1. Introduction
Polymers are widely used in many different fields due to their versatility and reasonable price-performance ratio. However, some specific applications, such as their use as biomaterials [
1], protective coatings [
2], and in thin film technology to name a few, require polymers with enhanced properties. Different approaches have been developed to address that need. In this work, we use two of them: (i) fabrication of nanocomposites by in-situ polymerization [
3,
4], where a small amount of a nanomaterial is added to a polymeric matrix giving the resulting material some of the remarkable properties of the nanomaterial; and (ii) surface nanostructuring, where nanostructures are generated on the surface of a material, which can be engineered to change wettability, adhesion, surface energy, tribological properties and other physicochemical properties [
5,
6].
Regarding polymer nanocomposites, one can find applications in electronics, packaging, biotechnology and many others [
1,
2,
7]. They are lighter than conventional composites, and their properties can be tailored by changing the additive and the polymer used, the percentage of additive, and the diffusion of the additive in the polymeric matrix [
3].
In this work, Poly(trimethylene terephthalate) (PTT) was used both as a stand-alone material and as a matrix for our nanocomposite. PTT is an interesting material to use as a polymer matrix because of its crystallization rate, and the variety of ways it can be processed (injection molding, film casting and film spinning) [
8]. Moreover, it can be fabricated out of a high percentage of renewable materials [
9]. Tungsten disulfide (WS
2) inorganic nanotubes were used as the additive of the nanocomposite. They are an alternative additive to carbon nanotubes to manufacture nanocomposites. Among their advantages is the low manufacturing expense, since catalysts are not needed for their fabrication and the precursors are relatively inexpensive [
10]. Besides, they present high impact resistance and good tribological behavior. It is because of these properties that they have been applied successfully as reinforcing agents and to improve the tribological properties of a variety of materials [
9,
11].
As for the generation of surface nanostructures, it is a common practice employed to further improve the surface properties of any material and provide it with some degree of functionality. Although lithographic techniques [
12,
13] are the usual method to achieve this, there has been a growing interest in laser processing techniques [
14,
15,
16,
17], specifically, in Laser Induced Periodic Surface Structures (LIPSS) generation [
18,
19,
20], due to its versatility, cleanliness and simplicity from the point of view of equipment requirements and operation.
LIPSS are ripples that appear on the surface of materials after being irradiated with a laser under specific processing conditions, whose period is in the order of magnitude of the laser wavelength. Birnbaum [
21] was the first to observe them at the bottom of an ablation crater produced with a ruby laser, and Sipe et al. later developed a first principle theory [
22,
23,
24] explaining their origin. According to this theory, the origin of LIPSS is the modulation of the intensity of the laser produced by the interference of the incident wave with a surface wave, formed by scattering on the rough surface of the material, and their period can be calculated as:
where
is the laser wavelength,
is the effective refractive index at the selvedge region [
25] and
is the angle of incidence of the laser [
26].
LIPSS formation has been reported in metals [
27], dielectrics [
28] and semiconductors [
29]. Moreover, the dependence of the LIPSS period and depth with fluence, number of pulses and laser polarization was studied in several works [
30,
31,
32].
Remarkably, LIPSS in polymers form below the ablation threshold fluence. The nanostructures are not formed as a result of material removal or ablation but following a reordering of the material on the surface. In amorphous polymers, LIPSS arise when the surface of the polymer is heated above the glass transition temperature (T
g). At this temperature, its viscosity greatly diminishes allowing the polymeric chains to flow. In contrast, in crystalline polymers, the surface must reach the melting temperature to break the crystalline structure and be able to flow. Many polymers, and in particular PTT, are not crystalline but semicrystalline, which means that only a percentage of them is in a crystalline phase [
33,
34]. The higher this percentage, the higher the threshold temperature that allows the polymer to flow, and, therefore, the higher the fluence and number of pulses necessary for LIPSS formation [
19].
In this work, we generated LIPSS with femtosecond laser pulses. fs-LIPSS are interesting as far as nonlinear absorption is the key phenomenon behind energy transfer to the material and it depends in a far less decisive way on wavelength compared to linear absorption. This provides more freedom when choosing the irradiation wavelength and thus allows to produce LIPSS with a period on demand. Moreover, since the interaction is much faster than the thermal relaxation times, all the phenomenology is dominated by the laser-matter interaction, and the thermally affected area is much smaller than with longer pulses, which enhances our control over the LIPSS area. Despite these advantages, few articles are dedicated to fs-LIPSS [
35,
36,
37,
38,
39,
40] and picosecond LIPSS [
41,
42] generation on polymers, in contrast with the abundant information on their generation with nanosecond pulses [
19,
43,
44,
45,
46,
47,
48].
