A technique combining low-coherence reflectometry, laser ablation and microfluidics in a single microstructured fiber is developed. Experimental results demonstrate the possibility to ablate thin aluminum foil samples with fiber-guided Nd:YAG laser light, to collect liquid in the holes of the fiber and to simultaneously monitor the positioning of fiber for ablation and the fluid collection process with low-coherence reflectometry. Potential applications of the technique include minimally invasive retrieval of liquid samples with low contamination risk.
© 2009 OSA
Minimally invasive techniques for the collection of fluids are useful in a number of application areas. In leukaemia diagnosis, for instance, bone marrow aspiration and biopsy involve the collection of marrow for cell analysis. Minimally invasive techniques that allows for fluid retrieval inside the bone offering low contamination risk, the ability to open very small holes and possibly inflicting less pain are of great interest. In geosciences, fluid inclusions in rocks – liquids and gases trapped during rock solidification - are studied to reconstruct the environment prevailing during rock formation and in the search for organic material from the past. One can envisage a microstructured fiber system which allows for minimally invasive collection of fluid with low contamination risk. To this end, the combination of three functions, normally used separately, is required. On one hand, high power laser light can be guided by the fiber to ablate the hard material that contains the fluid of interest [1–3]. No mechanical contact is used, and this reduces the contamination risk significantly. The same fiber can also be used for fluid collection if it is equipped with longitudinal microcapillary holes [4,5]. Finally, positioning and monitoring can benefit from using low-coherence reflectometry [6–9], to determine the distance between the fiber tip and the sample, when the sample has been perforated, and when the liquid inside the sample comes in contact with the fiber end and starts being sucked. The development of such a tool is a long term endeavor. In the present letter, we take the first steps and describe a proof-of-principle experiment where these three functions are implemented simultaneously in a single fiber. We use low-coherence reflectometry for monitoring and positioning the microstructured fiber tip in front of the sample, fiber-guided high-power laser delivery for ablating a perforation in the sample, and microfluidics in the holes of the microstructured fiber for fluid retrieval. In section 2, we characterize the fiber-based reflectometer. In section 3, a combination of laser light delivery and fluid recovery in a microstructured fiber is demonstrated, and in section 4 we investigate fiber delivered laser ablation. In section 5, we describe how all three techniques can be combined utilizing the multifunctionality of the microstructured fiber, and in section 6 we draw some conclusions.
2. Low-coherence reflectometry in glass samples
An Ando AQ7410B reflectometer incorporating a short coherence length LED-source at the operating wavelength 1310 nm and a scanning Michelson interferometer is used for the low-coherence reflectometry studies, measuring the distance between reflective surfaces. The nominal spatial resolution is 20 µm and the range scanned for any given reference arm length equal to 1.3 m. The reflectometer is preliminarily characterized in its ability to measure distances with micrometric resolution and detect the presence of liquids. To this end, silica fiber capillaries are used, with inner/outer diameter 25/125 µm, 45/125 µm and 129/300 µm. The probe signal from the interferometer and red light from a HeNe laser are combined in a fused coupler to allow for visual alignement and to identify the object under study. In some experiments, a 10 × microscope objective collimates the beam that exits the fiber probe, as illustrated in Fig. 1 . Proper alignement minimizes the sensitivity of the reflectometer to sample surface roughness.
Figure 2 shows one example of measurement in the radial direction of a 25 µm fiber capillary that is empty (filled with air). The vertical axis shows the amplitude of the signal reflected at various interfaces, obtained from the interferometric fringe contrast. The trace is normalized to the signal from the entrance surface, which gives the largest reflection. The fiber capillary has its acrylate primary coating on, and the index mismatch to the silica glass gives a reflection only ~1 dB above the background, 25 dB below the main reflection peak. The relative position of the major peak can be determined to within ± 2 µm. The hole diameter measured from the black trace in Fig. 2 (optical distance 18 µm) translates to 26 µm when the refractive index is corrected from the instrument-assigned value 1.46 to 1. When the fiber capillary is filled with distilled water, the measured optical distance changes to 23 µm (dashed red trace) and the average amplitude of the reflections is reduced by 41 times because of index-matching.
Note that the expected ratio for a planar surface at normal incidence Fresnel reflection is 21 times, but here the surface investigated is cylindrical. Again correcting the refractive index from 1.46 to 1.33 one infers a 25-µm physical hole diameter, in good agreement with the actual fiber capillary hole diameter. The dramatic reduction in amplitude of the reflection when the capillary holes are filled can be used for sensing the presence or absence of liquid.
Figure 3 illustrates the resolution obtained in measurements of an air inclusion in a 0.47 mm thick glass sample with the low-coherence interferometer, using the setup shown in Fig. 1. The air inclusions are trapped during the cooling of the glass to solid phase. The depth and position of the air inclusion in relation to the surface of the glass sample is determined with an accuracy <5 µm. The low-coherence reflectometer is a good tool to measure the distance between a fiber tip and the sample to be ablated.
