at different temperatures for 24 hours. After 24 hours a DSC thermogram was obtained to determine the level of crystallinity that had developed in the 24 hour period by measuring the heat of fusion, [delta H.sub.m], a proportional indication of the amount of crystallinity. Because all samples were compared after 24 hours and because crystallization was not complete in that period of time, relative rates of crystallization could be determined.

The results of the short term crystallization experiment are depicted in figure 2. As expected, a bell shape curve resulted in the [delta H.sub.m] versus crystallization temperature curve with a maximum at approximately the median between the glass transition temperature at -35[Degrees]C and the melting temperature at 61[Degrees]C. The location of the peak indicates that the fastest crystallization rate occurred at 0 to 5[Degrees]C.

At temperatures above 25[Degrees]C, scatter in the data was noted. The three open symbols indicate data points from single samples. The solid symbols indicate an average of the values from three to six samples. The extent of crystallinity in the majority of samples crystallized from 25[Degrees]C to 37[Degrees]C was higher than expected. Impurities can act as nucleating agents and even though prior to casting films the polymer solutions were filtered, all impurities may not have been removed. These undesired nucleating agents would have a more profound effect at the higher temperatures where nucleation can limit the rate of crystallization. These nuclei would enable a higher than expected extent of crystallization to occur during the 24 hours, thereby contributing to the data scatter.

Crystallization at lower temperatures is generally limited by chain mobility and not by nucleation, so impurities acting as nucleation sites would not have as much effect on the extent of crystallization in 24 hours. In this case data scatter would not be expected, and in fact, was not observed. Annealing effects, which could be argued as a reason for the higher than expected levels of crystallinity at the higher temperatures, were not considered significant neither during the DSC scan nor during the initial 24 hours of crystallization, as will be shown in the long term crystallization study results.

The melting temperature varied with crystallization temperature in a manner common for all polymers. The line shown through the data was determined by linear regression, and the intersection of this line with the theoretical [T.sub.m] = [T.sub.c] line was used to estimate the equilibrium melting point. The equilibrium melting point, the expected melting temperature if crystallization was thermodynamically controlled instead of kinetically controlled, was determined to be approximately 68[Degrees]C using this construction.

Long term crystallization study

Since the crystalline regions of PHO are acting as the physical crosslinks for this elastomer, a study on the effect of differing thermal histories on the mechanical properties of PHO was conducted. PHO films were crystallized at three different temperatures -20[Degrees]C, 5[Degrees]C and 20[Degrees]C, which represent slow, fast and medium crystallization rates respectively, as determined by the short term crystallization study.

Figure 3 shows the results of the long term crystallization study. Different maximum levels of crystallinity (using [delta H.sub.m] as the indicator) were reached in each film during the 24 weeks of crystallization, reaching constant levels after approximately 10 weeks. The maximum heat of fusion obtained in the films was directly proportional to the crystallization temperature; larger heats of fusion occurred in films crystallized at higher temperatures.

Annealing effects are expected over the 24 weeks because the films were maintained at temperatures above the [T.sub.g] and below the [T.sub.m]. Annealing effects observable from DSC experiments are manifested as changes in the shape of the melting endotherm, in location of the peak melting temperature, and in the area under the endotherm peak. Monitoring the change in the peak melting temperature over time gave an indication of when annealing effects became significant. No increase in the melting temperature was observed for PHO crystallized at any temperature evaluated during the initial 250 hours. Annealing effects were considered significant only after 250 hours and not during the 10 minute DSC scans nor during the 24 hour crystallization study.

The thermograms in figure 3 illustrate the changes typically observed in the shape of the melting endotherm peaks due to a combination of crystallization and annealing effects. In general, a broad melting endotherm with a low temperature shoulder was at first observed. As crystallization/annealing continued, this shoulder became more pronounced, eventually coalescing into the main peak. The endotherm shown after 24 weeks illustrates the final state of the films prior to mechanical testing.

As expected, the time required for these changes to occur depended on the crystallization temperature, with changes occurring faster at higher temperatures. Table 3 lists the times taken to maximize the extent and stabilize the melting endotherm shape at the different temperatures.

