Open Access
Add to BibSonomyAbstract
We report on the fabrication of shape-controlled microchannels in fused silica by femtosecond laser irradiation at 600 kHz repetition rate followed by chemical etching. The shape control is achieved by suitable wobbling of the glass substrate during the irradiation process. Cylindrical microchannels with uniform cross-sections are demonstrated with an unprecedented length of 4 mm. Some applications are also addressed: connection of two microchannels with a smaller one, 3D microchannel adapter and fabrication of O-grooves for easy fiber-to-waveguide coupling.
©2009 Optical Society of America
1. Introduction
Microfabrication techniques play a key role in modern science and technology and paved the way to the emergence of many interdisciplinary applications [1]. In particular, the advent of microfluidics, through the concept of lab-on-a-chip, is revolutionizing fields such as biology and chemistry [2,3]. Microfluidic channels, which form the backbone of such integrated devices, are conventionally manufactured by the combination of photolithography and chemical-etching [1], but this approach is primarily limited to the fabrication of two-dimensional patterns on the surface. In order to create a true 3D structure several layers of glass substrates need to be processed and fused together. This makes the fabrication procedure complex and multi-stepped. A relatively new microfabrication technique is Femtosecond Laser Irradiation followed by Chemical Etching (FLICE), which has gained interest in the last few years, due to its simplicity and ability to produce buried 3D structures with high aspect ratios [4–9]. As FLICE is a maskless technology, it enables rapid prototyping of new complex device schemes [10,11]. Due to its simplicity and flexibility it can complement existing technologies with ease. Moreover, since the same femtosecond laser can be used to produce low loss optical waveguides [12], femtosecond laser based technologies could become a one-stop solution for fabrication of microfluidic channels integrated with optical circuits [13].
The FLICE technique on fused silica samples consists of two steps: 1) irradiation of the sample with focussed femtosecond laser pulses; 2) etching of the laser modified zone by a hydrofluoric acid (HF) solution in water. When fused silica is irradiated, the modifications induced by the femtosecond laser pulses can be classified into three categories depending on the laser processing conditions [6]: a) for a low fluence, a smooth modification is achieved, resulting mainly in a positive refractive index change with a very weak selectivity in etching; b) for a moderate fluence, sub-wavelength nanocracks are produced, yielding a high etching selectivity of the irradiated volume with respect to the pristine one (up to two orders of magnitude); c) for high fluence, a disruptive modification is obtained with the creation of voids and microexplosions. In particular, regime a) is typically suited for waveguide fabrication, while regime b) is the one employed in the first step of the FLICE technique for microchannel production. Regime c) can be used for direct laser ablation and it will be not be considered in this paper.
The increase in HF etching rate of fused silica has been correlated to the decrease of the Si-O-Si bond angle induced by the hydrostatic pressure or compressive stress created in the irradiated region [5,14]. This explanation is particularly suited for regime a), where the formation of waveguides is due to a local increase of the density of the material; in fact, the etching rate is shown to be proportional to the increase of the refractive index [15]. However, in this regime a very modest increase in the etching rate is obtained after irradiation. In case (b), a much higher selectivity in etching of the irradiated regions is achieved. This can be attributed to the formation of self-ordered nanocrack structures, perpendicular to the laser polarisation direction [16]. During the etching process, the nanocracks act as channels for the diffusion of the acid deeper into the fused silica. Hence the etching process is a combination of two simultaneous phenomena, the diffusion of the acid along the irradiated region and the etching of the fused silica that gets in contact with such acid. Therefore, the etching process should not be thought as the acid carving its way by progressively removing the irradiated material but as a fast diffusion of acid in the irradiated region that causes an etching of material along the diffusion path. This is the reason why the microchannels produced in regime b) yield much higher aspect ratios than regime a), where the nanocracks are absent. As a further proof, if a polarization parallel to the sample translation direction is used in the irradiation process, a negligible etching rate is obtained since the nanocracks are transversal to the channel axis, thus blocking acid diffusion [6]. The buried microchannels fabricated by this technology are typically limited in length. A self terminating etching process is observed, probably due to the exhaustion of the HF acid in the etched microchannel and to the difficulty in refreshing it. The longest dead-end channels fabricated up to now are 1.8 mm-long [13]. This technology, with a suitable optimization of the etching conditions, is able to provide rather smooth surfaces with rms roughness in the 20-50 nm range [9].
