Moulding Wetting by Laser-Induced Nanostructures

In this account, we have exposed multilayer thin metal film samples (each layer of 17 nm thickness) of different materials to femtosecond laser beam at 1030 nm wavelength. The interaction generated high-quality laser-induced periodic surface structures (LIPSS) of spatial period between 740 and 790 nm and with maximal average corrugation height below 100 nm. The contact angle (CA) values of the water droplets on the surface were estimated and the values between unmodified and modified samples were compared. Even though the laser interaction changed both the surface morphology and the chemical composition, the wetting properties were predominantly influenced by the small change in morphology causing the increase of the contact angle for 20% which could not be explained classically. The influence of both surface corrugation and chemical composition to the wetting properties has been thoroughly investigated, discussed and explained. The presented results clearly confirm that femtosecond patterning can be used to mould wetting properties.

The causes of LIPSS generation and shaping are seen most probably in the emergence of the surface plasmon polariton (SPP), as well as in the hydrodynamic features [8][9][10]. If compared to single layer metal films, more regular LIPSS are generated with low-fluence femtosecond (fs) laser beam interaction with multilayer thin metal films, since the existence of the metal sublayer influences the quality and stability of LIPSS [11].
Changing wetting and tribological properties of the material by LIPSS formation opens new fields of application in nano/microfluidics, optofluidics, fluid microreactors, biomedicine, biochemical sensors, and thermal management [12,13]. Control of wetting properties and achieving super-hydrophobic surfaces by laser interaction has been reported for various materials: stainless steel, TiAl alloy, Si and materials coated with hydrophobic materials like chloroalkylsilane and fluoroalkylsilane [14][15][16]. In all mentioned cases super-hydrophobicity is achieved by chemical modification of the surface or microstructuring.
Laser interaction can induce the change in chemical composition of the surface, forming ultra thin oxide layers, which contributes to the wetting. Wetting is initially enhanced, but as time passes, it is ultimately reversed by surface chemistry phenomena that take place on the irradiated surfaces, i.e. hydrophobicity is increased [16,17].
Van der Waals and electrostatic forces play an important role in adsorption, adhesion and wetting phenomena [18]. The Casimir force shows application potential in the field of micro-and nanoelectromechanical systems -engineered devices, where controlling forces between microscopic bodies or surfaces are crucial for a variety of applications [19,20]. The influence of submicron surface corrugations on Lifshitz-van der Waals forces have been calculated for polyethylene, where surface nano-patterning is responsible for changing the forces from attractive to repulsive [21,22]. For metal surface nano-structuring at scales below the plasma wavelength, the Casimir interaction decreases faster than usual for large inter-surfaces separation while at short separations an equivalent pressure is larger [23]. The motivation of our study is to reveal the link between wetting and nanocorrugation [21,22] experimentally and to extend the potential application of nanostructures.
The morphology of surface micro/nanostructures obtained by fs laser is evaluated in aim to determine the influence of morphology to their wetting. By interaction with fs laser beam, we generated LIPSS on the surfaces of several metallic multilayer materials. We estimated the contact angle (CA) difference for different materials and also for materials before and after LIPSS forming thus examining the change in hydrophobicity.

Experimental part
The samples were multilayer thin films on Si substrate: 15x(Zr/Ti), 15x(Ti/Zr), 8x(Zr/Cr/Ti), with the topmost layer being Zr, Ti and Zr, respectively. The total thickness of the complete multilayer structures was 500 nm, where thicknesses of individual layers were about 17 nm. The samples were exposed to femtosecond beam which generated the LIPSS.
The laser source was Pharos SP Yb:KGW laser system from Light Conversion. The surface of thin films were irradiated by focused linearly p-polarised pulses of 1 kHz repetition rate, 160 fs pulse duration, 1030 nm central wavelength and 43 μm Gaussian spot diameter. The samples were laser processed in open air ambient environment and mounted on a motorised, computer controlled, X-Y-Z translation stage, at normal incidence to the laser beam. For higher precision, the irradiations were conducted at identical conditions covering surface of 5x5 mm at pulse energy of 2.5 μJ (fluence of 0.662 J/cm 2 ) and scan velocity of 3 mm/s with constant distance between lines of 15 μm. In each line, energy per pulse was assumed to be constant, since the pulse energy deviation was less than 1%. Equivalent exposure time for the area of the spot was 14.3 ms.
In order to determine the effects of chemical composition and morphology to wetting, the drops of distilled water have been placed onto the both unmodified and laser-processed surface areas in open air. The volume of each droplet was of 3 μL, controlled by a motorised syringe, while the shape has been determined by micro-photographing in the horizontal direction by using Data Physics OCA (optical contact angle) Series device. The values of contact angles have been estimated from the images by the utilization of the Gwyddion software.