Equilibrium modulus and [M.sub.c] calculations

From stress relaxation experiments and small strains (<50%) and the statistical theory of rubber elasticity, the equilibrium shear modulus was determined graphically from: [delta] z G([lambda]. - 1/[lambda.sup.2]). Where: G = equilibrium shear modulus (MPa); [lambda], = extension ratio, L/Lo; [delta] = engineering stress [MPa]; as shown in figure 4. Agreement between experiment and theory was good at strains <30% and the slope in this region was used as the equilibrium shear modulus.

The magnitude of the equilibrium shear modulus in the rubber theory is based on an ideal network of point crosslinks. In this physically crosslinked system, the crystalline regions would also act as filler particles due to their finite size which would increase the modulus substantially. This filler effect can be estimated using the GuthSmallwood equation (ref. 11): [E.sub.f]/[E.sub.0] = 1 + 2.5 [V.sub.f]+ 14.1 [V.sub.f].sup.2]2 where: E = modulus; [V.sub.f] = volume fraction of filler; 0 subscript refers to untilled material and f subscript refers to filled material.

The equilibrium shear modulus can then be modified to eliminate the filler effect of the crystalline regions. Assuming PHO is 30% crystalline by volume ([V.sub.f] = 0.30) results in [E.sub.f]/[E.sub.O] = 3 and the equilibrium shear modulus reduces to 0.67 MPa. An estimate of the molecular weight between crosslinks ([M.sub.c]) can then be calculated using the equation

[M.sub.c] = pRT/G = approximately 3,600g/mole where: p = polymer density (l.019g/[cm.sup.3]) (ref. 6); T = 298K The number of physical crosslinks per chain is then estimated to be: [M.sub.n]/[M.sub.c] = 84,000/3,600 on the order of 10 to 20 crosslinks/chain.

These network parameters compare favorably with another TPE, the segmented polyurethane, whose soft segments typically have a [M.sub.n] of 600-6,000 which is analogous to the [M.sub.c] and approximately 10 to 60 hard segments per chain which is analogous to the number of physical crosslinks per chain.

Mechanical properties

The stress-strain curve and calculated parameters for PHO crystallized from the melt at room temperature for approximately three weeks are depicted in figure 5.

The results of the effect of crystallization temperature on the mechanical properties are also shown in figure 5 where the Young's modulus, tensile strength at break, and ultimate elongation are plotted as a function of crystallization temperature. The range of values obtained was large, with Young's modulus showing a 200% increase, tensile strength at break a 60% increase and ultimate elongation a 50% decrease over the crystallization temperature range evaluated.

The trends in the modulus and ultimate elongation can be explained in terms of the differences in the maximum extent of crystallinity. Higher moduli are expected as the amount of crystallinity increases due to the physical crosslinking and filler effect crystalline regions have on the material. The lower ultimate elongation may be a result of the lower extensibility of crystalline regions and thus the whole material will become less extensible as the extent of crystallinity increases.

The values for Young's modulus and ultimate elongation are significantly different than the values published by an earlier study (17 MPa and 250-350%) (ref. 6). Factors such as sample geometry and strain rate would greatly affect the measured properties. Whereas dumbbell or ring samples were used for ultimate property evaluations in this study, strip samples were used in the earlier study. A 1 or 2 min-[1] strain rate was used in this study, however no strain rate was reported by the earlier study. Figure 8 indicates the sensitivity of the tensile modulus of PHO to strain rate. No strain rate was specified in the earlier study, so direct comparison of the earlier value of moduli with the values determined in this study is inappropriate.

Figure 6 shows the results of the comparative mechanical testing of the commercial TPEs with short term, room temperature crystallized PHO. As can be seen, the values for parameters evaluated fell within the range defined by the examples of commercial TPEs tested.

For PHO crystallized at various temperatures for a long time, the evaluated parameters also fell within the range defined by commercial TPEs. The classification of PHO as a thermoplastic elastomer appears accurate based on this comparative data.

Tensile set

Tensile set quantifies the deviation of a material from ideal elastic behavior. A high tensile set indicates poor elasticity and is an important consideration for any material considered to be an elastomer. As indicated in figure 6, PHO exhibited a high tensile set, 35% after 100% elongation fairing only better than the copolyester TPE evaluated. Figure 7 shows the full tensile set data on PHO crystallized for approximately three weeks at room temperature. Similar results were obtained on the long term crystallized PHO at all temperatures. A substantial increase in tensile set was noted as the elongation was increased.