A known feature of directly buried microchannels fabricated by the FLICE technique is their conical shape (Fig. 1 ) [6,7], which becomes more pronounced for longer channels. This conical shape formation can be understood according to the above explanation of the etching process. The sample edge (indicated by a ‘ + ’ symbol, Fig. 1) is exposed to the HF acid for longer time with respect to the buried regions where the etching is delayed by the diffusion time that the acid needs to access them. This results in conical shaped microchannels with larger radius at the channel entrance as compared to the buried end. This is a specific feature of microchannels fabricated by FLICE and is typically not wanted in applications where uniform channels are desired. This effect can be reduced by etching the microchannel from opposite sides. In this way, depending on the extent of overlap between the two etched cones, channels with different shapes can be obtained with a more uniform cross-section the shorter the final microchannel [13]. The apex angle α of the cone (Fig. 1(e)) depends on the etching rate of unirradiated fused silica and diffusion rate of the acid in the irradiated channel. It is worth noting that there is a trade-off between achieving microchannels with high aspect ratio and with greater length. Indeed, a low HF concentration [5,6] decreases the etch rate and results in a better aspect ratio, but shorter channels are achieved; in fact, a lower quantity of HF is present in the channel and self termination of the etching process happens earlier. On the other hand, high HF concentration [7] results in a lower aspect ratio, due to a faster lateral etching, but delays the self termination of the etching process and increases the channel length.
Fig. 1 (a) Microscope image of a modified line by a 1 kHz repetition rate laser; (b) corresponding microchannel obtained after chemical etching. (c) modified line by a 600 kHz repetition rate laser; (d) corresponding microchannel obtained after chemical etching. Symbol ’ + ’ indicates the sample edge. (e) Schematic diagram of an etched cone.
In this paper we demonstrate a method to compensate for the conical shape of the microchannels in a single side etching process with high HF concentration. This method consists in irradiating a reverse cone with respect to the one normally obtained with the FLICE technique. With this Shape Controlled FLICE technique (SC-FLICE) dead-end microchannels with uniform and cylindrical shape are obtained. This opens the way to several applications for this technology and in addition it allows one to produce a through microchannel with unprecedented length and uniform cross-section.
2. Experimental, principle and feasibility of the shape control of microchannels
2.1 High repetition rate laser irradiation and chemical etching
All previous studies on microchannel fabrication by the FLICE technique are based on amplified Ti:sapphire lasers, with repetition rate ranging from 1 to 200 kHz [4–9]. Irradiation of fused silica with those lasers results in processing speeds of tens to hundreds of µm/s. Figure 1(a) shows an irradiated line produced by a 1 kHz Ti:sapphire laser with 4 µJ pulse energy, focused by a 0.6 NA microscope objective and with a 10 µm/s translation speed [7]. Subsequent etching in an ultrasonic bath of 20% solution of HF in water for 3 hours provided the microchannel shown in Fig. 1(b). Recently, diode-pumped femtosecond ytterbium lasers, both in bulk and fiber formats, have proved particularly suitable for optical waveguide writing [17,18]. These lasers have several advantages with respect to Ti:sapphire and in particular are much more compact and suited for an industrial environment. In this work we apply such a laser also to the fabrication of microchannels by the FLICE technique. We use the second harmonic (515 nm) of a cavity-dumped Yb:KYW oscillator providing 350-fs laser pulses at repetition rates up to 1 MHz [19]. The irradiated line reported in Fig. 1(c) was produced using such laser with 150 nJ pulse energy, polarization orthogonal to the irradiated path, 600 kHz repetition rate, 0.6 NA microscope objective (focusing to a beam diameter of about 1 µm) and 1cm/s translation speed. The use of 600 kHz instead of 1 MHz was adopted in order to have a higher energy per pulse (of about 1.5 µJ at 1030 nm) which enabled a wider range of available pulse energies at the second harmonic. Applying the same etching conditions to this irradiated path an almost equal microchannel is obtained (Fig. 1(d)). As a matter of fact, the use of a different laser and the higher repetition rate provided microchannels with the same quality, but with an increase in the processing speed of up to three orders of magnitude. It is nevertheless true that the second step of the FLICE technique still requires a few hours etching, but all the structures are etched out in parallel, while as the complexity of the microchannel network increases the irradiation time may rapidly become the most time consuming step.
