Open Access
Add to BibSonomyAbstract
Ultrashort pulse ablation has become a useful tool for micromachining and biomedical surgical applications. Implementation of ultrashort pulse ablation in confined spaces has been limited by endoscopic delivery and focusing of a high peak power pulse. Here we demonstrate ultrashort pulse ablation through a thin multi-core fiber (MCF) using wavefront shaping, which allows for focusing and scanning the pulse without requiring distal end optics and enables a smaller ablation tool. The intensity necessary for ablation is significantly higher than for multiphoton imaging. We show that the ultimate limitations of the MCF based ablation are the nonlinear effects induced by the pulse in the MCFs cores. We characterize and compare the performance of two devices utilizing a different number of cores and demonstrate ultrashort pulse ablation on a thin film of gold.
© 2017 Optical Society of America
1. Introduction
Ultrashort pulse laser ablation has proven to be a powerful tool in high precision manufacturing [1,2] and surgical applications [3,4]. The highly nonlinear absorption process allows for ablation with high precision and resolution with minimal collateral damage [1,5,6]. These benefits have allowed ultrashort ablation for micromachining in a variety of materials, including metals, semiconductors, glass, polymers and others [2,5–7]. Furthermore, ultrashort pulse ablation has become indispensable to ophthalmic surgery [3,4], while new biomedical applications are being developed [3].
A unique challenge for enabling new ultrashort pulse ablation applications is delivering the high peak power, ultrashort pulses in confined spaces where we cannot reach with conventional microscope objectives. Optical fibers have the possibility of light delivery, but the pulse would induce strong nonlinear effects in conventional optical fibers and the fibers themselves would be susceptible to ablation by the spatially confined pulse. The last decade has seen a rapid growth in the development of photonic crystal fibers [8], of which hollow core photonic crystal fibers have shown an unsurpassed ability to deliver high energy, ultrashort pulses with minimal loss and dispersion [9–11]. However, after propagation through a fiber, the emitted pulse needs to be focused and scanned onto the targeted material. This approach has been followed in the development of several two photon fluorescence [12–17] and ablation endoscopes [17–19], which utilize microlenses with MEMS mirrors or piezoscanners. Nevertheless, the addition of distal end optics and scanning mechanisms significantly add to the size of the endoscope.
To bypass the use of optical components at the distal end of an ablation fiber endoscope, we propose the use of wavefront shaping through a lensless multi-core fiber (MCF) to control a focused high peak power pulse. The MCF mitigates fiber damage and nonlinearities by spreading the pulse energy amongst the many cores of a densely packed MCF. In such a device, the pulses in each core propagate and emerge from the MCF with different phase delay, due to variations between cores and core-to-core coupling. To recombine and focus the pulse after the fiber, the phase of the radiation coupled into each core should be modulated to compensate for the phase delays in the MCF and create a quadratic phase across the distal facet after propagation. The quadratic phase will focus the light after free space propagation some distance from the MCF. Several efforts have been made to realize this system for light control and multiphoton imaging with various MCF types [20–23]. However, ablation through MCF requires much higher peak powers than multiphoton imaging, which leads to nonlinear effects in the fiber cores. GRIN multimode fibers have also been utilized with wavefront shaping for multiphoton imaging [24], however the internal periodic refocusing of modes could induce damage within a GRIN fiber at high optical powers and thereby limits the maximum deliverable pulse energy.
In this paper, we demonstrate, for the first time, the use of MCFs for delivery and control of high peak power (more than 1012 W/cm2), ultrashort pulses for ablation. The peak intensity required for ultrashort pulse ablation via laser induced optical breakdown is between 1011 W/cm2 and 1013 W/cm2. These values are more than two orders of magnitude higher than previously attained using MCF and wavefront shaping for two-photon imaging [23]. We analyze the nonlinear degradation of the created focus spot as a function of input pulse energy and investigate how this affects the focused pulse in terms of pulse width and peak intensity. Finally, we demonstrate ultrashort pulse ablation of a thin gold film with proximal control through a MCF. The development of such small diameter ablation tools could be useful for enabling new microsurgery or microfabrication applications.