DSC analysis was conducted to determine why PHO exhibited such a high tensile set. Thermal analysis would reveal changes that had occurred in the crystalline regions which are the important tie down points in the physically crosslinked network.

Tensile set can result from many sources such as:

* irreversible orientation or permanent displacement (flow) of the physical crosslinks in the amorphous matrix;

* strain induced crystallization which does not melt upon release of the deforming stress;

* stress induced break-up or rearrangement of the physical crosslinks.

In all cases, a permanent change to the material occurs because the undeformed reference state has been irreversibly changed. Tensile set will be one consequence of the permanent changes.

The results are shown in figure 8. After small deformations (< 50% strain) no change occurred in the parameters [T.sub.g], [T.sub.m], or [delta H.sub.m], and no change in the shape of the melting endotherm (not shown) was noted in the stretched material. These observations indicate that no change to the size, distribution or amount of crystallinity occurred upon small deformation. Some change did occur in the undeformed reference state since a small amount of tensile set was observed at these small strains as shown in figure 8. These observations support irreversible orientation or permanent displacement (flow) of the physical crosslinks in the amorphous matrix was a source of tensile set.

The thermal analysis results of the samples after undergoing large deformations (100% to 300% strain) show the heat of fusion, [delta H.sub.m], increased by 60% after 300% elongation. This result supports strain induced crystallization, which does not melt upon release of the stress, as a source of tensile set.

The melting point of PHO decreased as the elongation increased. In addition, a change in the endotherm peak shape was observed as the elongation increased; the melting endotherm peak became more narrow. These observations imply that overall, the size of the crystalline regions is decreasing (lower [T.sub.m]) while the homogeneity and perfection is increasing (peak narrowing). These observations support the third mechanism mentioned as a source of tensile set, stress induced break-up or rearrangement of the crystalline regions.

Surprising results were obtained from the thermal investigation of the stretched PHO after long term crystallization at 20[Degrees]C. Table 4 lists the DSC results obtained. No large increase in the heat of fusion or decrease in the melting temperature occurred in the long term crystallized PHO as had for the short term crystallized PHO, but unusual changes were observed in the shape of the melting endotherm. A shoulder formed on the melting peak after 100% and 150% elongation and after 200% elongation a new distinct melting peak emerged at a lower melting point, 45[Degrees]C. This new peak may indicate that a different crystal structure was formed with stretching. Investigations are continuing to explain the changes in the endotherm peak shape as a function of elongation on long term crystallized PHO. Overall, many irreversible changes to the crystalline regions occurs upon deformation resulting in permanent changes to the physical crosslinks which results in significant tensile set.


PHO, at Shore A 60, is a relatively soft elastomer compared to the other TPEs.


The statistical rubber elasticity theory with a correction for the filler effect of the crystalline physical crosslinks indicates PHO forms a physically crosslinked network when crystallized from the melt at room temperature with a molecular weight between crosslinks of approximately 4,000 and with approximately 10 to 20 physical crosslink points per chain. This network structure is similar to the segmented polyurethanes.

The short term PHO crystallization study which used the heat of fusion as an indicator of crystallinity, in&cated crystallization occurred fastest at 0 to 5[Degrees]C. The equilibrium melting point of PHO was determined to be approximately 68[Degrees]. Long term crystallization which used the attainment of stable values for the heat of fusion and an unchanging melting endotherm peak shape as indicators showed that PHO crystallized very slowly requiring approximately seven weeks at 20[Degrees]C and 16 weeks at 5[Degrees]C.

PHO can be considered a thermoplastic elastomer unique in its source, bacteria. Because of this natural source, PHO has a unique chemical structure for a TPE, a stereoregular, random copolymer. The polymer is semi-crystalline (30%) and because of a low [T.sub.g] at -35[Degrees]C and a [T.sub.m] at 61[Degrees]C, the polymer exhibits elastomeric behavior within this temperature range. The commercial TPEs involved in this study were structurally different being either block copolymers (amides, copolyesters, olefinics, styrenics, urethanes) or blended dissimilar polymers (elastomeric alloys). The stress-strain properties, hardness and tensile set of PHO were found to be in the range of these structurally different, commercial TPEs.