All the results presented in this paper are obtained employing a Yb:KYW oscillator with the above irradiation parameters followed by chemical etching in an ultrasonic bath with a 20% HF concentration in water and at a temperature of 35 ± 1 °C.
2.2 Conical shape compensation
The standard FLICE technology can provide microchannels with conical shape, which may be needed in some specific applications, but it is typically desirable to have microchannels cylindrical in shape. In order to obtain channels with uniform cross section, we modified the first step of the FLICE technique by irradiating more complex structures than just a straight line. The novel idea thus is to irradiate a conical spiral, as that represented in Fig. 2(a) , forming a laser-modified conical surface, which is complementary to the cone that is created when etching a straight line. Additionally a straight line forming the axis of the cone is also irradiated, which helps in removing the inner volume of the cone. The laser polarization is always orthogonal to the axis of the cone. The expected result is shown in Fig. 2(b), where the blue cone is that of typical channel etched starting from a straight line, while the red line represents the conical irradiation, which would compensate for this effect, and the dotted line shows the cylindrical microchannel that should be obtained with the SC-FLICE.
Fig. 2 (a) Schematic diagram of a conical spiral inscribed into the substrate; (b) representation of a conical microchannel (blue), of the compensating conical spiral (red) and of the final cylindrical microchannel (grey).
When a straight line is etched, the resulting void cone can be characterized by the length L, the radius r and the angle α, given by α = tan−1 (r / L) (see Fig. 1(e)). For a given set of etching parameters (20% HF acid in aqueous solution, 3 hours of etching in an ultrasonic bath at a constant temperature of 35°C) the values obtained for L and r are typically 1.5 mm and 52 µm respectively. The value of α was calculated to be 1.9°. By irradiating a conical spiral with the reverse shape, but with the same L and r as above (Fig. 3(a) ), the etching yielded a microchannel with uniform diameter of 100 µm throughout the entire channel length of 1.5 mm (Fig. 3(b)). By keeping α constant at 1.9°, longer or shorter microchannels, again with uniform cross section, can be obtained by varying the length of the irradiated cone and consequently the etching time
Fig. 3 Microscope image of (a) conical spiral inscribed in the glass and (b) the etched microchannel .
The periodicity Λ of the spirals determines the extent of overlap between two successive laser irradiated arcs of the conical spiral. When there is an insufficient overlap, the acid encounters unexposed regions, which will slow down the etching process. Given our 0.6 NA focusing objective, we find that for Λ = 5 µm the etching rate is already three times slower than the optimal case of conic spirals with Λ = 2 µm, i.e. 150 µm/hour and 500 µm/hour respectively. It is also worth noting that the nanocracks, that are formed in the irradiation of successive overlapping arcs, are naturally self-aligned. This is indeed necessary to provide nanopaths for the acid diffusion in the irradiated region, as discussed in the Introduction. The self-alignment of subsequently written nanocracks, even with a few seconds delay as in the conical spiral irradiation, is quite surprising but consistent with the observation in Ref. 16, where self-alignment of the nanocracks created by subsequent pulses at a 1 kHz repetition rate is reported. In fact, the 1-ms delay between the two pulses is already sufficient for the material to cool down completely and therefore the same mechanism for the nanocracks self-alignment will take place also for much longer delays.
In order to irradiate conical spirals in the volume of the material, the laser focus is kept fixed while the sample is moved by a suitable software that drives a 3-axes air-bearing translation stage (Fiberglide 3D, Aerotech). The use of a 600 kHz repetition-rate laser allows the quite high processing speed of 1 cm/s, as reported in par. 2.1. Currently, in the irradiation of conical spirals the tangential speed is limited to 1 mm/s by the translation stages in order to keep a high control of the irradiated shape. Given the conical spiraling in the irradiation process, the effective speed along the Z-axis of the cone is much lower than the tangential one. The effective speed along that axis is not uniform but we calculate a typical average value of 20 µm/s (e.g. the pre-etching irradiation of a 2-mm long channel requires about 2 minutes). This is in the same range as the processing speed with low repetition rate lasers and line irradiation (Fig. 1(a)), but with SC-FLICE a significant enhancement in the control of the microchannel shape is achieved.