3. Combining light delivery and fluid collection in microstructured fibers
A 125-µm diameter microstructured silica fiber made by Acreo AB with four 26-µm diameter holes around a standard single-mode 8.3 µm Ge-doped core is used in the experiments of fluid retrieval and laser light delivery, as shown in Fig. 4 (a) . A special fiber arrangement is required for fluid collection without hindering light guidance in the fiber. In particular, very low-loss coupling to the ablating laser is necessary, to prevent leakage of high optical power. Here, the microstructured 4-hole fiber is fusion spliced to a piece of single-mode fiber (SMF28) etched to an outer diameter of ~78 µm. A few-centimeters long piece of SMF is etched in 50% concentrated hydroflouric acid (HF) for 15 minutes before low-loss (<0.2 dB) splicing to the microstructured 4-hole fiber, as seen in Fig. 4 (b). Capillary force alone can draw water-based liquids into the holes of the microstructured fiber, as illustrated in Fig. 4 (c). Fluid flow with only capillary forces is allowed for, since the air is not trapped in the holes of the microstructured fiber because of the partial overlap of the holes and the SMF.
4. Fiber delivered laser ablation experiments
A Q-switched and mode-locked Nd:YAG laser is used for ablation, with an operating wavelength 1064 nm and repetition rate up to 4.2 kHz. The peak power of each individual ~200 ps pulse is estimated at 40 kW. A frequency doubling KTP crystal is used to generate second harmonic (SH) green light. In some cases, the combination of second harmonic and fundamental radiation can reduce the ablation threshold , but the SH is useful here mainly for visualization of the ablating beam. Each fiber-coupled Q-switched train of pulses contains ~100 µJ energy distributed in ~20 pulses and ~30% of the energy is contained in the pre-lase. This regime of laser operation has been reported to result in efficient ablation . The maximum energy density used for the ablation process is calculated to be 0.6 J/cm2 in the individual mode-locked pulses. The maximum average power coupled through the fiber is ~600 mW. An average power of ~300 mW is sufficient for ablation of a 23 µm thick aluminum foil. The choice of target material for this proof-of-principle experiment is based on various considerations, including heat conduction in a metallic sample, the relatively low power available from the laser source and the observation that little or no debris is formed from aluminum targets during ablation. It is observed that ablation at half repetition rate takes approximately twice longer. No hole is created in the aluminum foil illuminated by 500 mW average power beam when the laser is operated continuous-wave. This observation implies that the ablation process is dominated by the high intensity of the short pulses, and not by a simple temperature increase of the target.
Due to the divergence of the laser beam, the required distance for ablation is <1 mm. To avoid damaging the fiber tip, the distance between fiber tip and sample is kept >50 µm. In the experiments that follow, a distance of 100 µm is chosen for ablation of the aluminum foil. From a numerical aperture 0.14 and a mode-field diameter 8.0 µm at 1.06 µm, it is calculated (in a Gaussian approximation) that the beam diameter spreads to 27 µm on the target. However, the diameter of the circular hole produced in the aluminum foil is measured to be ~40 µm, implying that the ablation area is approximately twice wider than the size of the Gaussian spot, see Fig. 5 .
5. Experiment combining reflectometry, microfluidics and ablation
The steps described previously are then combined in a technique where a microstructured fiber is positioned, as shown Fig. 6 (a) , by reflectometry and used for ablation and fluid retrieval, as shown in Fig. 6 (b). An artificial cell is designed, consisting of a small plastic container filled with a Rhodamine 6G solution in distilled water and covered with a membrane, a piece of 23 µm thick aluminum foil. The cell can be pressurized by adding more solution through an opening with a syringe.
A non-achromatic 10 × microscope objective is used to couple the high-power ablating light into the microstructured 4-hole fiber, via a section of etched SMF28 fiber, as illustrated in Fig. 6 (c), together with light from the reflectometer at 1.3 µm wavelength and green light. Focusing is optimized to maximize the 1.06-µm radiation at the expense of the green light power launched. Coupling of the reflectometer signal into the fiber is fine tuned by adjusting the position and divergence of the probe light onto the 10 × microscope objective with a collimating lens at the exit of the fiber-coupled reflectometer. It is observed that the few-mW average power green light incident on the sample does not reduce the ablation threshold as hoped. However, the green light is still useful for visualization of the perforation, from the transmission of the green radiation through the otherwise opaque sample.
