Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

The morphological and optical characteristics of femtosecond laser-induced large-area micro/nanostructures on GaAs, Si, and brass

Open Access Open Access

Abstract

We systematically study the morphological and optical characteristics of the large-area micro/nanostructures produced by femtosecond laser irradiation on GaAs, Si, and brass. The experimental results demonstrate that along with the increase of laser fluence, significant changes in the surface morphology can be observed, and the most prominent phenomenon is the enlarging of the feature size of formed structures. Interestingly, by the fourier analysis of the treated areas, a peculiar phenomenon can be revealed: as laser fluence increases, the spatial frequencies of the structures change following a specific law – the allowed main frequencies are discrete, and appear to be a sequence of 2f, f, f/2, f/4, and f/8 (f is the fundamental frequency corresponding to the near-subwavelength ripples). In our opinion, the new frequency components of f/2, f/4, and f/8 originate in the 2-order, 4-order, and 8-order grating coupling. The law can offer us new insights for the evolving mechanisms of a variety of laser-induced micro/nanostructures in different scales. Furthermore, the optical characteristics of the treated surface are strongly dependent on the morphological characteristics that are mainly determined by laser fluence, such as the feature size of the micro/nanostructures, the topology of the surface morphology, the surface roughness, and the irregular degree of the formed structures. In general, as laser fluence increases in a moderate range, the specular reflectance of the structured surface would be significantly reduced. However, if laser fluence is excessive, the anti-specular-reflection effect would be much weakened. In ideal laser fluence, the micro/nanostructures produced by the near-infrared laser can achieve an ultra-low specular reflectance in the visible and near-infrared spectral region, which exhibits an attracting application prospect in the field of utilizing solar energy.

©2010 Optical Society of America

1. Introduction

Recently, the method of preparation of micro/nanostructures on material surface by direct femtosecond (fs) laser irradiation has aroused great interest of researchers in the fields of laser processing and optoelectronics [110]. This is mainly because after fs laser processing, the electro-optical characteristics of such a micro/nanostructured surface change greatly and on occasion exhibit excellent features. For example, “black Si” prepared by fs laser has superior absorption properties, for which the technique has achieved a successful application in solar cell [2,46]. In recent years, due to the world's energy shortages, the enthusiasm of development of new energy sources continues to rise, such as utilizing the renewable energy (solar power, wind power, hydroelectric power, tidal power, etc) as an alternative to fossil fuels and nuclear power. For the promising way of harvesting solar energy – using solar cells to convert sunlight into electricity, an important bottleneck is the low solar photovoltaic conversion efficiency. Therefore, how to improve the efficiency of solar cells is a priority issue for the researchers. In a variety of techniques to improve the efficiency of solar cells, micro/nanostructured treatment of solar cell superficies to enhance the efficiency of sunlight absorption is an important means, such as the micro pointed cones on “black Si” produced by fs laser irradiation. Besides “black Si”, nowadays, the interesting phenomenon that the formation of micro/nanostructures on metals induced by direct fs laser irradiation would significantly change the optical properties of metals [79] (formation of “black metal” or even “color metal” [9]), have attracted great interest of researchers. In addition, as an inverse effect, by applying the blackening technique to a tungsten incandescent lamp filament [10], the tungsten lamp can be dramatically brightened. These intriguing results indicate that the direct fs laser processing technique have opened up a new road for material modification.

In laser processing, the appearance of various micro/nanostructures is a universal phenomenon, such as kinds of laser-induced periodic surface structures (LIPSSs) [1140]. In order to understand the curious optical characteristics of the laser-processing surfaces as reviewed above, the key is to inspect the structures of the treated surface in a microcosmic view. But because the micro-morphologies of the treated surfaces are always complex, it is not easy to obtain a direct and clear relationship between the optical properties and the morphologies. In particular, for the processing of ultrafast laser, under different irradiation conditions, the characteristics of the formed structures change greatly [1126,3040]. This mainly embodies in the characteristic dimension of the formed structures, which can vary from sub-100 nm to a few μm. For instance, the deep-subwavelength ripples (DSRs) are apt to be observed in the case of multi-photon fs irradiation with fluence near ablation threshold [1126], and the near-subwavelength ripples (NSRs) [11,13,16,17,21,22,25,31,3340], which would frequently appear accompanied with DSRs, tend to form at slightly higher fluence (note that the different abbreviations LSFL (low spatial frequency LIPSS) and HSFL (high spatial frequency LIPSS) for the structures are widely used in the literatures, which refers to the ripples with a spatial period close to the irradiation wavelength and the ripples with a spatial period significantly smaller than the irradiation wavelength, respectively; we use the new abbreviation NSR instead of LSFL in order to refer in particular to “near-subwavelength” characteristics of certain non-classical LSFLs, and in accord with NSR we use DSR instead of HSFL). To figure out the respective effects of the different scale structures on the optical characteristics of the complex surface is the primary way to understand the curious optical characteristics of the treated surface.

Up to now, although abundant experimental results concerning the fs laser-induced micro/nanostructures on semiconductors, metals, and dielectrics have been reported, systematical studies on the morphological and optical properties of such micro/nanostructures, especially on the dependence between the unique optical properties and the surface morphologies, are still lacking. Actually, such studies can provide us new insights into the origin of a variety of exotic morphological and optical properties of the fs laser-induced micro/nanostructures and useful guides for the preparation of structures with excellent optical characteristics for specific applications.

In the paper, with the techniques of fabricating large-area uniform micro/nanostructures by direct fs laser irradiation, we study the morphological characteristics of the induced structures on semiconductors (GaAs and Si) and metal (brass) at different laser fluence. Our results indicate that the surface morphologies of the treated area are mainly determined by laser fluence. With the laser fluence varies in a wide range, surface morphologies of the formed structures have a very significant change. By the fast fourier transform (FFT) of the scanning electron microscope (SEM) images of the treated areas, we can investigate the evolution of the surface morphologies in detail via the spatial frequency information, which is able to provide us with more profound views to the implicit messages adherent to the micro/nanostructured surfaces. Interestingly, by the FFT analysis method, a peculiar phenomenon can be revealed: as laser fluence increases, the spatial frequency of the micro/nanostructures changes following a specific law – the main frequency divides by 2 serially, which appears to be a sequence of 2f, f, f/2, f/4, and f/8 (f is the fundamental frequency corresponding to the NSRs). To our knowledge, such a strange phenomenon in laser ablation has not yet been reported before, which can offer us new insights for the evolving mechanisms of a variety of laser-induced micro/nanostructures in different scales.

Furthermore, by measuring the relative specular reflectance, we deliberate the influence of surface morphologies on the optical characteristics, particularly on the anti-specular-reflection characteristics. The systematic study on the dependence of surface morphology and spectral characteristics would help us to look into the decisive factor for the anti-specular-reflection characteristics, which in turn could give us guidance for fabricating surface structures with excellent anti-specular-reflection performance. By the comparison of the morphological and spectral results, it can be seen that the pronounced morphological change owing to the varying of laser fluence in a wide range gives rise to the prominent change of the optical characteristics, which are strongly dependent on certain essential elements of surface morphology, such as the feature size of the micro/nanostructures, the topology of the surface morphology, the surface roughness, and the irregular degree of the formed structures. In general, as laser fluence increases in a moderate range, the anti-specular-reflection efficiency of the structured surface would be significantly improved, exhibiting a lower specular reflectance and a wider spectrum of responses. However, if laser fluence is excessive, the anti-specular-reflective effect would be much weakened. In ideal laser fluence, the near-infrared laser-induced micro/nanostructures can achieve an ultra-low specular reflectance in the visible and near-infrared spectral region.