The formation mechanisms of fs-LIPSS are not yet fully understood. Bonse et al. tried to explain fs-LIPSS by assuming a variation in the refraction index of the material in Sipe’s theory, following the Drude model, creating the so-called Sipe–Drude model [
30]. However, this model does not explain all the phenomena present in fs-LIPSS, such as High Spatial Frequency LIPSS (HSFL)—LIPSS that have a period smaller than laser wavelength and are aligned perpendicular to the regular LIPSS, also called Low Spatial Frequency LIPSS (LSFL). New theoretical models have been proposed to explain these phenomena, including a model based on self-organization from highly electrostatic instabilities originated by the laser [
49], one based on thin-films hydrodynamics [
50], one based on the analysis of an electronic excitation when short-lived plasma is created [
51,
52] or another model that uses Finite Differences Time Domain methods to solve Maxwell’s curl equations for linear, isotropic, dispersive materials with no magnetic losses [
53].
The objective of this work is to modify the surface properties of PTT and the composite PTT/tungsten disulfide inorganic nanotubes (PTT-WS2) by means of LIPSS formation with a femtosecond laser and measure the triggered physicochemical changes. In this way, we can shed light on whether the creation of this nanocomposite and further LIPSS generation on the surface may be used together to enhance the properties of PTT and to check if LIPSS formation is as effective in PTT-WS2 as in raw PTT.
2. Materials and Methods
2.1. Materials
The samples investigated are free-standing films, 200 ± 10 μm thick, of PTT and PTT-WS
2 nanocomposite, prepared by in situ polymerization. The nanocomposite is made out of PTT as the matrix and WS
2 nanotubes as the additive (0.5% in weight). The preparation procedure and characterization of the samples have been reported in previous work [
9]. The more relevant properties for our work given there are shown in
Table 1 and UV-Vis absorption spectrum of PTT is shown in
Figure S1.
2.2. Laser Irradiation
All the irradiations were carried out in air with the third harmonic of a femtosecond laser system. This system consists of a Ti:Sa oscillator (Tsunami, Spectra Physics®, Mountain View, CA, USA) and a regenerative amplifier (Spitfire, Spectra Physics®, Mountain View, CA, USA). The output pulses have 260 fs (FWHM), λ = 265nm, a repetition rate of 1 kHz, an energy up to 1mJ and are linearly polarized.
Regarding the irradiation set-up, the fs laser beam is focused perpendicularly on the surface of the sample, which is placed on a motorized XYZ translation stage. To control the number of pulses we used an electromechanical shutter, and to control the fluence we use neutral filters for coarse tuning as well as a λ/2 plate and a linear polarizer for fine tuning. A thermopile detector (407 A, Spectra Physics®, Mountain View, CA, USA) was used to measure the average power of the beam. We calculated the fluence from this average power measurements assuming the laser transversal mode is Gaussian TEM00, which is a reasonable approximation for this system.
For this study, we irradiated the sample varying the number of pulses at fluences per pulse ranging from 8.5 to 33.9 mJ/cm2.
2.3. AFM Measurements. Topography
Atomic Force Microscopy (AFM) technique was chosen to measure the surface topography. Due to the soft nature of our samples, we measured in tapping mode using an AFM Multimode 8 (Bruker®, Karlsruhe, Germany) system with the controller Nanoscope V (Bruker®) and the software Nanoscope Analysis 1.50 (Bruker®) for image analysis. The tips were silicon NSG30 (NT-MDT®) with a curvature radius of ~6 nm, a nominal resonant frequency of 320 kHz and a typical spring constant of 40 N/m.
2.4. Contact Angle Measurements. Surface Energy
Intending to determine the changes in wettability and surface energy of the samples, we measured their contact angle (CA) with different liquids both before and after irradiation. The contact angle is the angle formed by the liquid-solid and the air-liquid interfaces when a drop is deposited in a planar solid. It gives a measure of the wettability of the material. Moreover, if we measure the contact angle of liquids of known surface energy in our material, we can obtain the components of the surface energy using the Owens, Wendt, Rabel and Kaelble (OWRK) model, which separates the total surface energy (
) in a polar component (
, due to acid-base interactions) and a dispersive component (
, due to Lifshitz-van der Waals forces) [
55,
56], or the van Oss, Chaudhury and Good model, which further separates the polar component in a polar negative component (
, associated with electron donors) and a polar positive component (
, associated with electron acceptors) [
57]. This last method requires the use of at least three liquids of different nature (apolar and polar). We used paraffin oil as the apolar liquid and deionized water and glycerol as the polar ones (surface energy components detailed in
Table A1 [
58]).