The positioning of the fiber for ablation and the fluid collection process can be monitored with reflectometry. The fiber tip gives rise to a large and initially constant reflection peak. The fiber tip is brought towards the aluminum foil cover of the liquid-filled cell and positioned 100 µm away from it, as illustrated in Fig. 7 (a) . The ablation process is so rapid that it cannot be followed in real-time here by the reflectometer, but after ~1 second of exposure to the 1.06 µm radiation the reflection of the aluminum foil surface decreases by more than 10 dB, evidencing the opening of the hole. Green light is then transmitted into the cell, a process that can be visually followed by inspection from the side of the arrangement. The reflectometer trace then shows a reflection peak from the liquid surface ~120 µm away from the fiber tip reflection, shown in Fig. 7 (b). The liquid infiltrates the ablated area, a process monitored with the reflectometer and characterized by the liquid surface reflection moving closer and closer to the fiber tip, as seen in Fig. 7 (c). Figure 7 (d) evidences that the reflection of the fiber tip decreases significantly (~13 dB), due to index-matching, when the tip gets in contact with the liquid. This also marks the beginning of fluid retrieval by the microstructured 4-hole fiber, shown in Fig. 7 (e). Note that ablation and fluid collection take place here without mechanical movement of the fiber tip. If and when the fiber-liquid contact is broken (for instance, by depletion of the liquid on the top surface of the cell) the fiber tip reflection increases (~13 dB) and the reflection from a liquid surface also appears, separated from the fiber tip by ~70 µm as illustrated in Fig. 7 (f), evidencing completion of fluid collection.
A technique has been developed for the retrieval of liquids contained in an artificial cell, following laser ablation using a microstructured optical fiber. In the proof-of-principle demonstration, the same fiber used for ablation and collection is employed with low-coherence reflectometry for positioning and monitoring. The experiments show that microfluidic flow can be monitored by studying the amplitude and the position of a low-coherence reflected signal, the peak of which can be determined to a resolution of a few microns. The experiments described here can be extended to ablating pulses of higher power, e.g., with a few nanoseconds Nd:YAG laser pulses. The low attenuation in silica fiber of the light signals used for ablation and monitoring make it possible to extend significantly the range of the collection technique. To this end, the fiber-based reference arm of the reflectometer has to be extended to match to within ~1 m the length of the signal fiber. A potentially useful extension of the results described here is the collection in the fiber holes for further analysis of a liquid used as a conveying means of ablated material. The liquid can be dispensed from the holes prior to ablation and collected after the material is ejected by light.
It is a pleasure to thank F. Laurell and L. E. Berg from the Royal Institute of Technology (KTH) in Stockholm for use of facilities and equipment, and Oleksandr Tarasenko, Ralf Koch and Leif Kjellberg for experimental help. This work is carried out in the framework of Acreo Fiber Optic Center and partly sponsored by the Swedish Research Council.
References and links
1. B. M. Kim, M. D. Feit, A. M. Rubenchik, E. J. Joslin, P. M. Celliers, J. Eichler, and L. B. Da Silva, “Influence of pulse duration on ultrashort laser pulse ablation of biological tissues,” J. Biomed. Opt. 6(3), 332–338 (2001). [CrossRef] [PubMed]
3. H. Pratisto, M. Frenz, M. Ith, H. J. Altermatt, E. D. Jansen, and H. P. Weber, “Combination of fiber guided pulses erbium and holmium laser radiation for tissue ablation under water,” Appl. Opt. 35(19), 3328–3337 (1996). [CrossRef] [PubMed]
5. C. Monat, P. Domachuk, and B. J. Eggleton, “Integrated optofluidics: A new river of light,” Nat. Photonics 1(2), 106–114 (2007). [CrossRef]
6. S. A. Kingsley and D. E. N. Davies, “OFDR diagnostics for fibre and integrated-optic systems,” Electron. Lett. 21(10), 434–435 (1985). [CrossRef]
7. H. H. Gilgen, R. P. Novak, R. P. Salathe, W. Hodel, and P. Beaud, “Submilimeter optical reflectometry,” J. Lightwave Technol. 7(8), 1225–1233 (1989). [CrossRef]
8. D. Stifter, “Beyond biomedicine: a review of alternative applications and developments for optical coherence tomography,” Appl. Phys. B 88(3), 337–357 (2007). [CrossRef]
9. S. A. Boppart, J. Herrmann, C. Pitris, D. L. Stamper, M. E. Brezinski, and J. G. Fujimoto, “High-resolution optical coherence tomography-guided laser ablation of surgical tissue,” J. Surg. Res. 82(2), 275–284 (1999). [CrossRef] [PubMed]
10. S. Zoppel, R. Merz, J. Zehetner, and G. A. Reider, “Enhancement of laser ablation yield by two color excitation,” Appl. Phys., A Mater. Sci. Process. 81(4), 847–850 (2005). [CrossRef]
11. S. M. Klimentov, S. V. Garnov, T. V. Kononenko, V. I. Konov, P. A. Pivovarov, and F. Dausinger, “High rate deep channel ablative formation by picosecond–nanosecond combined laser pulses,” Appl. Phys., A Mater. Sci. Process. 69(7Suppl.), S633–S636 (1999). [CrossRef]