2.3 Compensation for the refraction
When a laser beam is focused inside a glass slab of refractive index n by means of a dry microscope objective, the focus position is different from that obtained with the same focusing in air. This well-known effect is due to the refraction of light beams when crossing the air-glass interface. It can be shown that the depth of the focus in the glass slab is equal to n times the depth it would have if only air were present. As a consequence, when a conical spiral is inscribed into the glass, the cross section of the cone will not be circular as the sample motion (Fig. 4 (A1)) but will come out elliptical showing a stretching in the depth axis by a factor n with respect to the lateral axis (Fig. 4(A2)). This will also affect the microchannel cross-section after etching, which will be uniform along the channel but elliptical (Fig. 4(A3)).
Fig. 4 The dotted lines A1 and B1 represent the cross sections of the path along which the sample is moved. A2 and B2 represent the actual cross-sections irradiated in the sample due to refraction at the interface. A3 and B3 are the microscope images of the corresponding microchannel cross-sections.
In order to pre-compensate for this effect, we irradiated a conical spiral with an elliptical cross-section (Fig. 4(B1)); in this way the stretching of the depth axis provides a perfect circle (Fig. 4(B2)) and finally the etched channel will have the desired circular cross-section (Fig. 4(B3)).
3. Applications of the shape controlled microchannels
3.1 Long cylindrical channel
The main drawback of the FLICE technique is the limited length of the microchannels, which is about 1.5-2 mm when etching from a single side [7]. Etching from both sides of the microchannel could in principle provide double the length; as a matter of fact, this is not feasible because the standard conical shape of the etched microchannels requires some overlap to obtain a through channel. The maximum length demonstrated so far is 3 mm but that channel had a reduction factor of 2 in diameter from edge to center [13]. In order to achieve more uniform channels a much higher overlap between the etching from the two sides is required, yielding a maximum length of 2 mm. The use of the SC-FLICE technique presented in this work allows one to double the microchannel length when etching from two sides with respect to a single side, while also achieving a uniform cross section. Indeed, with the present technique almost no overlap is required to obtain through channels and the channel uniformity is achieved irrespectively of the amount of overlap. To obtain the longest microchannel two conical spirals are written back to back, as shown in Fig. 5(a) . The length of each conical spiral is 2 mm and the radius in the middle is r = 70 µm. After etching from both sides in an ultrasonic bath of 20% solution of HF in water for 4.5 hours, a 4 mm long microchannel is achieved, which is to our knowledge the longest channel fabricated with this technology (Fig. 5(b)). The radius of the cross-section is 90 µm ± 5 µm along the whole microchannel.
Fig. 5 Microscope images of (a) the irradiated double spiral and (b) the 4-mm long cylindrical microchannel achieved by etching the above structure.
3.2 Microchannels with access holes
In previous paragraphs all microchannels were produced by etching from the sides of the glass slab. In most applications, however, the microchannel is connected to the external world by access holes on the top surface. With standard technologies the access holes are fabricated in a subsequent step after sealing the surface microchannel with a cover slide. By using the FLICE technique the access holes can be fabricated in a single step together with the microchannel. A non-uniform microchannel with access holes has been fabricated with a 1 kHz laser by irradiating and then etching a U-shaped path [20]. In this work we apply the SC-FLICE technique to produce a uniform and buried microchannel with access holes reaching the substrate surface. The schematic representation of the irradiated path is shown (not to scale) in Fig. 6(a) and the microscope image of the actual structure is reported in Fig. 6(b). The double conical spiral is fabricated at a 500 µm depth; the two vertices of the conical spirals are then connected to the substrate surface by two uniform helixes with a 20 µm radius. After etching these two helixes provide the access holes, with 110-μm radius, connecting the buried and uniform cross-section microchannel to the top surface (Fig. 6(c)). The advantage in fabricating the access holes by irradiating two helixes instead of two lines as in the previous work [20] is that in this way the size of the access holes can be defined by a proper choice of the helix radius. This structure again demonstrates the true 3D capability of the SC-FLICE technique.
Fig. 6 (a) 3D representation of conical spirals with the two helixes (not to scale), Microscope image of (b) the irradiated conical spiral with two helixes (XY view) and (c) the same structure after etching. The openings serve as entrance columns for the microchannel
3.3 Three-dimensional microchannel adapter
The shape control method described in par.2.2 can also be exploited for the tapering of a microchannel. It is worth noting that the SC-FLICE technique has the unique capability of performing a three-dimensional tapering, where the microchannel cross-section changes with respect to both the transverse axes symmetrically. In particular, we demonstrate a microchannel adapter where two uniform microchannels with different diameters are connected through a tapered portion (Fig. 7(a) ).