2. Experiments

The experimental setup used here for fabricating large area uniform nanostructures by direct fs laser ablation was similar to that reported in Ref. 24. A 125-fs Ti:Sapphire laser system (Spectra Physics Hurricane) operating at a 1-kHz repetition rate was used in the study. The central wavelength of 800 nm exported by the ultrafast amplifier was used in the experiments of brass, and the infrared laser of 1280 nm exported by the optical parametric amplification (OPA) was used in the experiments of Si and GaAs. An infrared laser for Si and GaAs was of great benefit to get a more abundant hierarchy of the formed micro/nanostructures, such as the formation of fine ripples and fine-coarse hybrid ripples. In the study, we focused on the ripple structures fabricated by linearly polarized laser with different laser fluence at normal incidence. The processing technique resembled that in Ref. 24 with the following parameters: scanning areas of 400 × 400 μm2 for GaAs and Si and 200 × 200 μm2 for brass, focal lengths of 150 mm for GaAs and Si and 300 mm for brass, scanning interval of 8 μm, and scanning velocity of 125 μm/s. For the lenses with the focal lengths of 150 mm and 300 mm, via the knife-edge method the focal spot sizes (D) (width at waist defined by 1/e2 points) of the Gaussian beam, which are close to the beam diameters in the sample processing plane under good focused condition, were measured as 36 μm for the wavelength of 1280 nm and 45 μm for the wavelength of 800 nm, respectively. Then, the laser fluence defines as the average fluence of E/A, where E is the energy of fs laser pulse, and A is the focal spot area of πD2/4. For the ablation experiments of semiconductors, a polished GaAs single crystal sample ((100) surface, surface roughness < 5 Å) and a polished Si single crystal sample ((100) surface, surface roughness < 5 Å) with dimensions of 10 × 10 × 1mm3 were used in this study. For the ablation experiments of metal, a polished brass (CuZn) sample with dimensions of 20 × 20 × 2mm3 was used (the surface of brass was first roughly polished with SiC emery paper to remove the oxide layers, and then was polished with polishing cloth in order to achieve a superior mirror finish).

After the samples were treated, two kinds of measurements were carried out for the investigation of the morphological and optical characteristics. Above all, the surface morphologies of the formed nanostructures were analyzed with SEM (JEOL JSM-6380). Then, the optical characteristics of the micro/nanostructured surface were surveyed by a spectral method with a micro-region spectral detection system built by ourselves. Concretely, the spectrum method is through measuring the relative specular reflectance Rr of incident light as a function of wavelength, which is defined as

Rr(λ)=It(λ)Ii(λ)
where It is the intensity of specular reflecting light of the treated area and Ii is the intensity of specular reflecting light of the intrinsic surface (It and Ii are measured under the same light irradiation condition, and always Ii are measured near by the treated area for the corresponding It). The measurement of the specular reflectance was carried out via the micro-region spectral detection system as shown in Fig. 1 . The sample was placed in a three-dimension micro-positioning stage with the surface normal oriented horizontally (the wave vector of the ripples on the treated surface was in the incidence plane of the illumination). A broadband incandescent lamp equipped with a condenser reflective concave mirror was used as the illumination source, which was able to cover a wide spectral range from 400 nm to 1000 nm and emitted unpolarized light with a small divergence angle. The emitted light irradiated the sample surface horizontally with an incidence angle θ of 34° for GaAs and Si and 44° for brass, and then the specular reflecting light was collected by a long-focus microscope objective. There were two purposes for the use of the optical microscope (OM). One was to determine the measuring position and take the photo of the treated area through the OM under the same condition as the spectral measurement, which was accomplished by a CCD monitoring system. The other was to collect the specular reflecting light, which was focused and coupled into the fiber optic probe of the fiber optic spectrometer (Ocean Optics, HR2000). Here the spot size of the analyzing optical beam was about 50 μm, which was smaller than the size of the treated areas but large enough to achieve stable results for spectral analysis. Then the fiber optic spectrometer recorded the spectral response curve (the dependence of wavelength and light intensity) of the detected area. Through dividing the data of the treated area by the data of the intrinsic area, we could get the curve of the relative specular reflectance of the treated area. By the means, we were able to investigate the relationship between the surface morphologies and the optical characteristics of the micro/nanostructures prepared under different fs-laser irradiation conditions.

 figure: Fig. 1

Fig. 1 A schematic view of the experimental setup for the spectral method.

Download Full Size | PDF

3. Results

3.1 The experimental results of GaAs

3.1.1 The morphological characteristics of the treated areas on GaAs

With the irradiation wavelength of 1280 nm, whose photon energy is smaller than the bandgap of GaAs (1.43 eV), we are able to get various surface morphologies in the treated area by varying laser fluence. As the SEM observations shown in Fig. 2 , where the laser fluence for area (a), (b), (c), (d) and (e) is 0.09, 0.11, 0.15, 0.19 and 0.25J/cm2, respectively, when laser fluence is approaching the ablation threshold of GaAs (the case of (a)), the irradiated area is only locally damaged and such localized ablated areas covered by mixed fine-coarse ripples are formed randomly. In the peripheral area, due to the longer irradiation time, it is easier to induce micro/nanostructures with a large area. When laser fluence is slightly higher than the ablation threshold (the case of (b)), the induced ripples cover the whole irradiated area, and the configuration is still compound – contain NSRs with period of about 700 nm and DSRs with period of about 300 nm. With regard to semiconductors, owing to the high electronic mobility the thermal effect in laser processing is far more dramatic than wide bandgap materials, and consequently it is far harder to produce pure fine ripples on GaAs. As laser fluence continues to rise (the case of (c)), the population of the fine ripples decreases obviously, and the coarse ripples dominate the ablated area. Meanwhile, it can be seen that as laser fluence increases, the ripples become disordered due to the enhancement of the thermal effect. But in the periphery, large-area regular fine ripples still can be observed (no shown here), resembling the situation in fixed point ablation. When laser fluence becomes higher (the case of (d)), the surface morphology continues to evolve – the degree of ripple irregularity further raises, and some longitudinal vein-like structures appear and interlace with the ripples. As laser fluence further increases, such vein-like structures keep on strengthening, thus make the whole ablated area become a complex pattern in a large scale (the case of (e)). From the results, it can be seen that in fs laser processing the influence of laser fluence on the surface morphology for semiconductors is more serious than for wide bandgap materials [24]. On the other hand, on account of the significant correlation between the optical characteristics and the surface morphology, it can be expected that the optical properties of the micro/nanostructured surfaces on GaAs ablated in various laser fluences differ greatly.

 figure: Fig. 2

Fig. 2 The experimental results of the micro/nanostructures on GaAs produced by 1280-nm fs laser of linear polarization with different fluences. The laser fluence for area (a), (b), (c), (d) and (e) is 0.09, 0.11, 0.15, 0.19 and 0.25J/cm2, respectively. From the following processing parameters: the scanning velocity of 125 μm/s, the scanning interval of 8 μm, the pulse repetition frequency of 1 kHz, and the focal spot size of 36 μm, the average number of laser pulses per spot can be calculated to be about 1000 pulses. The sub-figures of left column are the OM images of the 400 × 400-μm2 treated areas taken under the same condition as spectral measurement. The sub-figures of the middle two columns are SEM images of the treated areas with different magnification factors. In (a) the double-headed arrow indicates the direction of laser polarization (in (b) to (e) the polarization directions are the same as in (a)). The sub-figures of the right column are the FFT images of the corresponding SEM images, which transform from the SEM images with small magnification factor except (a) and rescale to have the same spatial frequency scale.

Download Full Size | PDF

The FFT images of the SEM images of treated areas are shown in the right column, which definitely demonstrate certain typical characteristics for the morphologies of different treated areas. For the area irradiated by fs laser with threshold fluence (the case of (a)), the FFT image reveals three principal frequency components, corresponding to the NSRs (f component), the second harmonic of NSRs (2f component, mean that the profile of NSRs markedly deviates from ideal sine envolope), and the DSRs (>2f component), respectively. The increase of laser fluence would make the DSR component approach 2f (the case of (b)) – it seems that certain DSRs come from the splitting of the NSRs in half. Such a 2f DSR component has also been observed in some recent literatures [3840]. Whereas, along with the further increase of laser fluence, the NSR component is greatly enhanced (the case of (c)), and exhibits a red-shift – the period become larger. Meanwhile, the increase of laser fluence would weaken the DSR component, which ultimately almost disappears for laser fluence large enough (the cases of (c), (d), and (e)). Interestingly, when laser fluence reaches certain range (the case of (c)), a new frequency component appear, which is close to half (f/2) of the spatial frequency of NSRs. More curiously, with the further increase of laser fluence, such a new frequency component changes discretely to be a quarter (f/4) and an eighth (f/8) of the spatial frequency of NSRs (the cases of d and e, respectively). In addition, the intensity of the f/2, f/4, and f/8 components increases in sequence and eventually dominates in the FFT image, whereas the intensity of the f component decreases accordingly. On the whole, a distinct law for the evolution of the surface apertures arises – the spatial frequency of the structures gets smaller as laser fluence raises, and divides by 2 serially to be a sequence of 2f, f, f/2, f/4, and f/8. This conclusion is further confirmed by the results of Si.