2. Wavefront control through MCF
In our previous work we utilized digital phase conjugation for delivering focused ultrashort pulses through a MCF for two-photon fluorescence imaging [23]. In this paper, we use a transmission matrix (TM) approach for controlling the propagation of light through the MCF [25–27]. Experimentally we found that the TM outperformed the digital phase conjugation method in terms of the focusing efficiency, defined as the percentage of output power that is in the focus. The TM bypasses the complexity of alignment between the hologram capturing detector array and the phase conjugating spatial light modulator (SLM) for digital phase conjugation. Like digital phase conjugation, the TM accounts for the core-to-core coupling and allows the use of densely packed MCFs [23,28]. Although for propagation of ultrashort pulses, the core-to-core coupling should be weak to minimize temporal dispersion.
The TM is measured by capturing the output fields of orthogonal input modes after propagation through the MCF [27]. Here we define the output field modes spatially, like pixels (where mode m represents a point in space of the output field), at a distance from the distal facet, while the input field is described at the fiber facet and the modes n are the fundamental modes of the individual cores. The ultrashort pulse limits the TM measurement temporally through coherence gating due to the short coherence length of the laser [23,29]. Once the TM is measured it can be used to calculate the necessary input wavefront to shape arbitrary intensity patterns at the output field [27]. In our case, we are interested in maximizing the focus intensity by creating a single focus spot at output mode m, whose intensity Im is equal to:
where N is the number of cores, An is the amplitude at core n, ϕn is the phase on core n at the input, tmn is the transmission matrix element which relates the input at core n to output mode m, and C is a factor which accounts for reduction of the focusing efficiency due to diffractive effects from the quasi-periodically spaced cores. The intensity Im is maximized when ϕn compensates for the phase delay in core n measured in tmn.Figure 1 shows experimentally obtained data to demonstrate this process. Figure 1(a) shows the facet of a MCF. The MCF is designed with high variability in the core size to reduce core-to-core coupling which allows a high core density [30]. Figure 1(b) shows the phase of the wavefront at the proximal facet of the MCF for generating a focus. The input wavefront appears to be completely random on the facet. Figure 1(c) shows the quadratic wavefront created at the distal end of the MCF using Fig. 1(b) as the input wavefront. The phase was extracted using digital holography. After free space propagation, the quadratic phase focuses the wavefront at some distance away from the fiber facet (Fig. 1(e), intensity in log scale). Because the quadratic phase is discretized by the MCF cores, diffraction peaks appear around the focus [21,23]. However, the relatively close spacing of the cores and their quasi-periodic arrangement separates the diffractive effects far from the focus and spreads out the diffraction peak intensity [23].
Fig. 1 Visualization of wavefront shaping for focusing through MCF. (a) The facet of MCF with white light illumination. (b) The phase field input to the MCF facet to create a focus. The color wheel in the upper right indicates the relative phase value from zero (blue) to 2π (red). The wavefront at the distal end of the MCF with (c) low peak power and (d) high peak power input after the wavefront shown in (b) has propagated through the MCF. Both clearly show the quadratic phase field, although the high peak power case shows deterioration of the wavefront. The focus formed 800 µm from the facet of the MCF with (e) low peak power and (f) high peak power input. The intensity is shown in a log scale. The scale bar in each figure is 50 µm.
In the following sections, we experimentally evaluate how the focus intensity depends on input pulse energy to understand the limitations imposed by nonlinearities.
3. Optical setup
We implemented an optical system for TM measurement and light control through a MCF as shown in Fig. 2. A high pulse energy, ultrafast laser source (Satsuma, Amplitude Systèmes, λ = 1030 nm, 330 fs pulse width, 40 µJ pulse energy) is collimated and split into reference and illumination paths. The illumination path images an SLM (Pluto-NIR2, Holoeye) onto the proximal facet of a MCF (FIGH-10-500N, Fujikura, 10,000 cores) with two lenses in a 4f configuration. We illuminate 4,400 cores of the MCF by using L3 with focal length of 200 mm and O1 was a 20X objective (NA 0.40, Newport). At the distal end of the MCF, another 4f system images a plane at a short distance away from the facet onto a CMOS detector array (MV1-D1312IE-G2-12, PhotonFocus). The second 4f utilized a 200 mm focal length for L4 and O2 was a 20X objective (NA 0.40, Newport). The reference path interferes with the illumination path to make an off-axis digital hologram on the CMOS detector array. The reference path length can be adjusted by a delay line to tune the coherence gating of the pulse transmitted through the MCF [23,29]. The TM is measured by applying a series of plane waves with varying spatial frequencies to the input of the MCF. For each such input, the output of the MCF is recorded holographically. The set of all these input-output measurements forms the transmission matrix. For all experiments, the TM was measured at low power to avoid nonlinear effects during calibration.