Fig. 7 (a) Schematic representation of the microchannel adapter in 3D (not to scale). Microscope image of (b) the irradiated structure and (c) of the final microchannel adapter.
To achieve the desired shape, the pattern shown in Fig. 7(b) was irradiated. Two opposite conical spirals are produced, but while the one on the right starts from a point, that on the left starts from a non-zero radius. Since both the conical spirals have the same optimized slope they compensate for the conicity of the etching process and result in uniform channels after etching. However, a wider microchannel is obtained on the left where the conical spiral already starts with a non-zero radius (Fig. 7(c)). In particular, the left channel has a 200-µm diameter with respect to the 100 µm of the one on the right. As the etching further proceeds the region between the two channels is etched out creating a tapered connection between the two uniform channels. As previously mentioned such tapering takes place along both the transversal dimensions, providing a really three-dimensional microchannel adapter. The length and slope of the tapered region can be tailored by modifying the conical spirals design.
3.4 Interconnecting microchannels
The SC-FLICE technique creates uniform channels in three dimensions that can be in turn connected by smaller channels at arbitrary positions. This allows the creation of more complex microsystems based on a 3D network of microchannels. It is however worth noting that the production of this network is still accomplished in just two steps, first the irradiation of the complete structure and then the removal of the irradiated volume by chemical etching.
As an example Fig. 8(a) shows two long cylindrical microchannels connected, at a distance of about 1 mm from the sample edge, by another channel with a smaller radius. In order to achieve this result two conical spirals, according to the details given in par. 2.2, have been irradiated parallel to each other with a distance of 500 µm. A straight line orthogonally connecting the two axes of the conical spirals has been also irradiated. When etching the structure, the two conical spirals provide the two parallel microchannels with a uniform cross-section while the straight line creates the interconnecting channel. Albeit no conical irradiation is performed in the latter channel, its cross-section is quite uniform due to its limited length and to the fact that the etching is performed from both sides [13]. In addition, the size of the connecting channel (30 µm diameter) is much smaller than that of the two main ones, (100 µm). To understand the different sizes of the channels one has to keep in mind that these structures are completely buried and that the etching process always starts from the edge of the sample. Therefore, the interconnecting channel is etched by the acid only after some delay with respect to the two main ones. As an example, in the structure shown in Fig. 8(a) the connecting channel is etched only during the last one of the 3 hours in the etching process. This explains why its diameter is about 1/3 that of the two main channels. According to this discussion, it becomes evident that the size of the connecting channel depends on its distance from the sample edge, with a diameter that gets smaller and smaller as it is placed toward the end of the two main channels. This structure may have interesting sieving applications, with a size selective connection between two channels at different positions.
Fig. 8 (a) Two microchannels connected by a third smaller one. (b) Waveguide centered to the microchannel; the blow-up shows the end of the etched channel and the optical waveguide.
3.5 ‘O’-grooves for fiber to chip plug-in connector
A new research field called optofluidics is rapidly emerging [21]. This requires integration of micro-optics and microfluidics. Femtosecond laser micromachining is a technology particularly suited for fabricating both microchannels and optical waveguides and for their integration [13]. SC-FLICE can fabricate cylindrical wells in the glass (arranged in arrays or matrices) and each well could be addressed by optical waveguides; this could pave the way to integrated optical sensing of biomolecules.
On the other hand, a cylindrical microchannel coupled to an optical waveguides could be exploited as an ‘O’-groove for easy fiber to waveguide coupling. In this case the waveguide should be perfectly centered with respect to the microchannel and should arrive as close as possible to the end of the channel (Fig. 8(b)). The alignment between the two structures is quite straightforward since the optical waveguide writing is performed with the same laser and in the same processing run as the conical spiral irradiation, but with lower pulse energy and processing speed, i.e. 100 nJ and 100 µm/s. Thus the alignment has a submicron resolution. On the other hand, care should be taken in producing the waveguide as close as possible to the microchannel since etching along the waveguide is to be avoided. For this reason a gap of 30 µm between the irradiated cone edge and the waveguide was left, which acts as a buffer region in order to completely etch the microchannel without attacking the waveguide. In addition, a careful calibration of the etching process can reduce the final gap between the waveguide and the channel to a value close to zero, resulting in a 3.5 h etching time.
