Selective laser crystallization and amorphization in polymer fibers

Polyester fiber manufacture represents over half of all fiber production worldwide, and this may be attributed to its high strength, low cost, and ease of modification, with uses spanning engineering, apparel, and medical fields.¹ However, polyethylene terephthalate (PET) has poor chemical adhesion due to low wettability caused by low surface free energy and chemical inertness.² Modification of fiber properties is often necessary to increase chemical adhesion and to alter fiber properties to meet final requirements.

Figure 1 shows a process diagram displaying the main stages in polyester textile production, from raw materials to a finished product. Fiber properties can be adjusted at different stages, as shown in Fig. 1, with techniques such as chemical modification of the polymer melt and selection of fiber cross-sectional shape being used prior to fiber extrusion, unlike surface modification techniques, which are performed after drawing. Later-stage modification allows decisions surrounding the final properties matching with the end use of the fiber to be delayed, thereby increasing agility within the supply chain and providing a commercial advantage.³ This makes surface modification a popular choice.

Surface modification techniques such as plasma, UV, and chemical and physical modification, which are employed after fiber drawing, are used to increase fiber adhesion, allowing finishes to be absorbed and properties to be altered, with surface treatments reported to decrease the water contact angle to the minimum of 0°, resulting in maximum wettability.⁴ Treatment of PET with aqueous sodium hydroxide is a popular chemical treatment, increasing adhesion and reducing static; however, fiber-breaking strength and fiber dimensions are reduced, impeding its use in technical applications.⁵ Plasma and UV processing are also common to alter the functional groups on the fiber surface to increase reactivity. UV radiation can also be applied via lamps or lasers to alter fiber topology, creating rippled structures providing a greater surface area to the fiber and changing wettability.^6,7 Despite the successful application of these techniques in industry, these are destructive processes that either remove materials or cause chemical changes in the fiber surface.

At present, infrared laser use in the textile industry is primarily for denim fading, replacing the more environmentally harmful stone-washing process, and these lasers are applied at the garment finishing stage, as shown in Fig. 1. Lasers are used to destructively alter the appearance of fibers within the fabric, resulting in a distressed effect caused by irreversible damage.⁸ This is very different from fiber surface modification, which takes place much earlier in the manufacturing process. Here, we consider laser surface treatment for nondestructive, reversible fiber modification at the fiber manufacturing stage.

To develop a laser process for controlling fiber properties without material degradation, we need to identify how the fiber microstructure can be controlled. The important factors are the crystalline structure of PET fibers and thermally induced physical transitions.

PET fibers are semicrystalline structures that can be described using a simplified two-phase model, consisting of crystalline and amorphous regions. Crystalline regions are formed when polymer chain segments run parallel to each other, resulting in a tightly packed, regular structure, allowing greater interchain forces imparting strength and rigidity to the fiber. Amorphous regions are found between crystallites and are areas of higher disorder, with average distances between adjacent polymer chains much greater than in the crystalline regions. Weaker interchain interactions arise from the more loosely packed arrangement, giving fibers flexibility and extensibility due to a slippage of the molecules against one another. Greater distances between chains also allow a penetration of other molecules into the fiber, unlike crystallites, which are impenetrable. Tie molecules run between adjacent crystallites through amorphous regions, and on loading, give strength to the fiber because of the presence of strong covalent bonds in the main chain bearing the load.

The spinning and subsequent drawing processes during fiber production are used to increase the degree of crystallinity and orientation of the polymer chains in the direction parallel to the fiber axis by extending polymer chains, increasing load bearing on the strong covalent bonds in the backbone of the polymer chain, and increasing fiber strength. In PET, polymer chains can exist in the extended trans or the twisted gauche conformation, with crystalline regions containing only the trans conformation due to larger bond angles promoting the close packing required for crystallization, whereas amorphous regions contain both conformations. The drawing process increases the number of molecules in the trans conformation; however, stress formation in the fiber caused by a stretching of tie molecules can cause crystallite imperfections.⁹ Annealing of fibers after manufacture allows a relaxation of tie molecules, reducing dimensional changes further downstream.

Annealing of PET fibers above the glass transition temperature induces thermodynamic relaxation, causing morphological changes to the fiber that differ depending on whether the fiber is heated under tension. Annealing in the relaxed state is reported to reduce Young’s modulus and yield strength because of a reduction of the extended tie molecules between the amorphous and the crystalline regions,^10–12 which in PET is due to a contraction of the polymer chains from the trans to gauche conformation.