Because PHO is a semicrystalline elastomer, the thermal history greatly affected the mechanical properties of the material with Young's modulus varying from 2.5 MPa to 9 MPa, tensile strength at break from 6 MPa to 10 MPa, and ultimate elongation from 450% to 300% depending on the crystallization temperature and time allowed for crystallization/annealing.

As an elastomer, PHO has a shortcoming - a substantially high tensile set - approximately 35% after 100% elongation. However, the commercial copolyester evaluated exhibited an even higher tensile set, approximately 55%, but is still classified as a thermoplastic elastomer. Tensile set in PHO appears to be due to the deformation induced changes in the crystalline regions which are the physical crosslinks in this elastomer. Elongation of PHO films which had been crystallized for extended times showed unusual changes in the melting endotherm peak shape with a new distinct, lower melting point peak emerging at 45[Degrees]C after 200% elongation. Further studies are being conducted to explain this unusual finding.


1. A.J. Anderson and E.A. Dawes, "Occurrence, metabolism, metabolic role and industrial uses of bacterial polvhydroxyalkanoates," Microbiological Reviews, 54,(4), 450 (1990).

2. P.A. Holmes, "Applications of PHB - a microbially produced biodegradable thermoplastic," Phys. Technol., 16, 32 (1985).

3. R.W. Lenz, Y.B. Kim, R.C. Fuller, "Polyesters produced by microorganisms, "J Bioactive and Compatible Polymers, 6, 382 (1991).

4. Y. Doi, "Microbial polvesters," VCH Publishers Inc., NY, 1990.

5. H. Brandl, R.A. Gross, R.W. Lenz, R.C. Fuller, "Pseudomonas oleovorans as a source ofpoly([beta]-hydroxyalkanoates) for potential applications as biodegradable polyesters," Appl. Environ. Microbiol., 54, (8), 1977 (1988).

6. R.H. Marchessault, C.J. Monasterios, F.G. Morin, P.R. Sundararajan, "Chiral poly([beta]-hydroxyalkanoates): An adaptable helix influenced by the alkane side chain," Int. J. Biol. Macromol., 12, April, 158 (1990).

7. N.R. Legge, G. Holden and H.E. Schroeder (Eds), "Thermoplastic elastomers: A comprehensive review," Hanser Publishers, NY, 1987.

8. K.D. Gagnon, D.B. Bain, R.W. Lenz, R.C. Fuller, "Yield study of the poly(beta-hydroxyalkanoate) produced by Pseudomonas oleovorans grown on sodium octanoate," in "Novel biodegradable microbial polymers," E.A. Dawes (ed.), Kluwer Academic Publishers, The Netherlands, 449, 1990.

9. Note: A more detailed description of this study can be found in: K.D. Gagnon, R.W. Lenz, R.C. Fuller, R.J. Farris, "Crystallization behavior and its influence on the mechanical properties of a thermoplastic elastomer produced by the bacterium, Pseudomonas oleovorans," accepted for publication in Rubber Chem. Tech.

10. Note: A more detailed description of this study can be found in: K.D. Gagnon, R.W. Lenz, R.C. Fuller, R.J. Farris, "The mechanical properties of a thermoplastic elastomer produced by Pseudomonas oleovorans," Macrornol., 25, 3723 (1992).

11. J.J. Aklonis and V.J. MacKnight, "Introduction to polymer viscoelasticity," John Wiley and Sons, NY, 1983.
COPYRIGHT 1992 Lippincott & Peto, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Farris, R.J.
Publication:Rubber World
Date:Nov 1, 1992
Previous Article:Understanding strain sensitivity effects in vibration isolators.
Next Article:Designing for plunger-type transfer molds.

Related Articles
TPE production and use in the 1990s.
Injection molding thermoplastic elastomers.
Impact modifiers: product lines reviewed.
Motor vehicle recovery spurs growth of specialty elastomers and TPEs.
Styrenic thermoplastic elastomers.
Synthesis and morphology of TPEs.
European compounders moving toward TPEs.
'Polyolefin elastomers' bow in.
Reclaimed tire rubber in TPE compounds.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters