ARTICLE pubs.acs.org/Langmuir
Laser-Induced Nanoscale Superhydrophobic Structures on Metal Surfaces R. Jagdheesh,† B. Pathiraj,† E. Karatay,‡ G. R. B. E. R€omer,† and A. J. Huis in‘t Veld*,† †
Chair of Applied Laser Technology, Faculty of Engineering Technology, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands ‡ Soft Matter Fluidics and Interfaces Group, Faculty of Engineering Technology, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands
bS Supporting Information ABSTRACT: The combination of a dual-scale (nano and micro) roughness with an inherent low-surface energy coating material is an essential factor for the development of superhydrophobic surfaces. Ultrashort pulse laser (USPL) machining/structuring is a promising technique for obtaining the dualscale roughness. Sheets of stainless steel (AISI 304 L SS) and Ti-6Al-4V alloys were laser-machined with ultraviolet laser pulses of 6.7 ps, with different numbers of pulses per irradiated area. The surface energy of the laser-machined samples was reduced via application of a layer of perfluorinated octyltrichlorosilane (FOTS). The influence of the number of pulses per irradiated area on the geometry of the nanostructure and the wetting properties of the laser-machined structures has been studied. The results show that with an increasing number of pulses per irradiated area, the nanoscale structures tend to become predominantly microscale. The top surface of the microscale structures is seen covered with nanoscale protrusions that are most pronounced in Ti-6Al-4V. The laser-machined Ti-6Al-4V surface attained superhydrophobicity, and the improvement in the contact angle was >27% when compared to that of a nontextured surface.
1. INTRODUCTION Hydrophobic surfaces have attracted much attention because of their potential in microfluidics, such as lab-on-chip devices, and as functional surfaces, for example, in automotive and aerospace components.1 Surface morphology is a key factor determining the wettability of a solid surface. The wettability of a solid surface is usually defined by measuring the static contact angle between a water droplet and the surface. In general, a hydrophilic surface has a static contact angle of 90°. Superhydrophobic surfaces have contact angles of >150°.2 The wettability of surfaces is described by either the Cassie Baxter model3 or the Wenzel model.4 According to the Wenzel model, the liquid is in contact with the whole surface and the hydrophobicity of the solid surface is improved by an increase in the solid liquid interface area due to the surface roughness. In the Cassie Baxter model, the surface has voids with trapped air, into which the liquid cannot penetrate. In this case, instead of a solid liquid interface alone, both solid liquid and liquid vapor interfaces are considered. If the surface roughness were to consist of patterns of square or conical protrusions (pillars), air would be trapped inside the valleys and the Cassie Baxter model could be used to describe the hydrophobicity. It is well-known that water repellency is governed by the surface morphology and surface chemistry.5,6 Patterning is one of the effective ways to change the surface morphology and to improve the wetting properties. In recent years, inspired by the topography of the r 2011 American Chemical Society
Lotus leaf, numerous methods such as plasma surface modification, radiation grafting, electrochemical deposition, electroless replacement deposition, lithographically patterned substrates, vertically aligned carbon nanotubes, natural oxidation, and laser patterning have been adopted for the development of hydrophobic surfaces.7 29 Among these surface modification techniques, laser patterning using an ultrashort pulse laser source is a unique technique that can modify the surface morphology with very limited distortion of the bulk material. Moreover, it is a noncontact method, and complex patterns can be created. The laser material interaction of ultrashort laser pulses on metals causes nanoscale ripples under certain processing conditions. If laser energy is above the ablation threshold of the material and below the melting threshold, these so-called laser-induced periodic surface structures (LIPSS) are obtained. LIPSS have been observed since the late 1960s on all kinds of materials. Such LIPSS are usually termed nanoscale “ripples” and consist of wavy surfaces with a periodicity and an amplitude equal to or smaller than the wavelength of the laser beam. The nanoscale ripples found on the laser-patterned surfaces are obtained when processing near the ablation threshold. Two kinds of ripples, viz., high-frequency and low,frequency LIPSS, are observed. In most cases, the orientations of such Received: March 24, 2011 Revised: May 10, 2011 Published: May 31, 2011 8464
dx.doi.org/10.1021/la2011088 | Langmuir 2011, 27, 8464–8469