From the FFT images, we can transform the spatial frequency components in the direction of ripple wave vector to the real space, as shown in Fig. 3 . The principal frequency components are tagged in the corresponding positions, from which we can intuitively see that the components of DSRs, NSRs, f/2, f/4, and f/8 have the real space scales of about 0.3, 1, 2, 4, and 8 μm, respectively. Clearly, the real space representation of the frequency components of DSRs and NSRs is consistent with the morphological characteristics of the direct SEM observation. From the figure, it is easy to see that with the increasing of laser fluence the dominant spatial scales of the laser-induced structures shift from a few hundred nanometers to a few micrometers. In detail, a phenomenon for the NSRs can be inspected – the peak of NSR moves toward large scale as laser fluence increases and tends to approach a maxima that is related to laser wavelength of 1280 nm.

 figure: Fig. 3

Fig. 3 The real space representation of the spatial frequency components of the treated GaAs surfaces in the direction of ripple wave vector.

Download Full Size | PDF

3.1.2 The optical characteristics of the treated areas on GaAs

From OM images (in Fig. 2) of the 400 × 400-μm2 treated areas taken under the same condition as spectral measurement, it can be seen that as laser fluence increases, the treated area becomes darker and darker, which exhibits quite a contrast to the untreated area. It means that the specular reflectance of the treated surface significantly reduces. In detail, for area (a) the specular reflectance is only slightly reduced; nevertheless, for area (b)) where the whole treated area is covered by ripples, the specular reflectance is evidently reduced (in the case trivial reflected light still exists). As the surface roughness further increases (for areas (c), (d), and (e)), the specular reflectance becomes fairly low, and through the direct OM observation we already cannot distinguish the difference among these areas. In the following, with the spectrum method we can see that the specular reflectance of these surfaces keeps on decreasing.

The relative specular reflectances (Rr) of the treated areas on GaAs fabricated by 1280-nm fs laser in various fluence conditions are shown in Fig. 4 . It is noticeable that the results of spectral measurement are consistent with direct observations. On the whole, as laser fluence increases, Rr of the micro/nanostructured surface obviously decreases. In the situation of partial damage (curve a), the low specular reflectance distributes in the purple range, hencethe surface shows a slight yellow bias, as shown in Fig. 2(a). As long as the whole treated area is covered by ripples (curve b), the specular reflectance is greatly reduced, especially for the short-wavelength range, which coincides with the morphological characteristics of the structured surfaces (the feature sizes of the ripples). This is because the micro/nanostructures tend to act on incident light with wavelength close to their feature size. That is, for the case DSRs dominate in the treated area, Rr in the wavelength range near the DSR scale would reduce more significantly. When the coarse ripples (include NSRs and certain larger-scale structures (the frequency components of ≤f/2)) dominate the treated area (curve c), Rr in the red and infrared band is also reduced greatly. In addition, because the ripples are more random (period distributes in a wide range), in a wide spectrum range Rr is at a low value. Along with the further increase of laser fluence (curve d), which induces the enhancement of the irregularity of the ripples and the deepening of the ripple grooves, in a wide spectrum range Rr of the treated surface reduces obviously, especially in the infrared band (through the directOM observation, because of the ultralow specular reflectance, we are already not able to distinguish the difference between area (c) and area (d)). At last, when the surface turns to be more uneven and irregularity in a larger scale (curve e), Rr further decreases. Here the spectral characteristic is quite similar with the curve d. The phenomenon originates in the fact that in the two situations the surface morphologies have an analogous topological structure, which determines the spectral response. But in the case of area (e), due to the deeper grooves on the surface, curve e displays a lower Rr.

 figure: Fig. 4

Fig. 4 The relative specular reflectances (Rr) of the treated areas on GaAs fabricated by 1280-nm fs laser in various fluence conditions as a function of laser wavelength (λ). Here curves a to e correspond to areas (a) to (e) in Fig. 2, respectively. The dashed line represents the reference spectrum of the untreated surface that is always equal to 1.

Download Full Size | PDF

3.2 The experimental results of Si

3.2.1 The morphological characteristics of the treated areas on Si

By the same techniques, we processed the surface of Si sample with different laser fluence as shown in Fig. 5 , which displays the treated areas with different morphological characteristics. Here the laser fluence for area (a), (b), (c), (d), (e), (f) and (g) is 0.16, 0.2, 0.26, 0.32, 0.4, 0.48, and 0.6J/cm2, respectively. When laser fluence is approaching ablation threshold (the case of (a)), the localized ripples appear randomly, to some extent resembling the case of GaAs. It should be noted that on Si the localized ablated areas dominate by coarse ripples instead of mixed fine-coarse ripples on GaAs. In addition, different from the case of GaAs, here a new kind of self-organized particle structures emerges in the irradiated area without ablation. With laser fluence slightly higher than ablation threshold (the case of (b)), ripples are formed in the whole treated area like the case of GaAs. Although some localized DSRs arise due to the splitting of NSRs that is similar to the case of GaAs in Fig. 2(b), here it is harder to find DSRs and as a result NSRs still dominate. As laser fluence keeps on a steady increase (the cases of (c) and (d)), NSRs are gradually enhanced mainly via the deepening of the grooves. When laser fluence reaches 0.4J/cm2 (the case of (e)), another new kind of morphological characteristic can be observed. The features of the structures remain certain characteristics of ripples. Whereas, the continuity of the ripples becomes poor; meanwhile, in the grooves a number of deep-hole structures appear. Nevertheless, with the further increase of laser fluence (the case of (f)), such a structured surface begins to be polluted by the re-deposited materials from laser plumes. Such re-deposited materials initially appear on the tops of ripple ridges and the tips of cones, and cover the whole treated area along with the increase of laser fluence (the case of (g)). Comparing the results on GaAs and Si, we can see that towards different semiconductors, the morphologies of the fs laser-induced structures demonstrate a clear distinction, which is more complicated than the case of wide bandgap materials [24]. This difference is because of the more pronounced thermal effect in the process of laser ablating semiconductors, which relates to the intrinsic electronic transport properties. In addition, the bandgap of Si (1.12eV) is smaller than that of GaAs (1.43eV), and more near the photon energy (0.97eV) of the laser light used in experiments. As a result, under the same irradiation condition, the thermal effect for Si is more prominent than for GaAs. Thus, the formed structures show different morphologies.

 figure: Fig. 5

Fig. 5 The experimental results of the micro/nanostructures on Si produced by 1280-nm fs laser of linear polarization with different fluences. The laser fluence for area (a), (b), (c), (d), (e), (f), and (g) is 0.16, 0.20, 0.26, 0.32, 0.40, 0.48, and 0.60 J/cm2, respectively. From the following processing parameters: the scanning velocity of 125 μm/s, the scanning interval of 8 μm, the pulse repetition frequency of 1 kHz, and the focal spot size of 36 μm, the average number of laser pulses per spot can be calculated to be about 1000 pulses. The sub-figures of left column are the OM images of the 400 × 400-μm2 treated areas taken under the same condition as spectral measurement. The sub-figures of the middle two columns are SEM images of the treated areas with different magnification factors. In (a) the double-headed arrow indicates the direction of laser polarization (in (b) to (g) the polarization directions are the same as in (a)). The sub-figures of the right column are the FFT images of the corresponding SEM images, which transform from the SEM images with small magnification factor except (a) and rescale to have the same spatial frequency scale for ease of comparison.