Results and discussion
Wetting properties of surfaces are influenced both by their chemical composition and by their morphology i.e. corrugation [24,25]. The influence of the different chemical composition is represented by different materials of the topmost layers, while the influence of the morphology is caused by the interaction with laser. The role of more complex multilayer structure is represented by using 8x(Zr/Cr/Ti) samples. Exposing the samples to fs laser beam generated LIPSS on the surfaces, Fig. 1. LIPSS on the surface of 15x(Ti/Zr) on Si are shown for two magnifications in Figs 1 a and b; LIPSS on the 8x(Zr/Cr/Ti) are presented in Figs 1 c and d, also for two different magnifications. Besides wavy-patterned LIPSS, the occurrence of nanoparticles with dimensions less than the periods of LIPSS, is also noticeable. Since the ablation threshold for the components is lower than the applied fluence of 0.662 J/cm 2 significant ablation of the multilayer samples was expected [26]. The 2DFFT processing of LIPSS micrographs is presented in Figs 2 a (LIPSS from Fig. 1 a) and b (LIPSS from Fig. 1 c). The appearance of the maxima shows the periodicity of the structures. Clear distinction of the maxima indicates that LIPSS are highly regular, due to the multi-layer structure. By convenient combination of the materials, the existence of the underneath layer produces steep change of the layers' temperature and enables the transfer of thermal energy through the interface and away from the interaction zone [11].
The interaction of fs beam with multilayer thin film materials induces intermixing of initially separated layers, which could be attributed to the alloying between the components [27]. Due to the intense laser ablation of the material and laser-induced intermixing of layers, all metallic components are present in similar amounts as oxides on the top surface.
The images of water droplets, placed on the surfaces both unexposed and exposed to fs beams are presented in Fig. 3. For 15x(Ti/Zr), the droplets are shown in Figs 3 a (on unexposed surface) and b (on exposed surface). Different shapes indicate the influence of the laser-induced change of the surface, which is twofold: both in the morphology and in the chemical composition. Similar holds for the droplets on the 15x(Zr/Ti) unexposed (Fig. 3 c) and exposed (Fig. 3 d) surfaces, as well as on the 8x(Zr/Cr/Ti) unexposed (Fig. 3 e) and exposed (Fig. 3 f) surfaces. The formation of LIPSS is accompanied with intensive modification of multilayer thin film, without noticeable and distinct pattern formation, and LIPSS are somewhere covered with nanoparticles with the dimension of up to 100 nm. The values of LIPSS periods estimated from the micrographs, as well as measured values of contact angles of both surfaces are summarised in Table 1. The period matches between 72 and 77% of the value of the implemented wavelength. The uniformly distributed LIPSS are oriented normal to polarisation direction and their period is not close to used laser wavelength for all observed samples. These types of LIPSS originate from the interference of the incident laser beam with surface electromagnetic wave excited during the laser treatment [8]. The CA values indicate that non-corrugated surfaces are hydrophilic. Similar CA values for surfaces unmodified by laser suggest they are not much influenced by the difference in the composition. The difference in CA between untreated samples of Zr/Ti and Zr/Cr/Ti could be explained by long-range wetting transparency [28]. For surfaces modified by laser beam, the values also do not differ much. However, a strong increase after the interaction by the fs laser beam suggests it is dominantly caused by the structural change -the occurrence of LIPSS -and not by the composition change. Differently to the case where 40° rise in hydrophobicity of Ti surface was obtained by µm corrugations [17], better results in this work were obtained by nm corrugations. Corrugation induces significant changes in the CA values meaning that corrugated surfaces are almost superhydrophobic.
In this view, all the samples should have similar increase of the CA after the interaction. However, the samples with Zr as the topmost layer in as-deposited (unmodified) samples reached greater CA increase after the interaction than the samples with Ti on top (Table 1), which would mean that slightly different amounts of Zr and Ti components are removed by laser treatment depending on the initial order of the layers.
LIPSS filling factor -the ratio between the ripple width and the spatial period -was estimated from Fig. 1 by the Gwyddion programme. For 15x(Ti/Zr) it is ~0.6, while for 8x(Zr/Cr/Ti) it is ~0.72. Greater filling factor leads to greater Casimir pressure for the same distance between the surfaces, while the corrugation height does not play a significant role [23]. Topmost surface is covered with very thin oxide layer, because the ambient is open air. Laser interaction will enforce the oxidation, but only to the saturation level [29]. In this way, greater filling factor for the samples with Zr as the topmost layer causes significantly greater relative change of the CA values than for the ones with Ti on top. In addition, this causes greater hydrophobicity of the surface.
Corrugation is responsible for hydrophobicity (wetting). Classical theory demands the size of corrugation to be above micrometre level in order to achieve super-hydrophobicity. If we assume that multilayer structure can be imagined as optical cavity, quantum effects cause rise of super-hydrophobicity. 21,22 Corrugations of nanoscopic level dramatically affect the Van der Waals interaction energy (through quantum vacuum photon modes) and thus the wetting contact angle is modified. Up to now, to achieve super-hydrophobicity, different coating and dozens of micro-size structures were described elsewhere [30,31]. The increase of hydrophobicity caused by fs nano-patterning experimentally confirms the model developed by Dellieu et al. [21,22].
Furthermore, the presented study goes beyond the proposed model since super-hydrophobicity is achieved on the metallic surface (not only molecular solid). In the end, the possibility of attaining superhydrophobicity using submicron corrugation by controlling quantum vacuum modes opens large almost unlimited fundamental and technological applications. While the responsibility of the change in the dispersion component is doubtful, the corrugation is responsible for all the changes of the CA. However, the magnitude of the effects is too big for such small corrugations, which can be explained with above mentioned non-classical approach [21,22]. Untreated metal surface is generally hydrophilic, but in this experiment it changed to highly hydrophobic by the implementation of sub-micrometre-sized corrugation.

Conclusions
We have investigated wetting of laser modified multilayer thin films of different metal materials by using contact angle measurements. The values do not differ much for different materials (both as-deposited and laser modified) but significantly change after the laser interaction and the formation of LIPSS. Here, due to nanometre-sized structuring, wetting can't be explained without taking quantum effects into consideration. To our actual knowledge, this is the first example of significant wetting change by nanostructures and it is in good agreement with previous publication [21,22] which explains wetting by quantum vacuum modes. This suggests that the increase of the hydrophobicity is related to the morphological changes of the cavity. The presented results extend the application of the model previously developed [21,22] for molecular solids and show that nanocorrugation of metal surfaces can be used to significantly affect the wetting. Potential applications of presented research could be numerous.