By matching the inner diameter of the O-groove with the outer diameter of the fiber, the fiber can be easily plugged into the chip without the need of any special alignment techniques. Since the O-grooves can be made longer than 1 mm, the fiber is well inserted into the chip and has little freedom to move, thus precise and stable alignment can be achieved. Figure 9 shows a coupling experiment in a fused silica chip, performed using a 1.5-mm long O-groove, with a 125 µm diameter matching that of the fiber. A 4-µm diameter waveguide is written centered to the microchannel and runs across the 1.5-cm length of the chip. This waveguide supports a single guided mode at 1.55 µm with a 1/e2-diameter of 12 µm. The fiber is inserted into the microchannel, such that the end of the fiber is touching the end face of the microchannel, without any index matching fluid (Fig. 9(b) shows the fiber inserted halfway for better visualization). As the waveguide and the fiber core are centered with each other, the laser light from the inserted fiber couples into the waveguide and propagates to the other end of the chip, where a microscope objective collects and images it onto an infrared viewing card (magnified view in Fig. 9(c)). This demonstrates the use of the O-grooves as fiber to chip plug-in connectors.
Fig. 9 (a) Experimental setup for coupling the laser light from fiber to the waveguide through the O-groove; (b) sample with an O-groove (same dimensions as Fig. 8(b)) and a fiber inserted half way. (c) blow-up of the guided mode near field image.
4. Conclusions
In this paper the shape control functionality is added to the well assessed FLICE technique for fabricating uniform and directly buried microchannels. By irradiating more complex structures as conical spirals it is possible to compensate for the intrinsic conical shape of the etched microchannel. This fabrication process is enabled by the use of a high repetition rate laser (0.6 MHz), which allows one to irradiate a much longer and complicated path in the same time taken by a standard 1kHz laser to irradiate one straight line. Cylindrical microchannels with a 1.5-mm length are produced by single side etching. In addition, uniform microchannels with an unprecedented 4-mm length are demonstrated by double side etching. The SC-FLICE technique allows also the fabrication of more complex structures, such as microchannels with access holes, interconnecting channels, microchannel adapters, and O-grooves for fiber to chip fast connection.
The flexibility and three-dimensional capabilities of the technique make SC-FLICE very attractive for the production of microsystems. In particular, given the possibility to directly write optical waveguides with the same laser system, several applications in optofluidics can be foreseen.
Acknowledgements
The authors acknowledge financial support from the European Union through the project HIBISCUS (contract no. IST-2005-034562) and the CARIPLO Foundation through the project “Materiali a memoria ottica passiva realizzati mediante cristalli liquidi in microstrutture tridimensionali”. KCV acknowledges Italian Ministry of University and Research (MUR) for the scholarship grant.
References and links
1. M. J. Madou, Fundamentals of microfabrication: the science of miniaturization (CRC Press, 2002).
2. G. M. Whitesides, “The origins and the future of microfluidics,” Nature 442(7101), 368–373 (2006). [CrossRef] [PubMed]
3. A. J. DeMello, “Control and detection of chemical reactions in microfluidic systems,” Nature 442(7101), 394–402 (2006). [CrossRef] [PubMed]
4. A. Marcinkevi Ius, S. Juodkazis, M. Watanabe, M. Miwa, S. Matsuo, H. Misawa, and J. Nishii, “Femtosecond laser-assisted three-dimensional microfabrication in silica,” Opt. Lett. 26(5), 277–279 (2001). [CrossRef]
5. Y. Bellouard, A. Said, M. Dugan, and P. Bado, “Fabrication of high-aspect ratio, micro-fluidic channels and tunnels using femtosecond laser pulses and chemical etching,” Opt. Express 12(10), 2120–2129 (2004). [CrossRef] [PubMed]
6. C. Hnatovsky, R. S. Taylor, E. Simova, P. P. Rajeev, D. M. Rayner, V. R. Bhardwaj, and P. B. Corkum, “Fabrication of microchannels in glass using focused femtosecond laser radiation and selective chemical etching,” Appl. Phys., A Mater. Sci. Process. 84(1-2), 47–61 (2006). [CrossRef]
7. V. Maselli, R. Osellame, G. Cerullo, R. Ramponi, P. Laporta, L. Magagnin, and P. L. Cavallotti, “Fabrication of long microchannels with circular cross section using astigmatically shaped femtosecond laser pulses and chemical etching,” Appl. Phys. Lett. 88, 191107 (2006).