Download Full Size | PDF

Resembling the results of GaAs, the FFT results for Si also deliver significant information for the morphological characteristic of the treated surface. For the case of (a), the FFT image reveals two principal frequency components related to NSRs (f component) and the second harmonic of NSRs (2f component). The higher order harmonic of NSRs can also be observed in the FFT image with a larger frequency scale, but not shown here. The evident high order components of NSRs imply that the ripples are provided with a non-sinusoidal grating profile that is in favor of coupling with the incident laser in the ablating process and enhances absorption of laser light [35]. On the other hand, in the FFT image the DSRs component is hard to be observed, indicating that it is more difficult to produce fine ripples on Si than on GaAs. Along with the increase of laser fluence, the NSR component is enhanced, exhibits a red-shift (the case of (b)), and become saturated (the cases of (c)). Then, similar to the case of GaAs, when laser fluence raises to certain range (the case of (d)), a new frequency components clearly appear, which is equal to half (f/2) of the spatial frequency of NSRs. The f/2 component is dominant for higher laser fluence (the case of (e)), and as a result the f component is blanking. Such a phenomenon doesn’t occur on GaAs, where the f component is always one of the main components. As laser fluence further increases (the case of (f)), a distinctive feature appear – another new frequency arises, which locates in frequency near a quarter (f/4) of the spatial frequency of NSRs. Meanwhile, the f/2 component is blanking. When the treated surface is seriously polluted by re-deposited materials (the case of (g)), the f/4 component maintains the dominant frequency component of the FFT image. Moreover, a week continuous component spread to the high frequency range, which comes of the re-deposited materials. By comparing the SEM and FFT images between GaAs and Si, we can get an instructive insight for the difference of the results between the two kinds of semiconductors: the ability of generating more prominent low-frequency components is the reason for Si to form structures with more complicated morphology other than ripple structures in high laser fluence. Generally speaking, the law for the evolution of the spatial frequency is further confirmed by the results of Si: the spatial frequency evolves following a sequence of 2f, f, f/2, and f/4.

The real-space representation of the spatial frequency components in the direction of ripple wave vector is shown in Fig. 6 . The main frequency components are also tagged in the corresponding positions. Here the situation of the 2f and f components is similar to that of GaAs. In addition, for the case of Si we can observe more prominent peaks for components of f/2 and f/4, which definitely locates at about 2 μm and 4 μm, respectively.

 figure: Fig. 6

Fig. 6 The real space representation of the spatial frequency components of the treated Si surfaces in the direction of ripple wave vector.

Download Full Size | PDF

3.2.2 The optical characteristics of the treated areas on Si

The OM images of the 400 × 400μm2 treated areas are shown in the left column of Fig. 5. Obviously, difference can be easily observed for the images between Si and GaAs, which is mainly manifested in the areas irradiated by high laser fluence, such as the cases of (f) and (g). Under these irradiation conditions, the treated areas display a higher specular reflectance than the case of moderate laser fluence. Especially for the situation of (g), the treated surface appears to be a distinct brown. This phenomenon indicates that in the irradiation condition of high laser fluence, the re-deposited materials covering the pre-formed micro/nanostructures would seriously influence the optical characteristics of the treated surface, and make the anti-reflection effect of the micro/nanostructures decline. In addition, from the results (a) to (e), it can be seen that as long as the formation of the ripple structures, the specular reflectance of the treated surfaces reduces greatly; in a wide range of laser fluence, the micro/nanostructures produced by fs laser show a favorable anti-specular-reflective property. As shown in (a), the specular reflectance in the localized areas where ripples are formed is apparently lower than that in the areas with the intrinsic surface.

Corresponding to the direct optical observations, the spectral measurements of the treated areas demonstrate the expected characteristics as shown in Fig. 7 . In area (a), owing to the formation of localized ripples as well as nano-particles, Rr of the area is lower than that of the untreated surface, particularly in the spectral range of near-purple. From the other spectral curves, we can see that after the whole area is covered by ripples (curves from b to g), Rr decreases significantly – lies in a small range (Rr<0.1). There is somewhat different between Si and GaAs. For GaAs, Rr of the large-area ripples (admixture of DSRs and NSRs) produced with laser fluence near ablation threshold is not very low (curve b in Fig. 4). Whereas, for Si, once the large-area ripples (NSRs dominant) are formed, Rr becomes very low (curve b). It seems that in the two situations the different types of ripples that have different spectral response range are responsible for the phenomenon. Similarly, along with laser fluence increasing in a moderate range, Rr continues to decline (curve c). But when laser fluence reaches certain range (curve d), we can observe obvious change about the spectral response: Rr in the spectral range of purple rises, and Rr in the spectral range of red remains a decline trend. This phenomenon originates in the weakening of the fine ripples and the further enhancing of the coarse ripples (the deepening of grooves and the larger space-scale component emerges (f/2 component)). When laser fluence reaches the value for new morphology to appear (the case of (e)), Rr continues to decline in the whole measured spectral range. As the SEM image of Fig. 5(e) shown, such a structure has a large surface roughness (deep grooves and hole-like structures), which is the origin of the low specular reflectance. In particular, in the spectral range of red and near-infrared (near the scale size of the structures), Rr achieves an ultralow value. With regard to (c), (d), and (e), on the whole thanks to the enlargement (in vertical and horizontal directions) of the feature size of formed structures by virtue of the increase of laser fluence, Rr in the spectral range of red and near-infrared shows a downward trend. Nevertheless, as laser fluence gets higher (the cases of (f) and (g)), Rr rises remarkably. The adverse effect is due to the re-deposited materials on the treated area, which would pollute the micro/nanostructures induced by direct fs laser irradiation and make the roughness diminish (the grooves and holes are padded by the re-deposited materials). In addition, there is a distinct difference between the re-deposited materials and the original materials for the crystal structures and optical properties. These factors make the specular reflectance of the re-deposited surface increase significantly, as the case of (g) shown. This phenomenon indicates that in order to obtain a surface morphology with an excellence anti-reflection effect, the laser fluence should be in an appropriate range.

 figure: Fig. 7

Fig. 7 The relative specular reflectances (Rr) of the treated areas on Si fabricated by 1280-nm fs laser in various fluence conditions as a function of laser wavelength (λ). Here curves a to g correspond to areas (a) to (g) of Fig. 5, respectively. The dashed line represents the reference spectrum of the untreated surface that is always equal to 1.

Download Full Size | PDF

3.3 The experimental results of brass

3.3.1 The morphological characteristics of the treated areas on brass

For brass, with the same processing technique, we also prepare a series of micro/nanostructured areas corresponding to different laser fluence as shown in Fig. 8 . From (a) to (h) the laser fluence is 0.36, 0.45, 0.52, 0.60, 0.68, 0.82, 1.06, and 1.39 J/cm2, respectively. In the area (a) ablated with laser fluence near the ablation threshold, some localized periodic holes and grooves of subwavelength scale appear randomly, which should derive from the excitation of localized surface plasmons (SPs) [15,19,26,2830,32,3437]. Along with the increase of laser fluence, these structures spread (the cases of (b) to (e)) and cover the whole ablated area (the cases of (f)). Interestingly, to some extent the morphology of areas (e) and (f) resembles that of areas (d) and (e) in Fig. 2 for GaAs: similar ripple structures with somewhat random, and the ripples are truncated by centain vein-like structures. The similarity means that there is a similar dynamics process for the formation of such a structure on the two kinds of materials. With higher laser fluence (the case of (g)), the ripples become blurring and are isolated by large-scale vein-like structures to form an island structure, i.e., the surface roughness gets larger. In addition, some melted or re-deposited nano-particles distribute randomly on the surface. Finally, the further increase of laser fluence would prompt some large-scale structures of river-like (the case of (h)). Meanwhile, the ripples almost disappear, and the random nano-particles cover the areas of island-like.

 figure: Fig. 8

Fig. 8 The experimental results of the micro/nanostructures on brass produced by 800-nm fs laser of linear polarization with different fluences. The laser fluence for area (a), (b), (c), (d), (e), (f), (g), and (h) is 0.36, 0.45, 0.52, 0.60, 0.68, 0.82, 1.06, and 1.39 J/cm2, respectively. From the following processing parameters: the scanning velocity of 125 μm/s, the scanning interval of 8 μm, the pulse repetition frequency of 1 kHz, and the focal spot size of 45 μm, the average number of laser pulses per spot can be calculated to be about 1600 pulses. The sub-figures of left column are the OM images of the 200 × 200-μm2 treated areas taken under the same condition as spectral measurement. The sub-figures of the middle two columns are SEM images of the treated areas with different magnification factors. In (a) the double-headed arrow indicates the direction of laser polarization (in (b) to (h) the polarization directions are the same as in (a)). The sub-figures of the right column are the FFT images of the corresponding SEM images, which transform from the SEM images with small magnification factor and rescale to have the same spatial frequency scale for ease of comparison. The frequency components in the direction perpendicular to ripple wave vector are shown in the top right corner of the corresponding FFT images (the images have been rotated 90° for ease of comparison, and some of the color scales of the images have been reset in order to highlight the perpendicular frequency components).