Fig. 2 The optical setup for measuring the transmission matrix through the MCF. A collimated, pulsed beam is split into illumination and reference arms. The illumination path images an SLM onto the facet of a MCF. The light transmitted through the MCF is imaged onto a CMOS detector. The reference arm interferes with the illumination arm on the detector to make an off-axis digital hologram. A delay line in the reference path allows for tuning the temporal delay of the reference pulses. An interferometric autocorrelator is built into the system for pulse length measurement. MCF: multi-core fiber, HWP: halfwave plate, PBS: polarized beamsplitter, BS: beamsplitter, SLM: spatial light modulator, D: dichroic filter, S: sample, L: lens, O: objective.
For measuring the temporal pulse width of the laser focus, a second order interferometric autocorrelator was built into the system. A flip mirror directed the pulses transmitted through the MCF into a Michelson interferometer with an adjustable path length arm. The exit path from the interferometer refocused the two beams with an objective onto a fluorescent sample (Rhodamine 6G in SU8). Another objective captured the two-photon fluorescence and imaged it onto a CMOS detector (Chameleon3, Point Grey Research). A dichroic filter blocked the 1030 nm source radiation.
4. Nonlinear degradation of focusing efficiency
To understand the limitations imposed by nonlinearities in the MCF, we measured the focusing efficiency of the system across a range of input pulse energy and pulse widths. Since the nonlinear optical effects depend on intensity, we probed their effect in the MCF by varying the peak power of the input pulse. The Satsuma laser has the built in functionality to stretch the pulse width with a positive chirp, which also decreases the pulse peak power. The pulse energy coupled into the MCF was varied by an internal modulation device in the Satsuma as well as by adjusting the polarization of the beam incident on the SLM (which only modulates horizontally polarized light) with a half wave plate.
The experiment clearly illustrated the degradation of the focus efficiency due to nonlinearities in the MCF. Figure 3 shows the dependence of input pulse energy (measured at the input to the MCF) on focusing efficiency with the four pulse widths, measured by second order interferometric autocorrelation to be 500, 750, 1000, and 2000 fs before the MCF (autocorrelator not shown in Fig. 2). The measurement of the transmission matrix and the creation of the focus was repeated 5 times for each pulse width and the error bars in Fig. 3 indicate the standard deviation of the measurements. These measurements show that the focusing efficiency was highly dependent on the pulse width. As the input pulse energy increased, the focusing efficiency with the shorter pulses degraded more quickly.
Fig. 3 The focusing efficiency when focusing ultrashort pulses of varying pulse widths through a MCF compared to the input pulse energy. The three regimes of the 500 fs pulse focusing efficiency are marked by I, II, and III.
To describe the trends seen, we divide the data into three regimes. Using the 500 fs curve as an example (marked in blue on Fig. 3), we see that (I) a low input energy linear regime exists where the focusing efficiency is independent of pulse width and input pulse energy. (II) Above 0.30 µJ, the focusing efficiency quickly degrades with increasing input energy due to self-phase modulation (SPM), as explained in the following section. (III) Above 1 µJ, the focusing efficiency decreases more slowly with increasing input pulse energy compared to the second regime. We suspect that in the third regime nonlinear effects induced spectral broadening which, consequently, increased the effect of the group velocity dispersion (GVD) in the core and broadened the pulse [31]. The broadened pulse decreased the intensity in the cores and induced less nonlinear phase shift with increasing input pulse energy.