Langmuir low-frequency LIPSS are found to be perpendicular to the polarization direction of the incident laser beam.26 Kietzig et al.27 studied the time dependence of the hydrophobicity after machining on structures created by femtosecond laser-machining of different metals and alloys. They also observed that the laser-induced dual-scale (micro and nano) roughness structures play a significant role in the wetting properties of the metal surfaces. Wu and Zhou28 showed that surfaces with dual-scale structures had higher apparent contact angles during wetting. In general, the periodicity of the ripples is equal to or less than the wavelength λ of the laser radiation used for the structuring. This effect is explained on the basis of interference of incident and scattered laser radiation or excited surface waves.29 Formation of subwavelength ripples has also been observed and was attributed to the relaxation of highly excited unstable surface layers.30 The physical phenomena explaining the ripple initiation, growth, and transition toward other patterns are still not clear. The initiation of bubble formation and the subsequent transformation to low- and high-frequency ripples with an increase in the fluence have been studied on 800H alloyed steel by R€omer et al.31 To improve the wetting properties of metal surfaces, a combination of the two strategies, viz., altering both surface morphology and chemistry, has also been tried. Baldacchini et al.32 successfully attempted to switch from the hydrophilic state to the hydrophobic state by applying a fluorosilane coating on a laser-textured silicon surface. Hydrophobic surfaces were also reported to have been achieved without any additional coating on the surface after laser treatment.33 Water repellence was also obtained by Bhattacharya et al.34 on clustered copper nanowires without using a coating. They attribute the hydrophobicity mainly to a geometrical phenomenon related to the nature of clustering. Their studies imply the dominant importance of surface patterning in the enhancement of water repellence. Apart from alteration of the surface morphology, many studies have sought to enhance the hydrophobicity by changing the surface chemistry by application of coatings and by surface treatments. Such methods include the sol gel process, plasma surface modification and polymerization, physical vapor deposition, chemical adsorption, dispersion plating, and self-assembled monolayers (SAM) based on organosilanes.35 Among these methods, formation of SAM by chemical vapor surface modification is an advantageous, simple, and effective method for obtaining hydrophobic states of matter even when the substrate material is hydrophilic.35,36 To the best of our knowledge, none of the published work on LIPSS has related nanoscale ripples to the improvement of hydrophobicity. Therefore, this study has been initiated with the objective of studying the influence of the nanoscale ripples on the improvement in the hydrophobicity of a metal surface. In this study, the effect of laser-patterned nanoscale structures on the wetting behavior of silanized stainless steel (SS) and Ti-6Al-4V sheets was investigated. The metal substrates were laser-machined with picosecond laser pulses with different laser processing parameters. The surface morphology was characterized using scanning electron microscopy (SEM) and atomic force microscopy (AFM). A perfluorinated octyltrichlorosilane (FOTS) coating was applied on laser-machined surfaces. Wetting behaviors were evaluated by static and contact angle hysteresis measurements.
2. EXPERIMENTAL SETUP Flat stainless steel (AISI 304 L SS) and Ti-6Al-4V sheets (60 mm 20 mm 1 mm) were laser-machined by ultraviolet laser pulses, with
ARTICLE
Figure 1. Schematic illustration of 50% overlaps between successive lines.
Table 1. Laser Processing Parameters laser power
repetition rate
energy per pulse
fluence
(mW)
(kHz)
(μJ)
(J/cm2)
50
200
0.25
0.098
different laser processing parameters. Before being laser-machined, samples were mechanically polished with SiC grinding paper, followed by a final polishing using colloidal silica particles resulting in a surface roughness (Ra) of ∼23 nm. The samples were cleaned with acetone prior to being laser-machined. A Trumpf TruMicro laser source, with a central wavelength of 1030 nm, was used for generation of picosecond laser pulses. However, a frequency-tripled (wavelength of 343 nm) average maximal power of 15 W at 400 kHz was used for surface patterning. Use of a shorter wavelength allows smaller diffraction-limited spot sizes, which allow finer features in the surface pattern. The experiments were performed at a fixed pulse duration of 6.7 ps. The linearly polarized laser beam was guided over the samples by a twomirror galvo-scanner (SCANLAB AG) system equipped with a 100 mm telecentric fθ lens. The angle of incidence was perpendicular to the sample surface. The laser beam had a Gaussian power density profile. Lasermachining was performed in a clean room under atmospheric conditions with an apparent spot size of 18 μm. Four set of samples were produced with 12, 24, 36, and 48 pulses/irradiation spot on both SS and Ti-6Al-4V to study the influence of the number of pulses per irradiation spot on the nanoscale structures. Initially, a line has been machined with a pulse to pulse overlap of more than 90%. The successive lines were laid with an overlap of 50% as illustrated in Figure 1, to minimize the wavy structure between successive lines on the laser-machined surface. An area of 5 mm 5 mm was machined to perform the contact angle (CA) measurements. The wavy structures formed on the metal surface are termed as a whole as “ripples”, and an individual ripple is termed a “ridge”. The laser processing conditions are listed in Table 1. The laser-machined structures