Fiber Young’s modulus, E, is related to the two-phase structure by the model proposed by Takayanagi et al.,¹³ 

where E_c is the crystalline Young’s modulus, E_a is the amorphous Young’s modulus, and λ and φ are the parameters relating to the degree of stress and strain distribution through the fiber.

Equal stress distribution between the crystalline and the amorphous regions within the fiber is brought about by a series coupling of the two regions, which occurs under fibers annealed in the relaxed state. Series coupling results in a simplification of Takayanagi’s model, given by

where X_f is the volume fraction crystallinity of the fiber.

This model provides a useful link between the macro properties of the fiber, in the form of Young’s modulus, and the microproperties in the form of crystallinity of the fibers. If surface modification techniques vary the degree of crystallinity in accordance with this model, then it could provide a useful tool to predict fiber microstructure from mechanical properties associated with surface changes, something that is not currently possible.

This paper aims to understand the mechanism of structural change in PET after laser treatment at various fluences, using Young’s modulus as a way of measuring the macroproperties and relating these to the microstructural changes. Takayanagi’s model is used for the validation of crystallinity changes affecting fiber Young’s modulus.

A PET monofilament yarn with a diameter of 0.103 mm was used for all experiments.

A Synrad ti100 100 W infrared CO₂ laser operating at a 10.6 μm wavelength with a TEM00 mode beam was used for the treatment of the PET fiber at fluence values between 0.0088 and 0.086 J/mm². The fibers were secured in a frame at the focal distance of the laser in a relaxed state. The laser beam was operated with the beam in focus on the yarn with a beam size of 0.3 mm. The laser was scanned across the fiber at a velocity of 7000 mm/s. Laser power was measured using a Coherent LM-200 Laser Power Detector to ensure correct fluence values.

An Instron 3343 Single Column Testing System was used with a 50 N load cell to obtain tensile data for PET fibers. bluehill 3 software was used for programming the machine. Slack correction was applied at a value of 0.1 N to eliminate the effect of operator tension/slack applied during sample loading. The gauge length of the yarn was 60 mm and was extended at a constant rate of 60 mm/min, with measurements of force and extension taken every 2 ms. Stress and strain values were calculated from the data. Young’s modulus values were calculated using the least squares fit on all points between 0 and 0.2% strain on the stress-strain curve. An average Young’s modulus value was calculated from ten repeats and plotted against laser fluence.

Fiber images were taken using a Nikon Optiphot microscope. Fiber cross-sectional images were obtained by cutting the fibers with a razor blade.

The melting behavior of the PET fibers was analyzed using differential scanning calorimetry (DSC). A TA Instruments Q200 Differential Scanning Calorimeter was used for heating at a rate of 10 °C/min up to 290 °C. The enthalpy of fusion, ΔH_f, is the energy required to overcome the intermolecular forces binding crystallites together and was obtained from the area under the melting endotherm on the DSC plot using ta advantage software. The mass fraction crystallinity, X_m, was determined from the ratio of enthalpy of fusion of the yarn sample to the enthalpy of fusion of a 100% crystalline sample of PET with ΔH_f° = 140.0 J/g,¹⁴ given by Eq. (3),

Mass degree crystallinity values were converted to volume degree crystallinities for comparison with Takayanagi’s model using Eqs. (4)–(6), where V_c and V_a are volumes, ρ_c and ρ_a are the densities of the crystalline and amorphous regions, respectively, and m is the mass of the fiber sample used in DSC. Literature values for crystalline and amorphous densities were taken as 1.455 and 1.33 g/cm³,¹⁵ 

Young’s modulus data were calculated from stress strain data at 0.2% strain, and an example of a stress-strain curve for polyester is given in Fig. 2. Modulus was calculated at a low strain value to minimize the effect of change in the cross-sectional area during loading.

Young’s modulus data of the laser-treated PET fiber samples are plotted against laser fluence with error bars at one standard deviation in Fig. 3. PET Young’s modulus remained stable with laser treatment up to 0.037 J/mm², with the error bars indicating that there may be no real difference in yarn Young’s modulus within this range. The first five data points all lay within 5% of the Young’s modulus of the untreated yarn. Above 0.06 J/mm², Young’s modulus decreased at a higher rate. Laser treatment at 0.086 J/mm², the highest fluence value tested, achieved a 38.5% decrease in Young’s modulus; however, at this fluence, the fiber showed clear signs of thermal degradation and the surface roughness increased.