5. Self-phase modulation induced phase shift in MCF
The focal spot at the distal end of the MCF has maximum intensity Im when the input wavefront compensates for the phase delay in the cores to create the ideal quadratic wavefront at the distal facet (Fig. 1(c)), which is the situation in the first regime shown in Fig. 3. To explain the behavior exhibited in the second regime, we will consider the consequence of the phase shift caused by SPM in the MCF. The phase shift in each core is dependent on the nonlinear index n2 and the intensity of the propagating mode. Assuming no pulse broadening and no core-to-core coupling, the phase shift in core n is:
where λ is the wavelength, ωn is the mode radius of core n, L is the length of the MCF and Pn is the peak power of the pulse in the core, determined by the location of the core in the MCF relative to the Gaussian shape of the beam incident on the MCF.By adding the SPM phase shift for core n from Eq. (2) into Eq. (1) the focal spot intensity becomes:
ϕn is the phase of the wavefront at the input of core n that in the absence of nonlinearities would generate a focus after propagating through the MCF. With the MCFs we use, each core induces a unique due to the variability of the core characteristics (e.g. size, shape, material properties). In turn, the phase shift in the cores degrades the focus intensity. For example, Fig. 1(d) shows how by increasing the input pulse energy, the phase shift in the cores degrades the quadratic phase profile. The SPM phase shift only occurs near the MCF center where the intensity in the cores is highest due to the Gaussian profile of the beam. The SPM induced in the cores results in the decrease in focus efficiency seen in Fig. 3. Figure 1(f) shows how this causes a slight increase in the background intensity. Although obscured by the log scale, the maximum intensity of the focus in Fig. 1(f) is half that of Fig. 1(e).In the previous analysis, for simplicity we neglected the effect of core-to-core coupling in the MCFs. While the MCFs were designed to limit the core-to-core coupling in the fiber [30], in practice the coupling modulates power and phase along the length of the MCF. When the input power induces a SPM phase shift, the core-to-core coupling affects the phase and intensity of the pulses propagating in the neighboring cores and further degrades the quadratic phase at the distal facet.
6. Pulse broadening in MCF and peak intensity in focus
As mentioned previously, we suspected that the peak intensity in the cores became high enough to induce spectral and temporal broadening, which led to the third regime shown in Fig. 3. Measuring the pulse duration and spectrum of the output field both in and away from the focus provides some insight into the strength of the spectral and temporal broadening in the MCF.
The pulse width and spectrum were measured with three different input pulse energies (0.6 µJ, 2.5 µJ and 4.5 µJ) and the four previous input pulse widths. For the spectroscopic measurement, a fiber input to a spectrometer (HR4000CG-UV-NIR, Ocean Optics) placed in a plane conjugate to the focus plane could probe either the spectrum of the background or the focus. The pulse width of the focus was measured with an autocorrelator built after the MCF (shown in Fig. 2). Since we have temporally broadened pulses, the pulse durations were estimated by Fourier filtering the second order interferometric autocorrelation trace to isolate the intensity autocorrelation of the pulse, as explained in [32]. The laser emits a squared hyperbolic secant shaped pulse, so the pulse width was calculated by multiplying the FWHM of the intensity autocorrelation by a deconvolution factor of 0.65. However, it is important to note the temporal pulse shape of the wavefront shaping generated focus after the MCF is unknown, but we use the 0.65 deconvolution factor for consistency.
The spectral characterization verified that at high intensities the nonlinear effects were strong enough to broaden the pulse spectrum. The spectral broadening could be induced by SPM, cross phase modulation, four wave mixing, or other nonlinear effects [31]. The spectrum of the background captured for the 500 fs pulse (Fig. 4(a)) showed significant blue-shifted spectral broadening as the input pulse energy increased. For input pulse energies of 0.6 µJ, 2.5 µJ, and 4.5 µJ the respective linewidth (1/e) in the background were 10 nm, 14 nm, and 20 nm. In contrast, the linewidth of the in focus radiation remained unchanged (8 nm) regardless of input pulse energy. Figure 4(b) shows the in and out of focus spectrum with 750 fs input pulse width, which shows that the SPM induced less spectral broadening than the 500 fs case. The spectral broadening for the 1000 fs and 2000 fs input pulse width further decreased.
Fig. 4 The measured spectrum of the in focus (blue curve) and away from focus (orange, yellow and purple curves) light for different input pulse energies with (a) 500 fs and (b) 750 fs input pulse width, also showing the in focus spectrum which did not change with input energy. (c) The in focus pulse width dependence on input pulse energy.