Download Full Size | PDF

The FFT results for brass are somewhat different from the results of GaAs and Si. For the cases with laser fluence near or slightly higher than ablation threshold (the cases from (a) to (d)), the FFT images always show a dominant frequency component related to NSRs (f component) and a weak frequency component related to f/2 spatial structures. As laser fluence continues to increase (the cases from (e) to (g)), NSR component gradually weakens, f/2 component is blanking, and the components approaching zero are growing and dominant. In detail, a weak component related to f/4 frequency emerges (the case of (e)), and is enhanced and dominant with the increase of laser fluence (the case of (f)). Eventually, with further increase of laser fluence (the cases of (h)), NSR component completely vanishes and the FFT image only leaves the frequency component close to f/8. In all, the morphological characteristics exhibited by the FFT images agree with the results of direct SEM observation: the surface morphologies show that as laser fluence rises the spatial scale of the structures on the treated surface enlarges discretely, whose dominant frequency components evolve obeying a sequence of f, f/2, f/4, and f/8. In addition, besides the frequency components in the direction of ripple wave vector, significant frequency components in the direction perpendicular to ripple wave vector can also be observed clearly, as shown in the top right corner of the corresponding FFT images (the images have been rotated 90° for ease of comparison, and some of the color scales of the images have been reset in order to highlight the perpendicular frequency components). Interestingly, the perpendicular frequency components evolve obeying a sequence of f/2, f/4, and f/8, which are generally half of the corresponding dominant frequency of the components in the direction of the ripple wave vector. Such perpendicular frequency components are related to the structures perpendicular to the ripples, for instance the vein-like structures. The results indicate that the structures perpendicular to the ripples are not random, but rather have some long range order. In addition, the similar law for the evolution of the two orthogonal frequency components means that they have a similar origin.

The real-space representation of FFT images in the direction of ripple wave vector is shown in Fig. 9 . There are two prominent characteristics for the curves: different from GaAs and Si, there is almost no shift for the NSR component, which gradually weakens as laser fluence increases; the structures with large spatial scale enhance greatly as laser fluence increases.

 figure: Fig. 9

Fig. 9 The real space representation of the spatial frequency components of the treated brass surfaces in the direction of ripple wave vector.

Download Full Size | PDF

3.3.2 The optical characteristics of the treated areas on brass

With the spectral measurement setup, the OM images of the whole 200 × 200μm2 areas treated by different fluences is shown in the left column of Fig. 8. In general, as laser fluence increases, the treated area becomes darker and darker, which means that the specular reflectance of the treated area decreases gradually. The trend of darkening continues in a wide range of laser fluence, and at last becomes saturated. In detail, as long as the whole treated surface is covered by ripples (the case of (f)) the specular reflectance is already very low. Then, with the further increase of laser fluence (the cases of (g) and (h)), the treated areas become quite dark, which indicates an extraordinary anti-specular-reflection effect. Besides, noting that the treated surface appears to be light red in the cases of low laser fluence (the cases of (a), (b), and (c)). We consider such a phenomenon is owing to certain nano-particles re-deposited on the surface, because after the sample surface had been cleaned several times with alcohol, the phenomenon of color change of the treated area could not be observed again.

The spectral responses of the micro/nanostructures on brass are shown in Fig. 10 . Along with the increase of laser fluence, Rr in the short wavelength range appears a rapid decline (curves from a to c). When laser fluence reaches a moderate range (curves d and e), Rr decreases in a wider spectral range. As the OM images shown, provided the whole treated area is covered by ripples, Rr in the near-infrared range (curve f) will also decrease intensively, which results in the good anti-specular-reflection effect with a broad spectral range (curve g). As laser fluence further increases, the broad-band anti-specular-reflection effect slightly strengthen (curve h), and is close to be saturated. In these situations (curve g and h), we can see that Rr in the near-infrared range is lower. Interestingly, here the evolution trend from e to h is similar to that in Fig. 4 from b to e for GaAs. The phenomenon may be on account of the resembling spatial frequency components of the treated areas on the two materials, which would induce similar spectral response. In addition, the further increase of laser fluence will bring about the rise of the specular reflectance (the result not shown here), especially for the short wavelength range, which indicates extravagant laser fluence may go against superior anti-specular-reflection performance, resembling the case of Si.

 figure: Fig. 10

Fig. 10 The relative specular reflectances (Rr) of the treated areas on brass fabricated by 800-nm fs laser in various fluence conditions as a function of laser wavelength (λ). Here curves a to h correspond to areas (a) to (h) of Fig. 8, respectively. The dashed line represents the reference spectrum of the untreated surface that is always equal to 1.

Download Full Size | PDF

4. Discussion

From the above experimental results for the morphological characteristics of fs laser-induced micro/nanostructures on GaAs, Si, and brass, we can see a significant characteristic for the evolution of the formed structures: as laser fluence increases, the feature sizes of the structures will enlarge following certain specific rules. Above all, in a large laser fluence range, the spatial frequencies of different kinds of structures transform non-continuously following a specific law – the frequency divides by 2 serially, which appears to be a sequence of 2f, f, f/2, f/4, and f/8. Secondly, for a certain kind of structure, as laser fluence increases, the feature size will also increase, and gradually approaches an extreme value – a typical example is NSRs, whose period would get close to laser wavelength and saturate with laser fluence increasing. The observation can be explained neatly by the calculation results of Ref. 36. As shown in Fig. 4 of Ref. 36, the period of NSR, which is dependent on the excited carrier density in the conduction band generated by laser irradiation that is monotonously related to the laser fluence, exhibits a dip close to the critical electron density for plasma resonance. In detail, in the “right hand side” of the dip, the NSR period is approaching the laser wavelength, whose characteristic is consistent with the observation of the increase of NSR periods with fluence in our experiments. Actually, the origin of the phenomenon is similar to that of the Gaussian field distribution effect that make NSR period decrease from the center to the periphery of the ablated crater as proposed in Ref. 35. By contract, for metals this phenomenon is not obvious, somewhat resembling that shown in Ref. 35 ― as pulse number increases the shift of the NSR period in conductors is far smaller than in semiconductors and dielectrics. It is because for metals the laser irradiation only redistributes the carriers, but not has a major impact on the carrier density due to the absence of bandgap in metals. Consequently, for increasing fluences in metals the NSR period that rests upon the carrier density [35,36] is not shifted evidently. Thirdly, for each kind of the structures, there is an appropriate laser fluence range for it to form. For example, DSRs tend to appear with laser fluence approaching the ablation threshold, whereas NSRs are apt to occur with laser fluence higher than ablation threshold, and have a wide laser fluence range for existing. Outside the appropriate laser fluence range, the structure would be blanking. Fourthly, the surface morphology of the treated surface is determined by the type of the structure. As the spatial frequencies of the structures transform non-continuously following the sequence of 2f, f, f/2, f/4, and f/8, the morphology of the treated surface exhibits a series of very different appearances, accordingly.