8. Y. Cheng, K. Sugioka, K. Midorikawa, M. Masuda, K. Toyoda, M. Kawachi, and K. Shihoyama, “Control of the cross-sectional shape of a hollow microchannel embedded in photostructurable glass by use of a femtosecond laser,” Opt. Lett. 28(1), 55–57 (2003). [CrossRef] [PubMed]
9. Y. Sikorski, C. Rablau, M. Dugan, A. A. Said, P. Bado, and L. G. Beholz, “Fabrication and characterization of microstructures with optical quality surfaces in fused silica glass using femtosecond laser pulses and chemical etching,” Appl. Opt. 45(28), 7519–7523 (2006). [CrossRef] [PubMed]
10. K. Sugioka, Y. Cheng, and K. Midorikawa, “Three-dimensional micromachining of glass using femtosecond laser for lab-on-a-chip device manufacture,” Appl. Phys. A: Mater. Sci. Process 81, 1–10 (2005).
11. Y. Bellouard, A. A. Said, and P. Bado, “Integrating optics and micro-mechanics in a single substrate: a step toward monolithic integration in fused silica,” Opt. Express 13(17), 6635–6644 (2005). [CrossRef] [PubMed]
12. R. R. Gattass and E. Mazur, “Femtosecond laser micromachining in transparent materials,” Nat. Photonics 2(4), 219–225 (2008). [CrossRef]
13. R. Osellame, V. Maselli, R. Martinez Vazquez, R. Ramponi, and G. Cerullo, “Integration of optical waveguides and microfluidic channels both fabricated by femtosecond laser irradiation,” Appl. Phys. Lett. 90(23), 231118 (2007). [CrossRef]
14. A. Agarwal and M. Tomozawa, “Correlation of silica glass properties with the infrared spectra,” J. Non-Cryst. Solids 209(1-2), 166–174 (1997). [CrossRef]
15. R. S. Taylor, C. Hnatovsky, E. Simova, D. M. Rayner, M. Mehandale, V. R. Bhardwaj, and P. B. Corkum, “Ultra-high resolution index of refraction profiles of femtosecond laser modified silica structures,” Opt. Express 11, 775–780 (2003). [CrossRef] [PubMed]
16. R. Taylor, C. Hnatovsky, and E. Simova, “Applications of femtosecond laser induced self-organized planar nanocracks inside fused silica glass,” Laser Photon. Rev. 2(1-2), 26–46 (2008). [CrossRef]
17. R. Osellame, N. Chiodo, G. della Valle, S. Taccheo, R. Ramponi, G. Cerullo, A. Killi, U. Morgner, M. Lederer, and D. Kopf, “Optical waveguide writing with a diode-pumped femtosecond oscillator,” Opt. Lett. 29(16), 1900–1902 (2004). [CrossRef] [PubMed]
18. S. M. Eaton, H. Zhang, M. L. Ng, J. Li, W.-J. Chen, S. Ho, and P. R. Herman, “Transition from thermal diffusion to heat accumulation in high repetition rate femtosecond laser writing of buried optical waveguides,” Opt. Express 16(13), 9443–9458 (2008). [CrossRef] [PubMed]
19. A. Killi, A. Steinmann, J. Dörring, U. Morgner, M. J. Lederer, D. Kopf, and C. Fallnich, “High-peak-power pulses from a cavity-dumped Yb:KY(WO4)2 oscillator,” Opt. Lett. 30(14), 1891–1893 (2005). [CrossRef] [PubMed]
20. K. C. Vishnubhatla, J. Clark, G. Lanzani, R. Ramponi, R. Osellame, and T. Virgili, “Ultrafast optofluidic gain switch based on conjugated polymer in femtosecond laser fabricated microchannels,” Appl. Phys. Lett. 94(4), 041123 (2009). [CrossRef]
21. D. Psaltis, S. R. Quake, and C. Yang, “Developing optofluidic technology through the fusion of microfluidics and optics,” Nature 442(7101), 381–386 (2006). [CrossRef] [PubMed]