The pulse width of the focus displayed significant temporal broadening dependent on the input pulse energy. Figure 4(c) shows how the 500, 750, 1000, and 2000 fs pulses broadened with increasing input pulse energy. These results imply that the nonlinear effects modified the broadening rate of the pulse in the cores, in other words the pulse broadened beyond the effect of GVD alone [31]. Interestingly, the pulse width of the focus at low input pulse energy was consistently shorter than the width of the pulse measured directly from the laser. It is unclear whether the pulse has compressed due to material dispersion in the MCF or whether it was an unexpected effect of the coherence-gated hologram. Further work will need to be conducted to identify the source of the compression.
Ultimately, the performance of the device for ablation depends on the peak intensity of the focused pulse. We estimated the peak intensity with the measured pulse durations and the energy in focus. Figure 5 shows the peak intensity of the created focus. Regardless of the input pulse width, the maximum intensity is around 1.1x1012 W/cm2. Notably, the maximum obtained peak intensity occurs for all pulse widths at a focusing efficiency that is ~75% of the maximum obtained in regime I (comparing Fig. 3 and Fig. 5).
Fig. 5 The estimated peak intensity in focus plotted for increasing input pulse energy and pulse width.
7. Optimized 10,000 core system
The presented description of the limitations of using a MCF for high power ultrashort pulse delivery only used 4,400 cores of the 10,000 cores of the MCF. In this section, we present an optimized system that utilizes optimally the entire MCF and maximizes the peak intensity of the focus spot.
We modified the optical system for optimal usage of all cores in the MCF. Referring to the optical setup diagram in Fig. 2, the L3 lens was changed to a focal length of 125 mm, which caused the beam to slightly overfill the MCF facet. The second 4f system was also changed: L4 to a 150 mm focal length lens and O2 to a 40X objective (NA 0.65).
To maximize the peak intensity, a balance between the focusing efficiency and the size of the focus spot must be found by selection of the distance between the MCF facet and the focus (working distance). For use of all of the delivered pulse energy from the MCF, the minimum working distance should be where all of the cores contribute to the focus, a distance limited by the NA of the cores (~NA 0.3) and the diameter of the illuminated portion of the MCF [28]. Beyond this point, the NA of the created focus (NAfoc) is given by the ratio of the radius of the illuminated aperture of the MCF and the working distance. While a higher NAfoc results in a smaller focus spot size, a lower NAfoc increases focusing efficiency. In fact, the discretization of the wavefront by the cores results in undersampling near the edges that lowers the focusing efficiency with stronger quadratic phase fields [23]. Table 1 presents the working distance and NAfoc utilized with 4,400 cores and the optimized 10,000 core system.
Table 1. A comparison of the two optical systems described.
In contrast to the 4,400 core system, the 10,000 core system had significantly reduced nonlinear degradation of the focus (Fig. 6(a)). Of course, this was an expected result of dividing the pulse energy among more cores. Spreading the pulse energy into more cores decreased the spectral broadening (Fig. 6(b)) and temporal broadening (Fig. 6(c)). The focusing efficiency when using 10,000 cores was less than with 4,400 cores at low pulse energies, because of the higher focus NA (Table 1). Despite the lower focusing efficiency, the smaller focus size and decrease in nonlinear effects resulted in an increased maximum peak intensity (Fig. 6(d)), in this case above 4x1012 W/cm2 with a 500 fs pulse, or nearly 4 times higher than the 4,400 core system.
Fig. 6 The characterization of the 10,000 core system. (a) The focusing efficiency, (b) the measured spectrum of the in focus and background light for different input pulse energies with 500 fs pulse width, (c) the in focus pulse width dependence on input pulse energy, and (b) the estimated peak intensity.
8. Ultrashort pulse ablation of gold
As a demonstration of the high peak power ultrafast pulse focusing and control through the MCF, we ablated a thin gold film (60 nm) deposited on silica. The ablation threshold of gold is 5.5x1011 W/cm2 with a 750 fs pulse [33]. We utilized the 4,400 core system for this demonstration to show that gold ablation is possible even with the more constrained system. To enable the proximal control of the light field through the MCF, the TM output field had dimensions of 200 µm x 200 µm. TM calculations and wavefront shaping, allow control of the distributed light intensity within the output field. In other words, a focus spot can be created anywhere within the measured output field, which allows for sequential focusing at designated locations to create an ablated pattern in the gold. Because the size of the ablated spot on the gold is intensity dependent [6], it is desirable that the focus intensity be invariant across the scanning range. However, when focusing away from the center of the output field, the focus intensity diminishes [23]. To compensate, we modified the input wavefronts to make the focus intensities at the targeted locations uniform. To do this, we decreased the intensity of the higher intensity spots by adding noise to the input wavefront proportional to the targeted decrease in intensity. This is similar to adding random phase values to the input wavefront. Importantly, adding the noise to the phase mask does not affect the focus spot size.