For the morphological characteristics of the treated surface described above, the most amazing issue is the appearance of the sequence of 2f, f, f/2, f/4, and f/8 for the spatial frequencies of the laser-induced structures. Now we would propose the mechanisms for such a phenomenon. First of all, towards the 2f component, it is somewhat different from other component, and requires a separate treatment. As the results shown on GaAs, actually the 2f component can divide into two parts: the DSR component (the spatial frequency is slightly higher than 2f) and the second harmonic of NSRs (equal to 2f). In our opinion, the DSRs may result from the standing wave of SPs [26] or the splitting of NSRs due to the loss of grating-assisted SP-laser coupling, while the appearance of the second harmonic of NSRs mean that the profile of NSRs markedly deviates from ideal sine envelope, which is propitious to the grating-assisted coupling between SPs and laser light [35]. And then, the f, f/2, f/4, and f/8 components can be ascribed to 1, 2, 4, and 8 order grating coupling between surface scattering wave (include SPs) and laser light. The phase-matching condition for the grating coupling can be expressed as

mG=kiks
where G is the grating vector (|G| = 2π/Λ, and Λ is the grating period), k i is the wave vector component of incident laser in the interface, and k s is the wave vector of surface scattering wave (include SPs). In Ref. 35, we had discussed the feedback mechanisms of NSRs – grating-assisted SP-laser coupling, which is a special case of Eq. (2) – let m = 1, that is the 1 order grating coupling. In such a case with normal incidence of laser, we can get a grating period Λ1 (corresponding to the spatial frequency f) close to laser wavelength. As a matter of fact, in the case that the values of k i and k s are fixed, if we take into account the high order grating coupling, such as the cases of m = 2, m = 4, and m = 8, we would get the grating periods of Λ2 = 2Λ1, Λ4 = 2Λ2 = 4Λ1, and Λ8 = 2Λ4 = 8Λ1, which has the spatial frequency of f/2, f/4, and f/8, respectively. That is, the f, f/2, f/4, and f/8 component of surface structures corresponds to 1, 2, 4, and 8 order grating coupling. Then, one would ask why the 3-order grating coupling cannot be observed. We consider that the missing of the 3-order grating is due to the feedback mechanism: the formation of such high order components is a multiple-pulse ablation process, i.e., it is a feedback process, and the gratings of different order form according to priority. In detail, the formation of 2-order grating should base on the formation of 1-order grating, and the formation of 4-order grating should base on the formation of 2-order grating. That is to say the instability of 1-order grating would induce 2-order grating, and the instability of 2-order grating would induce the 4-order grating, and so on. Thus, the 3-order grating has a very small chance to occur. In fact, in certain special cases, such as on brass, the 3-order grating may exist with a very low intensity in a suitable narrow fluence range, which is hard to be observed and is prone to fade away. In other words, such kinds of coupling order are not conducive to the feedback.

From the above experimental results, we can see that for the fs laser-induced micro/nanostructures, the optical characteristics have the following distinctive features. Firstly, there is obvious relevance between the surface morphology (two-dimensional topology) and the spectral characteristic of the treated surface. The feature size of the structures determines the position of spectral response. For example, as shown by the results of GaAs, DSRs tend to take effect on the purple band (in this spectral range the finely-rippled surface is provided with a low specular reflectance), and have a little impact on the light with a longer wavelength, such as the red light. Whereas, the coarse ripples (include the frequency components of f, f/2, f/4, f/8, etc.) have an anti-specular-reflection effect in a wider spectral range. In addition, the disordering of the formed micro/nanostructures is help for widening the spectral range of optical response. Secondly, the specular reflectance of the treated surface is directly related to the surface roughness (the amplitude of the relief). As mentioned above, the spectral range of optical response is mainly determined by the two-dimensional topology of the treated surface. Further, with regard to the surfaces exhibited similar morphologies, the surface roughness determines the specular reflectance of such a surface. That is, more rough (larger fluctuation in the direction perpendicular to sample surface) is the surface, more obvious is the anti-specular-reflection effect. This is because the deep grooves, holes, and cavities have a strong confinement and capture effect for the incident light. As a matter of fact, the excellent anti-reflection performance for the pointed-cone structures of “black Si” just stems from such an effect. Towards the structures produced in our experiments, the deep grooves or holes produced with appropriate laser fluence is of benefit to the superior anti-specular-reflection effect of the treated surface. Thirdly, composite structure with different levels is beneficial to the anti-specular-reflection performance of the treated surface, especially for the bandwidth of the anti-specular-reflection characteristic. As our results shown, the treated surface that is provided with an excellent anti-specular-reflection performance is always composed of structures in a large range of spatial scales, such as the spatial frequency components of f, f/2, f/4, f/8, etc., which would have a spectral respose in a wide spectrum of light.

It should be noted that in the study we have carried out the spectral measurements that can determine the specular reflectance rather than the actual anti-reflection effect for the treated surface. Because for grating with period larger than the wavelength of incident light, in addition to the 0-order (specular) scattering, the scattering of some higher orders may also occur and increase the actual reflectance. Besides, the random nanostructures can scatter the incident light into all directions and also increase the actual reflectance. In other words, a low specular reflectance does not necessarily correspond to the strong anti-reflection effect. As a matter of fact, the low specular reflectance is a necessary but not sufficient condition for the strong anti-reflection effect. It means that our study on the specular reflectance of the treated surface can provide a useful guide rather than a deterministic criterion for preparing structures with a superior anti-reflective performance. In future, a further study on the reflectance of treated surface with larger area via utilizing the integrating sphere would provide a more reliable basis for the investigation of the exact anti-reflection effect.

5. Conclusion

On GaAs, Si, and brass, we have systematically studies the morphological and optical characteristics of the large-area micro/nanostructures produced by fs laser irradiation as well as the dependency of the optical characteristics and the morphological characteristics. Our experimental studies based on surface morphological analysis indicate that the morphology of the micro/nanostructured surface produced fs laser ablation is mainly determined by laser fluence. In the irradiation condition of laser fluence near ablation threshold, the structured surface is constituted by the composite configuration – for semiconductors mixed by NSRs and DSRs and for metal mixed by hole-chains and NSRs. As laser fluence increases, NSRs dominate the treated surface. When laser fluence reaches a certain higher range, the surface morphology turns to be a complex configuration, which composes of NSRs, horizontal textures, deep holes and channels, etc. On the whole, with the increase of laser fluence, the feature scale of the micro/nanostructures enlarges, the grooves deepen, and the roughness largens. Furthermore, the FFT analysis demonstrates an interesting phenomenon: as laser fluence increases, the spatial frequency of the micro/nanostructures changes following a specific law – the frequency appears to be a sequence of 2f, f, f/2, f/4, and f/8. In our opinion, the new frequency components of f/2, f/4, and f/8 originate in the 2-order, 4-order, and 8-order grating coupling, whereas the 2f component comes of DSRs and the second harmonic of NSRs. The revealing of such a law is significant for us to understand the evolving mechanisms of a variety of laser-induced micro/nanostructures in different scales.

On the orther hand, our spectral results indicate that the optical characteristics of the laser induced micro/nanostructures directly relate to the morphological characteristics, such as the feature size of the micro/nanostructures (DSRs, NSRs, the f/2, f/4, and f/8 components), the topology of the surface morphology (ripples, particles, holes, and other patterns), the surface roughness (vertical scale of apertures, for instance the depth of grooves and holes), and the irregular degree of the formed structures (the periodic distribution of a certain kind structure). In general, with the increase of laser fluence that induces the enlargement of the scale size, the randomness, and the roughness of the micro/nanostructures, the anti-specular-reflection effect of the structured surface would be significantly enhanced (a lower specular reflectance and a wider spectrum response). However, if laser fluence is excessive, the anti-specular-reflective effect would be weakened, which is due to the re-deposited materials that would pollute the pre-formed micro/nanostructures. In ideal laser fluence, the near-infrared laser-induced micro/nanostructures can achieve an ultra-low specular reflectance in the visible and near-infrared spectral region, which exhibits an attracting application prospect in the field of utilizing solar energy. In short, the investigation of the dependency between the optical characteristics and the morphological characteristics would provide us with a guide for preparing structures with a more excellent anti-reflective performance.