were analyzed using a scanning electron microscope and an atomic force microscope. The periodicity of the ripples was measured from the SEM pictures. The periodicity measurements were performed at different locations to reduce the measurement error. Water-stable, transparent, hydrophobic SAM based on perfluorinated octyl trichlorosilane (FOTS) were coated by chemical vapor deposition on a laser-machined surface. Details of the procedure can be found in the literature.35 37 The hydrophobicity of the samples was evaluated by measuring the static contact angle using the sessile drop technique, with a video-based optical contact angle measuring device (OCA 15 plus from Data Physics Instruments). A 2 μL droplet of distilled deionized water was dispensed on the laser-machined surface under atmospheric conditions, and the static contact angle was determined by analyzing droplet 8465
dx.doi.org/10.1021/la2011088 |Langmuir 2011, 27, 8464–8469
Langmuir
ARTICLE
Figure 2. SEM picturess of stainless steel samples after (a) 12, (b) 24, (c) 36, and (d) 48 pulses. The arrow indicates the direction of polarization of the laser beam. images. Similarly, we made the contact line of the liquid drop with the laser-machined surface advance or retreat by adding or withdrawing, respectively, a small volume of liquid. The advancing and receding contact angles were measured, and the difference between them is contact angle hysteresis.38
3. RESULTS 3.1. Microscopic Analysis. Figure 2 shows nanoscale ripples produced on SS sheets with 12, 24, 36, and 48 pulses per irradiation spot. LIPSS were formed with the direction perpendicular to the polarization of the laser beam. The lateral dimensions of the ripple structure were measured from the SEM images. For the SS samples, with the increasing number of pulses per irradiated spot, the periodicity (the distance between the successive ridges) of the ripples decreases and was found to be in the range of 200 300 nm. The regular/large ripples are retained at low-fluence areas. The disappearance of ripples at the center of the laser spot can be explained by the ablation threshold of SS. The ablation threshold of SS in the picosecond regime is 50 mJ/ cm 2.39 In this experiment, laser-machining was performed with a relatively high fluence (98 mJ/cm 2) compared to the ablation threshold of SS. Further, the ablation threshold decreases with an increasing number of pulses per irradiated spot, because of the accumulation effect. From panels c and d of Figure 2, it appears that the center of the irradiated spot has undergone some melting and contours of a nearly flat surface (encircled spot as shown in the figure) have formed, as the number of pulses per irradiated spot has increased. Between the regions of large ripples, nanoscale protrusions were also observed. The amount of such protrusions increases with the number of pulses per irradiated spot. The experiments were performed with a Gaussian beam, and it has a higher fluence at the center of the circular spot compared to the edges. In the center of the beam, where the fluence is high, the ripples are broken and transformed to globular-like structure (white spots in Figure 2, a d), which is primarily found on top of the ridges. The extent of such structure increases with an increasing number of pulses per irradiated spot, and its dimension is a nanoscale length (20 30 nm). Such a structure is here termed a nanoscale protrusion. With an increasing number of pulses per irradiated spot, there is an increase in the amount of
Figure 3. SEM micrographs of the Ti-6Al-4V sample after (A1 and A2) 12, (B1 and B2) 24, (C1 and C2) 36, and (D1 and D2) 48 pulses. The white arrow indicates the direction of polaization of the laser beam.
debris (redeposited material) on the surface. This is perhaps caused by the rapid cooling and solidification of metal vapor. Ti-6Al-4V was machined with the same laser fluence that was used on SS. The fluence used for machining Ti-6Al-4V is well below the ablation threshold. The decrease in ablation threshold due to the incubation effect cannot be denied. As one can see in Figure 3, the topography of Ti-6A1-4V shows large LIPSS with nanoscale protrusions. The LIPSS are broken at high-fluence areas, and nanoscale protrusions are present in these places. The width of the ridges was in the range of 200 250 nm. In contrast to the surface morphology observed in SS, the Ti-6Al-4V surface shows more densely packed nanoscale protrusions. Further, the density of nanoscale protrusions on the surface and the size of the protrusions increase with the number of pulses per irradiated spot (Figure 3, A2, B2, C2, and D2). The top surfaces of the ridges are covered with nanoscale protrusions. Because the laser intensity is high at the spot center, under repetitious pulses, the material undergoes localized melting within the laser spot. The high radial temperature gradient in the localized melt region can induce a radial surface tension gradient that ejects the liquid to the periphery of the melt region.40 This may lead to the formation of nanoprotrusions due to the rapid cooling of the ejected liquid on the boundary 8466
dx.doi.org/10.1021/la2011088 |Langmuir 2011, 27, 8464–8469
Langmuir
ARTICLE
Figure 5. Average roughness of Ti-6Al-4V and SS as a function of the number of pulses per irradiated spot.