Figure 7 shows two examples of gold ablation through MCF with proximal control of the beam. An EPFL logo ablated with input pulse energy of 2 µJ, 750 fs pulse width, and a repetition rate of 1 kHz is shown in Fig. 7(a). The size of the pattern is 106 µm x 82 µm with a 2 µm separation between ablation focal spots. A video showing the ablation of the EPFL logo is found in the Visualization 1. Figure 7(b) shows an image of the Matterhorn and a Swiss shield ablated with input energy of 3 µJ, 750 fs pulse width, and 1 kHz repetition rate. The image is 120 µm x 120 µm and has a 2 µm pitch between focal spots. Both ablated images show good uniformity of ablation spot size across the image and show the capability of proximal control to scan a focus spot for ablation.
Fig. 7 Widefield transmission optical microscope images of ultrashort pulse ablated samples of thin gold film on glass. (a) EPFL letters, 106 µm x 82 µm. (b) The Matterhorn and Swiss shield, 120 µm x 120 µm.
9. Discussion and conclusion
We have demonstrated high peak power, ultrashort pulse focusing and control through MCFs and shown that the nonlinearities induced within the cores limit the ultimate peak intensity generated in a focus. Wavefront shaping allowed proximal control of a focus through the MCF, which eliminated the need for distal end optical components for focusing and scanning. We explored the nonlinear effects within the MCF and showed that they degrade the optical focus. Specifically, the nonlinearities degrade the percentage of output power in the focus and contribute to the temporal broadening of the pulse, both of which limit the peak intensity of the focused pulse. We found that with only 4,400 cores of the MCF, we could generate an optical focus with sufficient intensity to ablate gold.
To increase the intensity in a wavefront shaping generated focus a number of paths could be followed. Several solutions involve modification of the MCF. For example, a shorter MCF would decrease the nonlinear interaction length and increase focusing efficiency. Also, as shown in this paper, by spreading the pulse energy among more cores, the nonlinearity can be minimized and a higher intensity focus created. Increasing the MCF diameter or the density of the cores would allow the use of more cores. Although. with currently available MCFs it is not possible to use a smaller core spacing without significantly increasing the core-to-core coupling [30]. Alternatively, small distal components such as a GRIN lens could be included on the distal tip to focus with a higher NA and increase the intensity of the focus [24]. Other potential solutions involve further shaping of the wavefront and/or pulse to compensate for SPM [34]. As with other wavefront shaping through fiber techniques, the TM measurement is fiber conformation dependent. While the MCF shows some resilience to bending [28], new adaptive methods could overcome this obstacle [35].
The proposed ultrashort pulse ablation method opens the door for new applications in micromachining and microsurgery. By controlling the intensity field of the distal end of the MCF from the proximal side with wavefront shaping, the MCF essentially acts as a long, ultrathin objective with a high aspect ratio, in this case 200:1. It could become a useful tool to enable ultrashort pulse ablation in places currently unreachable by conventional objectives or with larger endoscopic devices.
Funding
The Bertarelli Program in Translational Neuroscience and Neuroengineering (10271); The Wyss Center for Bio- and Neuro-Engineering (10455).