Acknowledgments

The authors are grateful to Y. F. Liu and X. R. Zeng for their supports in the experiments. This work has been supported by grants from the Shanghai Institute of Optics and Fine Mechanics (SIOM), Chinese Academy of Sciences (CAS).

References and links

1. J. Perrière, É. Millon, and É. Fogarassy, Recent Advances in Laser Processing of Materials (Elsevier, Amsterdam, 2006).

2. T.-H. Her, R. J. Finlay, C. Wu, S. Deliwala, and E. Mazur, “Microstructuring of silicon with femtosecond laser pulses,” Appl. Phys. Lett. 73(12), 1673–1675 (1998). [CrossRef]  

3. J. Bonse, S. Baudach, J. Krüger, W. Kautek, and M. Lenzner, “Femtosecond laser ablation of silicon–modification thresholds and morphology,” Appl. Phys., A Mater. Sci. Process. 74(1), 19–25 (2002). [CrossRef]  

4. M. Y. Shen, C. H. Crouch, J. E. Carey, and E. Mazur, “Femtosecond laser-induced formation of submicrometer spikes on silicon in water,” Appl. Phys. Lett. 85(23), 5694–5696 (2004). [CrossRef]  

5. T.-H. Her, R. J. Finlay, C. Wu, and E. Mazur, “Femtosecond laser-induced formation of spikes on silicon,” Appl. Phys., A Mater. Sci. Process. 70(4), 383–385 (2000). [CrossRef]  

6. Z. Huang, J. E. Carey, M. Liu, X. Guo, E. Mazur, and J. C. Campbell, “Microstructured silicon photodetector,” Appl. Phys. Lett. 89(3), 033506 (2006). [CrossRef]  

7. A. Y. Vorobyev and C. Guo, “Enhanced absorptance of gold following multipulse femtosecond laser ablation,” Phys. Rev. B 72(19), 195422 (2005). [CrossRef]  

8. Y. Yang, J. Yang, C. Liang, and H. Wang, “Ultra-broadband enhanced absorption of metal surfaces structured by femtosecond laser pulses,” Opt. Express 16(15), 11259–11265 (2008). [CrossRef]   [PubMed]  

9. A. Y. Vorobyev and C. Guo, “Colorizing metals with femtosecond laser pulses,” Appl. Phys. Lett. 92(4), 041914 (2008). [CrossRef]  

10. A. Y. Vorobyev, V. S. Makin, and C. L. Guo, “Brighter light sources from black metal: significant increase in emission efficiency of incandescent light sources,” Phys. Rev. Lett. 102(23), 234301 (2009). [CrossRef]   [PubMed]  

11. J. Bonse, H. Sturm, D. Schmidt, and W. Kautek, “Chemical, morphological and accumulation phenomena in ultrashort-pulse laser ablation of TiN in air,” Appl. Phys., A Mater. Sci. Process. 71(6), 657–665 (2000). [CrossRef]  

12. J. Reif, F. Costache, M. Henyk, and S. V. Pandelov, “Ripples revisited: nonclassical morphology at the bottom of femtosecond laser ablation craters in transparent dielectrics,” Appl. Surf. Sci. 197–198, 891–895 (2002). [CrossRef]  

13. A. Borowiec and H. K. Haugen, “Subwavelength ripple formation on the surfaces of compound semiconductors irradiated with femtosecond laser pulses,” Appl. Phys. Lett. 82(25), 4462–4464 (2003). [CrossRef]  

14. Q. Wu, Y. Ma, R. Fang, Y. Liao, Q. Yu, X. Chen, and K. Wang, “Femtosecond laser-induced periodic surface structure on diamond film,” Appl. Phys. Lett. 82(11), 1703–1705 (2003). [CrossRef]  

15. Y. Shimotsuma, P. G. Kazansky, J. Qiu, and K. Hirao, “Self-organized nanogratings in glass irradiated by ultrashort light pulses,” Phys. Rev. Lett. 91(24), 247405 (2003). [CrossRef]   [PubMed]  

16. F. Costache, S. Kouteva-Arguirova, and J. Reif, “Sub–damage–threshold femtosecond laser ablation from crystalline Si: surface nanostructures and phase transformation,” Appl. Phys., A Mater. Sci. Process. 79(4-6), 1429 (2004).

17. J. Bonse, A. M. Munz, and H. Sturm, “Structure formation on the surface of indium phosphide irradiated by femtosecond laser pulses,” J. Appl. Phys. 97(1), 013538 (2005). [CrossRef]  

18. T. Q. Jia, H. X. Chen, M. Huang, F. L. Zhao, J. R. Qiu, R. X. Li, Z. Z. Xu, X. K. He, J. Zhang, and H. Kuroda, “Formation of nanogratings on the surface of a ZnSe crystal irradiated by femtosecond laser pulses,” Phys. Rev. B 72(12), 125429 (2005). [CrossRef]  

19. V. R. Bhardwaj, E. Simova, P. P. Rajeev, C. Hnatovsky, R. S. Taylor, D. M. Rayner, and P. B. Corkum, “Optically produced arrays of planar nanostructures inside fused silica,” Phys. Rev. Lett. 96(5), 057404 (2006). [CrossRef]   [PubMed]  

20. O. Varlamova, F. Costache, J. Reif, and M. Bestehorn, “Self-organized pattern formation upon femtosecond laser ablation by circular polarized light,” Appl. Surf. Sci. 252(13), 4702–4706 (2006). [CrossRef]  

21. E. M. Hsu, T. H. R. Crawford, H. F. Tiedje, and H. K. Haugen, “Periodic surface structures on gallium phosphide after irradiation with 150 fs-7 ns laser pulses at 800 nm,” Appl. Phys. Lett. 91(11), 111102 (2007). [CrossRef]  

22. T. Tomita, K. Kinoshita, S. Matsuo, and S. Hashimoto, “Effect of surface roughening on femtosecond laser-induced ripple structures,” Appl. Phys. Lett. 90(15), 153115 (2007). [CrossRef]  

23. M. Huang, F. L. Zhao, T. Q. Jia, Y. Cheng, N. S. Xu, and Z. Z. Xu, “A uniform 290 nm periodic square structure on ZnO fabricated by two-beam femtosecond laser ablation,” Nanotechnology 18(50), 505301 (2007). [CrossRef]  

24. M. Huang, F. L. Zhao, Y. Cheng, N. S. Xu, and Z. Z. Xu, “Large area uniform nanostructures fabricated by direct femtosecond laser ablation,” Opt. Express 16(23), 19354–19365 (2008). [CrossRef]  

25. D. Dufft, A. Rosenfeld, S. K. Das, R. Grunwald, and J. Bonse, “Femtosecond laser-induced periodic surface structures revisited: A comparative study on ZnO,” J. Appl. Phys. 105(3), 034908 (2009). [CrossRef]  

26. M. Huang, F. L. Zhao, Y. Cheng, N. S. Xu, and Z. Z. Xu, “Mechanisms of ultrafast laser-induced deep-subwavelength gratings on graphite and diamond,” Phys. Rev. B 79(12), 125436 (2009). [CrossRef]  

27. M. Birnbaum, “Semiconductor surface damage produced by ruby lasers,” J. Appl. Phys. 36(11), 3688 (1965). [CrossRef]  

28. F. Keilmann and Y. H. Bai, “Periodic surface structures frozen into CO2 laser-melted quartz,” Appl. Phys., A Mater. Sci. Process. 29(1), 9–18 (1982). [CrossRef]  

29. J. E. Sipe, J. F. Young, J. S. Preston, and H. M. van Driel, “Laser-induced periodic surface structure, I. theory,” Phys. Rev. B 27(2), 1141–1154 (1983). [CrossRef]  

30. A. J. Pedraza, Y. F. Guan, J. D. Fowlkes, and D. A. Smith, “Nanostructures produced by ultraviolet laser irradiation of silicon. I. rippled structures,” J. Vac. Sci. Technol. B 22(6), 2823–2835 (2004). [CrossRef]  