Figure 4 shows microscopic images of the top surface and a cross section of untreated PET fiber for comparison with the laser-treated PET fiber with fluences of 0.050 and 0.086 J/mm². The images show that there are no signs of thermal degradation to the fiber surface after laser treatment at 0.050 J/mm² and the appearance of the fiber remains very similar to that of the untreated fiber both on the surface and on the cross section. The laser-treated fiber at a fluence value of 0.086 J/mm² displays areas where damage has occurred because of the laser beam melting the top surface.

Images of a fiber cross section in Fig. 4 show that treatment at 0.050 J/mm² [Fig. 4(b)] laser fluence has no effect on the cross-sectional shape when compared with the untreated sample [Fig. 4(a)]. However, at a fluence of 0.086 J/mm² [Fig. 4(c)], a cavity is formed on the top surface of the PET fiber due to a variation of absorptivity with transverse distance from the centerline of the fiber. The highest absorption occurs at the centerline, where the beam and fiber are perpendicular, resulting in the greatest deformation of the fiber.¹⁶ The cross-sectional image in Fig. 4(c) shows that fiber modification is limited to the top surface of the fiber, which is expected due to a penetration depth of 0.0521 mm for PET at 10.6 μm wavelength,¹⁷ approximately equal to the radius of the fiber.

The DSC plot in Fig. 5 shows the melting endotherms for an untreated PET and fibers treated with laser fluences of 0.024 and 0.050 J/mm². Melting occurs at approximately the same temperature of 249 °C for all samples tested, indicating that laser treatment has no effect on the structure of individual crystallites in the bulk of the PET fibers. However, the size of the peaks appears to change, indicating a change in fractional crystallinity. Baseline shifts for the laser-treated fibers in comparison with those for the untreated are likely due to changes in the glass transition temperature caused by fractional crystallinity changes resulting from laser processing.¹⁸ 

The area under each of the melting endotherms in Fig. 5 gave ΔH_f values of 54.63 J/g for the untreated sample and 55.90 and 48.17 J/g for laser-treated PET at 0.024 and 0.050 J/mm², respectively. These values were converted to volumes using Eqs. (4) and (5). Equation (6) was used to calculate the volume fraction crystallinity of the fibers, which are plotted against Young’s modulus in Fig. 6. Takayanagi’s series model is plotted alongside the measured Young’s modulus data and a goodness of fit tested using an R² value.

Figure 6 shows that PET fiber Young’s modulus increases with volume fraction crystallinity, which is expected due to stronger interchain forces caused by a closer packing of polymer chains in the crystalline regions in comparison with the more disordered amorphous phase. The laser-treated PET fiber Young’s modulus appears to follow the two-phase model proposed by Takayanagi. A 4.8% decrease in Young’s modulus was found with a 7.2% decrease in crystallinity, indicating that treatment at a higher laser fluence increases amorphization. Uncertainty in Young’s modulus values is denoted by the vertical error bars, with a maximum uncertainty value of 0.18 GPa. Young’s modulus can be controlled within 1.8% of the measured value within the range tested.

The R² value for the measured Young’s modulus data compared with the predicted Young’s modulus using Takayanagi’s model is 0.9492, meaning that the model explains 94.92% of the variation in the data. From this, we can conclude that the model is a good fit for the laser-treated PET and that Takayanagi’s model can be used as a predictive tool linking Young’s modulus to crystallinity.

This research has shown that laser surface treatment on PET fibers at varying fluences can induce changes in the volume fraction crystallinity of the fiber. Takayanagi’s model was used to verify the link between macroproperties and microproperties of the fiber by linking fiber Young’s modulus data with fractional crystallinity.

Laser treatment at fluences up to 0.037 J/mm² had little effect on fiber Young’s modulus, whereas fluences above 0.06 J/mm² decreased Young’s modulus much more rapidly. Microscope images verified that fluences up to 0.050 J/mm² resulted in an undamaged fiber surface, indicating that laser treatment in this range would induce no changes to roughness or hand feel. It is widely reported that heat treatment of fibers in the relaxed state reduces Young’s modulus and strength,^10,11 with increased temperatures causing a greater reduction in fiber Young’s modulus due to polymer chains in the trans conformation transitioning to the gauche.¹¹ It is, therefore, expected that increasing laser fluence increases the temperature, heat-affected zone, and penetration depth within the fiber, causing a greater number of polymer chains to convert to the gauche conformation, decreasing fractional crystallinity, and resulting in amorphization of the fiber.