References and links
1. X. Liu, D. Du, and G. Mourou, “Laser ablation and micromachining with ultrashort laser pulses,” IEEE J. Quantum Electron. 33(10), 1706–1716 (1997). [CrossRef]
2. R. R. Gattass and E. Mazur, “Femtosecond laser micromachining in transparent materials,” Nat. Photonics 2(4), 219–225 (2008). [CrossRef]
3. C. L. Hoy, O. Ferhanoglu, M. Yildirim, K. H. Kim, S. S. Karajanagi, K. M. C. Chan, J. B. Kobler, S. M. Zeitels, and A. Ben-Yakar, “Clinical Ultrafast Laser Surgery: Recent Advances and Future Directions,” IEEE J. Sel. Top. Quantum Electron. 20(2), 242–255 (2014). [CrossRef]
4. H. K. Soong and J. B. Malta, “Femtosecond Lasers in Ophthalmology,” Am. J. Ophthalmol. 147(2), 189–197 (2009). [CrossRef] [PubMed]
5. J. Cheng, C. Liu, S. Shang, D. Liu, W. Perrie, G. Dearden, and K. Watkins, “A review of ultrafast laser materials micromachining,” Opt. Laser Technol. 46, 88–102 (2013). [CrossRef]
6. P. P. Pronko, S. K. Dutta, J. Squier, J. V. Rudd, D. Du, and G. Mourou, “Machining of sub-micron holes using a femtosecond laser at 800 nm,” Opt. Commun. 114(1-2), 106–110 (1995). [CrossRef]
7. M. D. Shirk and P. A. Molian, “A review of ultrashort pulsed laser ablation of materials,” J. Laser Appl. 10(1), 18–28 (1998). [CrossRef]
8. P. Russell, “Photonic crystal fibers,” Science 299(5605), 358–362 (2003). [CrossRef] [PubMed]
9. Y. Wang, M. Alharbi, T. D. Bradley, C. Fourcade-Dutin, B. Debord, B. Beaudou, F. Gerôme, and F. Benabid, “Hollow-core photonic crystal fibre for high power laser beam delivery,” High Power Laser Sci. Eng. 1(01), 17–28 (2013). [CrossRef]
10. F. Benabid, “Hollow-core Photonic Crystal Fibers: Guidances and Applications,” in Advanced Photonics, OSA Technical Digest (Online) (Optical Society of America, 2014), p. SoM4B.5.
11. Y. Y. Wang, X. Peng, M. Alharbi, C. F. Dutin, T. D. Bradley, F. Gérôme, M. Mielke, T. Booth, and F. Benabid, “Design and fabrication of hollow-core photonic crystal fibers for high-power ultrashort pulse transportation and pulse compression,” Opt. Lett. 37(15), 3111–3113 (2012). [CrossRef] [PubMed]
12. R. Le Harzic, M. Weinigel, I. Riemann, K. König, and B. Messerschmidt, “Nonlinear optical endoscope based on a compact two axes piezo scanner and a miniature objective lens,” Opt. Express 16(25), 20588–20596 (2008). [CrossRef] [PubMed]
13. S. F. Elahi and T. D. Wang, “Future and advances in endoscopy,” J. Biophotonics 4(7-8), 471–481 (2011). [CrossRef] [PubMed]
14. M. Gu, H. Bao, and H. Kang, “Fibre-optical microendoscopy,” J. Microsc. 254(1), 13–18 (2014). [CrossRef] [PubMed]
15. R. L. Harzic, I. Riemann, M. Weinigel, K. König, and B. Messerschmidt, “Rigid and high-numerical-aperture two-photon fluorescence endoscope,” Appl. Opt. 48(18), 3396–3400 (2009). [CrossRef] [PubMed]
16. S. P. Mekhail, G. Arbuthnott, and S. N. Chormaic, “Advances in Fibre Microendoscopy for Neuronal Imaging,” Opt. Data Process. Storage 2, (2016). [CrossRef]
17. C. L. Hoy, N. J. Durr, P. Chen, W. Piyawattanametha, H. Ra, O. Solgaard, and A. Ben-Yakar, “Miniaturized probe for femtosecond laser microsurgery and two-photon imaging,” Opt. Express 16(13), 9996–10005 (2008). [CrossRef] [PubMed]
18. C. L. Hoy, W. N. Everett, M. Yildirim, J. Kobler, S. M. Zeitels, and A. Ben-Yakar, “Towards endoscopic ultrafast laser microsurgery of vocal folds,” J. Biomed. Opt. 17(3), 038002 (2012). [CrossRef] [PubMed]