31. J. Bonse, M. Munz, and H. Sturm, “Structure Formation on the Surface of Indium Phosphide Irradiated by Femtosecond Laser Pulses,” J. Appl. Phys. 97(1), 013538 (2005). [CrossRef]  

32. J. Wang and C. Guo, “Formation of extraordinarily uniform periodic structures on metals induced by femtosecond laser pulses,” J. Appl. Phys. 100(2), 023511 (2006). [CrossRef]  

33. T. Tomita, Y. Fukumori, K. Kinoshita, S. Matsuo, and S. Hashimoto, “Observation of laser-induced surface waves on flat silicon surface,” Appl. Phys. Lett. 92(1), 013104 (2008). [CrossRef]  

34. G. A. Martsinovsky, G. D. Shandybina, Yu. S. Dement’eva, R. V. Dyukin, S. V. Zabotnov, L. A. Golovan’, and P. K. Kashkarov, “Generation of surface electromagnetic waves in semiconductors under the action of femtosecond laser pulses,” Semiconductors 43(10), 1298–1304 (2009). [CrossRef]  

35. M. Huang, F. L. Zhao, Y. Cheng, N. S. Xu, and Z. Z. Xu, “Origin of laser-induced near-subwavelength ripples: interference between surface plasmons and incident laser,” ACS Nano 3(12), 4062–4070 (2009). [CrossRef]   [PubMed]  

36. J. Bonse, A. Rosenfeld, and J. Krüger, “On the role of surface plasmon polaritons in the formation of laser-induced periodic surface structures upon irradiation of silicon by femtosecond-laser pulses,” J. Appl. Phys. 106(10), 104910 (2009). [CrossRef]  

37. E. V. Golosov, V. I. Emel’yanov, A. A. Ionin, Yu. R. Kolobov, S. I. Kudryashov, A. E. Ligachev, Yu. N. Novoselov, L. V. Seleznev, and D. V. Sinitsyn, “Femtosecond laser writing of subwave one-dimensional quasiperiodic nanostructures on a titanium Surface,” JETP Lett. 90(2), 107–110 (2009). [CrossRef]  

38. S. Sakabe, M. Hashida, S. Tokita, S. Namba, and K. Okamuro, “Mechanism for self-formation of periodic grating structures on a metal surface by a femtosecond laser pulse,” Phys. Rev. B 79(3), 033409 (2009). [CrossRef]  

39. Y. Han and S. Qu, “Uniform self-organized grating fabricated by single femtosecond laser on dense flint (ZF6) glass,” Appl. Phys., A Mater. Sci. Process. 98(1), 167–170 (2010). [CrossRef]  

40. X. Jia, T. Q. Jia, Y. Zhang, P. X. Xiong, D. H. Feng, Z. R. Sun, J. R. Qiu, and Z. Z. Xu, “Periodic nanoripples in the surface and subsurface layers in ZnO irradiated by femtosecond laser pulses,” Opt. Lett. 35(8), 1248–1250 (2010). [CrossRef]   [PubMed]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (10)

Fig. 1
Fig. 1 A schematic view of the experimental setup for the spectral method.
Fig. 2
Fig. 2 The experimental results of the micro/nanostructures on GaAs produced by 1280-nm fs laser of linear polarization with different fluences. The laser fluence for area (a), (b), (c), (d) and (e) is 0.09, 0.11, 0.15, 0.19 and 0.25J/cm2, respectively. From the following processing parameters: the scanning velocity of 125 μm/s, the scanning interval of 8 μm, the pulse repetition frequency of 1 kHz, and the focal spot size of 36 μm, the average number of laser pulses per spot can be calculated to be about 1000 pulses. The sub-figures of left column are the OM images of the 400 × 400-μm2 treated areas taken under the same condition as spectral measurement. The sub-figures of the middle two columns are SEM images of the treated areas with different magnification factors. In (a) the double-headed arrow indicates the direction of laser polarization (in (b) to (e) the polarization directions are the same as in (a)). The sub-figures of the right column are the FFT images of the corresponding SEM images, which transform from the SEM images with small magnification factor except (a) and rescale to have the same spatial frequency scale.
Fig. 3
Fig. 3 The real space representation of the spatial frequency components of the treated GaAs surfaces in the direction of ripple wave vector.
Fig. 4
Fig. 4 The relative specular reflectances (Rr ) of the treated areas on GaAs fabricated by 1280-nm fs laser in various fluence conditions as a function of laser wavelength (λ). Here curves a to e correspond to areas (a) to (e) in Fig. 2, respectively. The dashed line represents the reference spectrum of the untreated surface that is always equal to 1.
Fig. 5
Fig. 5 The experimental results of the micro/nanostructures on Si produced by 1280-nm fs laser of linear polarization with different fluences. The laser fluence for area (a), (b), (c), (d), (e), (f), and (g) is 0.16, 0.20, 0.26, 0.32, 0.40, 0.48, and 0.60 J/cm2, respectively. From the following processing parameters: the scanning velocity of 125 μm/s, the scanning interval of 8 μm, the pulse repetition frequency of 1 kHz, and the focal spot size of 36 μm, the average number of laser pulses per spot can be calculated to be about 1000 pulses. The sub-figures of left column are the OM images of the 400 × 400-μm2 treated areas taken under the same condition as spectral measurement. The sub-figures of the middle two columns are SEM images of the treated areas with different magnification factors. In (a) the double-headed arrow indicates the direction of laser polarization (in (b) to (g) the polarization directions are the same as in (a)). The sub-figures of the right column are the FFT images of the corresponding SEM images, which transform from the SEM images with small magnification factor except (a) and rescale to have the same spatial frequency scale for ease of comparison.
Fig. 6
Fig. 6 The real space representation of the spatial frequency components of the treated Si surfaces in the direction of ripple wave vector.
Fig. 7
Fig. 7 The relative specular reflectances (Rr ) of the treated areas on Si fabricated by 1280-nm fs laser in various fluence conditions as a function of laser wavelength (λ). Here curves a to g correspond to areas (a) to (g) of Fig. 5, respectively. The dashed line represents the reference spectrum of the untreated surface that is always equal to 1.
Fig. 8
Fig. 8 The experimental results of the micro/nanostructures on brass produced by 800-nm fs laser of linear polarization with different fluences. The laser fluence for area (a), (b), (c), (d), (e), (f), (g), and (h) is 0.36, 0.45, 0.52, 0.60, 0.68, 0.82, 1.06, and 1.39 J/cm2, respectively. From the following processing parameters: the scanning velocity of 125 μm/s, the scanning interval of 8 μm, the pulse repetition frequency of 1 kHz, and the focal spot size of 45 μm, the average number of laser pulses per spot can be calculated to be about 1600 pulses. The sub-figures of left column are the OM images of the 200 × 200-μm2 treated areas taken under the same condition as spectral measurement. The sub-figures of the middle two columns are SEM images of the treated areas with different magnification factors. In (a) the double-headed arrow indicates the direction of laser polarization (in (b) to (h) the polarization directions are the same as in (a)). The sub-figures of the right column are the FFT images of the corresponding SEM images, which transform from the SEM images with small magnification factor and rescale to have the same spatial frequency scale for ease of comparison. The frequency components in the direction perpendicular to ripple wave vector are shown in the top right corner of the corresponding FFT images (the images have been rotated 90° for ease of comparison, and some of the color scales of the images have been reset in order to highlight the perpendicular frequency components).
Fig. 9
Fig. 9 The real space representation of the spatial frequency components of the treated brass surfaces in the direction of ripple wave vector.
Fig. 10
Fig. 10 The relative specular reflectances (Rr ) of the treated areas on brass fabricated by 800-nm fs laser in various fluence conditions as a function of laser wavelength (λ). Here curves a to h correspond to areas (a) to (h) of Fig. 8, respectively. The dashed line represents the reference spectrum of the untreated surface that is always equal to 1.

Equations (2)

Equations on this page are rendered with MathJax. Learn more.

R r ( λ ) = I t ( λ ) I i ( λ )
m G = k i k s
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.