Figure 4. Noncontact AFM three-dimensional images obtained with the Ti-6Al-4V sample for a scan area of 10 μm 10 μm for 12, 24, 36, and 48 pulses per irradiated spot and corresponding cross-sectional profiles: (A1 and A2) 12, (B1 and B2) 24, (C1 and C2) 36, and (D1 and D2) 48 pulses.
with the solid state material. The random nanostructures induced during the first shot improve the absorption of laser light41 and enhance the growth of nanostructures with an increasing number of pulses. The increase in the number of pulses per irradiated spot can lead to the formation of microscale structures. It is important to note that the top of the microscale structure is covered with nanoscale protrusions. From the microstructure analysis of SS and Ti-6Al-4V, it is clearly proven that nanoscale structure (ripples and nanoscale protrusions) would be retained if the machining were conducted below the ablation threshold. 3.2. AFM Analysis. The topography analysis via AFM has been restricted to the Ti-6Al-4V sample that exhibited superhydrophobicity after laser micromachining and silanization. The surface morphology of the sample with 12 pulses per irradiated spot (Figure 4, A1) exhibits large ripples. The surface profile was analyzed for a cross section transverse to the large ripples (Figure 4, A2), which indicated the periodicity to be equal to or less than the wavelength of the laser beam. In the high-fluence region, as the number of pulses increases, the nanoscale protrusions become prominent. A complete breakdown of large ripples has been observed for the sample with 48 pulses per irradiated
spot (Figure 4, D1). Moreover, three-dimensional AFM images show the formation of the nanoscale protrusions on top of ridges for the samples with 36 and 48 pulses per irradiated spot (Figure 4, C1 and D1). Cross-sectional profiling of the 10 μm 10 μm scanned area (shown in the three-dimensional images) was performed to estimate the average height of the ridge as well as the nanoscale protrusions. The cross-sectional profile shows sharp and narrow peaks for the sample machined with 12 pulses per irradiated spot (Figure 4, A2), suggesting that the nanoscale protrusions are limited compared to the other samples. The average height of the ridges on the sample with 12 pulses per irradiated spot was ∼40 nm, and the nanoscale protrusions are not pronounced. The cross-sectional profiling of the samples obtained with 24, 36, and 48 pulses per irradiated spot indicated average ripple heights of 75, 85, and 105 nm, respectively. The nanoscale protrusions were in the range of 20 30 nm (Figure 4, B2, C2, and D2) for all these samples. Panels C1 and D1 of Figure 4 show that the tops of the ripples and ridges are covered with nanoscale protrusions. Formation of nanoscale protrusions on the top of the ridges is represented by nanoscale peaks at the top of the major peaks (Figure 4, C2 and D2). The size of the nanoscale protrusions is small in samples with 12 and 24 pulses per irradiated spot compared to that in samples with 48 pulses per irradiated spot. The trend observed from the cross-sectional profiles for the size of the nanoscale protrusions as well as the lateral dimensions of ridges indicates that, as the amount of energy added to the substrate increases, pure nanoscale structure will be transformed into dual-scale (nano and micro) structure. Figure 5 shows the average roughness (Ra) as a function of the number of pulses per irradiated spot. The average roughness value increased linearly with the number of pulses per irradiated spot, for the Ti-6Al-4V sample. In the case of SS, the average roughness increases almost linearly up to 36 pulses per irradiated spot and appears to stabilize. The average roughness values were nearly the same for 36 and 48 pulses per irradiated spot. 3.3. Contact Angle Measurements. The effect of lasermachining on the wetting properties of the FOTS-coated samples was evaluated by static sessile drop contact angle (CA) measurements, using a droplet size of 2 μL. Figure 6 shows 8467
dx.doi.org/10.1021/la2011088 |Langmuir 2011, 27, 8464–8469
Langmuir
ARTICLE
superhydrophobicity is perhaps related to the larger amount of nanoscale protrusions present on these samples.
Figure 6. Static contact angle as a function of surface roughness.
Figure 7. Hysteresis of Ti-6Al-4V as a function of surface roughness.
the static CA as a function of roughness for SS and Ti-6Al-4V samples. Regardless of the base metal used, hydrophobicity was found to increase with an increasing amount of roughness. Such an increase in CA is seen only up to 36 pulses per irradiated spot; with the 48 pulses per irradiated spot, no improvement was observed. The Ti-6Al-4V sample exhibited superhydrophobicity with an apparent CA of 152 ( 3°. This corresponds to an increase in hydrophobicity of ∼27% when compared to that of the unmachined Ti-6Al-4V surface, which shows a CA of ∼120 ( 3°. In case of SS, the apparent CA increased from 125 ( 3° to 140 ( 3° (a 12% improvement) because of the increased amount of roughness. Superhydrophobicity of Ti-6Al-4V substrates was also tested for water contact angle hysteresis (CAH), as CAH values of