19. O. Ferhanoglu, M. Yildirim, K. Subramanian, and A. Ben-Yakar, “A 5-mm piezo-scanning fiber device for high speed ultrafast laser microsurgery,” Biomed. Opt. Express 5(7), 2023–2036 (2014). [CrossRef] [PubMed]
20. A. J. Thompson, C. Paterson, M. A. A. Neil, C. Dunsby, and P. M. W. French, “Adaptive phase compensation for ultracompact laser scanning endomicroscopy,” Opt. Lett. 36(9), 1707–1709 (2011). [CrossRef] [PubMed]
21. E. R. Andresen, G. Bouwmans, S. Monneret, and H. Rigneault, “Two-photon lensless endoscope,” Opt. Express 21(18), 20713–20721 (2013). [CrossRef] [PubMed]
22. E. R. Andresen, G. Bouwmans, S. Monneret, and H. Rigneault, “Toward endoscopes with no distal optics: video-rate scanning microscopy through a fiber bundle,” Opt. Lett. 38(5), 609–611 (2013). [CrossRef] [PubMed]
23. D. B. Conkey, N. Stasio, E. E. Morales-Delgado, M. Romito, C. Moser, and D. Psaltis, “Lensless two-photon imaging through a multicore fiber with coherence-gated digital phase conjugation,” J. Biomed. Opt. 21(4), 45002 (2016). [CrossRef] [PubMed]
24. E. E. Morales-Delgado, D. Psaltis, and C. Moser, “Two-photon imaging through a multimode fiber,” Opt. Express 23(25), 32158–32170 (2015). [CrossRef] [PubMed]
25. S. M. Popoff, G. Lerosey, R. Carminati, M. Fink, A. C. Boccara, and S. Gigan, “Measuring the Transmission Matrix in Optics: an Approach to the Study and Control of Light Propagation in Disordered Media,” Phys. Rev. Lett. 104(10), 100601 (2010). [CrossRef] [PubMed]
26. D. Kim, J. Moon, M. Kim, T. D. Yang, J. Kim, E. Chung, and W. Choi, “Toward a miniature endomicroscope: pixelation-free and diffraction-limited imaging through a fiber bundle,” Opt. Lett. 39(7), 1921–1924 (2014). [CrossRef] [PubMed]
27. D. Loterie, S. Farahi, I. Papadopoulos, A. Goy, D. Psaltis, and C. Moser, “Digital confocal microscopy through a multimode fiber,” Opt. Express 23(18), 23845–23858 (2015). [CrossRef] [PubMed]
28. N. Stasio, D. B. Conkey, C. Moser, and D. Psaltis, “Light control in a multicore fiber using the memory effect,” Opt. Express 23(23), 30532 (2015). [CrossRef]
29. E. E. Morales-Delgado, S. Farahi, I. N. Papadopoulos, D. Psaltis, and C. Moser, “Delivery of focused short pulses through a multimode fiber,” Opt. Express 23(7), 9109–9120 (2015). [CrossRef] [PubMed]
30. X. Chen, K. L. Reichenbach, and C. Xu, “Experimental and theoretical analysis of core-to-core coupling on fiber bundle imaging,” Opt. Express 16(26), 21598–21607 (2008). [CrossRef] [PubMed]
31. G. Agrawal, Nonlinear Fiber Optics (Academic, 2001).
32. M. Wollenhaupt, A. Assion, and T. Baumert, “Femtosecond Laser Pulses: Linear Properties, Manipulation, Generation and Measurement,” in Springer Handbook of Lasers and Optics, F. T. Prof, ed. (Springer New York, 2007), pp. 937–983.
33. B. C. Stuart, M. D. Feit, S. Herman, A. M. Rubenchik, B. W. Shore, and M. D. Perry, “Optical ablation by high-power short-pulse lasers,” J. Opt. Soc. Am. B 13(2), 459–468 (1996). [CrossRef]
34. M. Tsang, D. Psaltis, and F. G. Omenetto, “Reverse propagation of femtosecond pulses in optical fibers,” Opt. Lett. 28(20), 1873–1875 (2003). [CrossRef] [PubMed]
35. S. C. Warren, Y. Kim, J. M. Stone, C. Mitchell, J. C. Knight, M. A. A. Neil, C. Paterson, P. M. W. French, and C. Dunsby, “Adaptive multiphoton endomicroscopy through a dynamically deformed multicore optical fiber using proximal detection,” Opt. Express 24(19), 21474–21484 (2016). [CrossRef] [PubMed]