Laser treatment at a fluence of 0.024 J/mm² may have caused a small increase in crystallinity, likely due to a relaxation of tie molecules allowing polymer chains to realign, removing crystallite imperfections, and increasing the degree of crystallinity.⁹ However, the error bars indicate that there may be no difference in Young’s modulus between the untreated and the treated fiber at 0.024 J/mm².

Selective amorphization of PET fibers increases chemical adhesion by increasing the volume in the fiber penetrable to additional molecules. This is advantageous across several sectors for the application of surface finishes and dyes. Applications in apparel design for increased dye absorption increasing color depth on specific areas of the fabric, limited only by laser beam diameter, allow high precision designs to be used not only in fibers but also in fabric and garments. Increased dye absorption after laser treatment on polyester fabric has been reported in a number of papers,^19–21 with fractional crystallinity reported as a potential reason for increased absorption. However, the results were inconclusive and the exact mechanism for increased dye uptake was not identified. This research has successfully identified that selective amorphization is possible with infrared laser treatment, with greater amorphization occurring with increased laser fluence, suggesting a possible mechanism for increased dye uptake reported in the work of other researchers.

Increased absorption can be applied to technical finishes for realizing improved performance in medical and engineering fields. PET fibers are particularly good for implanting within the human body due to chemical inertness and high strength. Increasing the amorphous fraction of the surface allows easier application of biocompatible finishes and could be used to increase porosity in scaffolds or to control drug delivery. In addition to benefits from increased adhesion, reduction in Young’s modulus increases fiber flexibility, useful for overcoming problems that arise from high stiffness such as knotting of sutures²² or for increasing flexibility in compression bandages. Textile composites require good interfacial contact between the fibers and the matrix because this determines the degree of reinforcement and effects of deformation behavior under external stress.²³ Increasing fiber adhesion by selective amorphization on the surface while leaving the bulk properties unchanged also increases composite strength and useability.

Laser treatment in this research was applied along the full length of the fibers, resulting in changes along the entire surface. However, laser processing involves high precision and is digitally controlled, allowing heat treatment to be applied selectively and at different fluences along the fiber. Varying the fiber architecture in this way allows selective crystallization where increased strength and rigidity is required and amorphization in areas where increased extensibility, flexibility, or chemical adhesion is necessary.

Modification on a microscopic level limited only by laser beam size allows a far greater level of control to fiber properties than is currently possible with other modification techniques.

Thermally induced changes to polymer crystallinity are not limited to PET fibers and, therefore, an extension of laser-induced Young’s modulus and crystallinity control is considered possible with other thermoplastic fibers such as polyamide. In addition, infrared lasers operate at several wavelengths, offering the possibility of optimization of the coupling coefficient with the polymer to increase or decrease absorption.

Crystallinity variation of PET fibers induced by laser treatment was successfully predicted from Young’s modulus data using the series form of Takayanagi’s model. The fiber microstructure was difficult to analyze and would require equipment unavailable to many manufacturers. A simple model linking fiber crystallinity to Young’s modulus allowed a prediction of microstructural properties from macro measurements, allowing decisions on surface enhancements to be balanced with mechanical properties required for end use.

This paper has shown that high precision crystallinity control of PET fiber can be achieved via fluence variation in CO₂ laser heat treatment. Young’s modulus was unaffected by treatment at fluences below 0.037 J/mm², with Young’s modulus decreasing rapidly over 0.06 J/mm². PET fiber Young’s modulus decreased in line with volume fraction crystallinity according to Takayanagi’s two-phase model, offering a prediction of microstructural changes induced by the laser from macro measurements, unlike any other surface modification techniques.

Laser treatment offers high precision and rapid and late-stage fiber crystallinity modification for the altering of macro and microstructural properties, which could be extended beyond fibers, to fabrics, films, and to other polymer materials. Lasers are currently underutilized within the textile industry and could be harnessed for use at more stages in the textile manufacturing process.

This work was jointly funded by Stretchline UK Ltd. and EPSRC DTP (No. EP/R513088/1). Special thanks go to Robert Pugh for his continued help.

The authors have no conflicts to disclose.

Francesca Wheeler: Writing – original draft (lead). John R. Tyrer: Supervision (equal). Lewis C. R. Jones: Writing – review & editing (equal).