The measurements were performed at room temperature and ambient humidity, by means of the sessile drop technique using a pocket goniometer PG2 (FIBRO system, Stockholm, Sweden). We carried out eight measurements for every different sample-liquid pair for non-irradiated samples and samples irradiated with 5000 pulses at 20.3 mJ/cm2, since preliminary AFM measurements guaranteed that those laser irradiation conditions generated well-ordered LIPSS.
2.5. PF-QNM Measurements. Adhesion and Elastic Modulus
Peak Force-Quantitative Nanomechanical Mapping (PF-QNM) [
59] is a protocol developed by Bruker
®, based on AFM, that measures the adhesion, elastic modulus, deformation and topography of a surface simultaneously, with the spatial resolution provided by AFM. This protocol was used to characterize the nanomechanical properties of the samples. Specifically, the surface elastic modulus was obtained by application of the Derjaguin–Muller–Toporov (DMT) model [
60]:
where
F is the force applied to the scanning tip,
Fadh is the adhesion force between the tip and the sample,
R is the tip radius, d is a parameter related to the tip penetration into the sample, called deformation and
E* is the reduced elastic modulus. The adhesion force,
E*, and the deformation were obtained from the Force-Distance traces of the PF-QNM nanoindentations.
E* is related to the Young’s modulus of the sample by its Poisson’s ratio which we fixed as 0.3 for both PTT and PTT-WS
2 since we assumed the nanocomposite would have the same Poisson’s ratio as PTT. For more in-depth information see
Appendix A.
To extract information, we must know the spring constant of the cantilever, the radius of the tip and the deflection sensitivity of the cantilever. We calculate the spring constant using Sader’s method [
61], and the radius of the tip and the deflection sensitivity of the cantilever are obtained using Bruker
® software and measuring standardized samples.
We used the aforementioned AFM device, but changed the tip to a RTESPA-300 (Bruker®) with a curvature radius of 13 nm, a nominal resonant frequency of 300 kHz, and a typical spring constant of 40 N/m. We carried out these measurements for samples at relative humidity (RH) and temperature values of 28 ± 1% and 21 ± 0.5 °C, respectively, as monitored by a temperature and RH data logger (EL-USB-2-LCD, Lascar Electronics, Whiteparish, UK).
4. Conclusions
We have induced LIPSS with UV femtosecond laser pulses in both PTT and PTT-WS2 surfaces. In all cases, the nanostructures are parallel to the polarization of the incident laser. The period of the structures is around 260 nm, close to the laser wavelength.
LIPSS emerge for fluences below the ablation threshold of the materials, from 15.9 to 31.3 mJ/cm2 for PTT, and from 19.1 to 33.9 mJ/cm2 for PTT-WS2, conditioned by the number of pulses (500–10,000). From this data, we can conclude that the presence of the nanoadditive leads to an increase of the energy density needed to trigger LIPSS formation. We explain this as the effect of the higher crystallization percentage and thermal dissipation of PTT-WS2.
The behavior of LIPSS period and depth with the fluence and number of pulses is similar to that reported for other polymers and was explained turning to the formation mechanism of LIPSS in polymers and the importance of feedback and incubation in the generation of LIPSS.
We have studied the wettability and surface energy of the samples, finding that the former increased with the formation of LIPSS and the total surface energy remained constant. However, its negative polar component increased heavily. This suggests the formation of polar hydrophilic species, caused by a reaction with the oxygen in the air, catalyzed by the ionization and high temperature of the surface of the sample while the irradiation took place.
We characterized the topography and mechanical properties of the sample, finding that the formation of LIPSS did not change the Young’s modulus remarkably, but it induces a decrease of the adhesion force in both materials by a factor of four. We attribute this effect to the change of surface chemistry, as also indicated by the contact angle measurements.
In conclusion, LIPSS emerged at slightly higher energies for the nanocomposite than for raw PTT but produced almost equal effects in both PTT and PTT-WS2. Therefore, LIPSS surface nanostructuring can be used in this nanocomposite without any demerit. Hence, we can use LIPSS to easily change the surface properties of the nanocomposite, specifically the mechanical ones. Moreover, given the high control over the nanostructured area, we could create small zones with different nanostructures and thus, different surface properties.