Open Access
Add to BibSonomy
Abstract
We report a diode-pumped femtosecond Ti:sapphire laser in which a suspension of single-walled carbon nanotubes spin-coated onto a cavity mirror is the saturable absorber element. Laser performance is presented for single-diode pumping at 462 nm with 51-fs pulses being achieved with an output power of 27 mW. The laser exhibits stable operation with a relative intensity noise of 0.26% and a pulse repetition frequency of 79.24 MHz. Measurements reporting saturable and non-saturable losses each of ∼0.1% illustrate that this saturable absorber element is particularly appropriate for use in a low gain system such as a diode-pumped Ti:sapphire laser.
Published by The Optical Society under the terms of the Creative Commons Attribution 4.0 License. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.
1. Introduction
Diode pumping of Ti:sapphire lasers [1], enabled by improvements in brightness and output power from visible laser diodes [2,3], massively reduces their complexity, size, power consumption and cost by eliminating the need for sophisticated diode-pumped solid-state (DPSS) pump lasers and the water-cooling systems that accompany them. Modelocking of diode-pumped Ti:sapphire lasers has achieved powers ranging from tens [4] to hundreds [5] of milliwatts and multi-nJ energies [6] for lasers operating with sub-100-fs, with modelocking achieved originally using a semiconducting saturable absorber mirror (SESAM) [7], and later by using Kerr-lens modelocking (KLM) [8], which can enable broader tunability [9]. KLM can be more difficult to achieve in such systems because the asymmetric, non-diffraction-limited beams from high-power visible diode lasers make soft-aperture modelocking [10] less reliable, although prescriptions for how to address this have been published [11]. Consequently, and particularly for lower power systems, SESAMs have been employed as the method of choice for achieving modelocking [12,13], but the increasing availability of two-dimensional materials with ultrafast recovery times has stimulated their use as an alternative saturable absorber element in femtosecond oscillators. One such example is single-walled carbon nanotube saturable absorbers (SWCNT-SAs), which present similar nonlinear optical responses to SESAMs and benefit from sufficiently high damage thresholds to be compatible with the Watt-level intracavity powers in a modelocked laser [14–21], with focal intensities at Ti:sapphire wavelengths estimated to be 700 MW cm−2 [15]. Typically, SWCNT-SAs are incorporated into a laser system by spin-coating a thin film of solvent-suspended carbon nanotubes onto either a cavity mirror (reflection type) or a window (transmission type). Both of these embodiments were first demonstrated in 2004 in a 1550-nm Er:fiber laser, whose operating wavelength matched the S1 absorption peak [14]. More recently, SWCNT-SAs were demonstrated as modelocking elements in DPSS laser-pumped Ti:sapphire lasers [15,16], in which operation at around 810 nm was matched to the shorter-wavelength S2 absorption. Directly diode-pumped femtosecond lasers incorporating SWCNT-SAs have since been reported in gain media including Er:doped-fiber [14], Nd:glass [17,18], Er/Yb:glass [17,19], Nd,Y:SrF2 [20] and very recently Cr:LiSAF [21], corresponding to wavelengths ranging from 850–1570 nm. Each of these systems reported modelocking at pump powers of less than 2 W, however Ti:sapphire lasers modelocked using SWCNT-SAs have so far required pump powers > 1.25W to avoid Q-switching. In this work we present a directly diode pumped Ti:sapphire laser incorporating a reflection-type SWCNT-SA and providing 51-fs pulses at 802 nm for pump powers as low as 1.01 W. This system requires no water cooling and is shown to be capable of stable, low-noise cw-modelocked operation.
2. SWCNT-SA fabrication
The SWCNTs used were IsoNanotubes-S from NanoIntegris, Inc., which are produced by an arc discharge process and are purified via density gradient centrifugation. They were 98% semiconducting, with lengths from 300 nm to 5 µm (mean length 1 µm) and diameters from 1.2–1.7 nm (mean diameter 1.4 nm) [22]. SWCNTs with these dimensions exhibit optical absorption from 700–1200 nm arising from electronic transitions (S22 van Hove singularity) [23], and the manufacturer's spectral data [22] show that near 800 nm the absorbance is 38% of the value measured at the peak of the S22 transition.
The SWCNT-SA was prepared by first suspending 100 µg of SWCNTs in 500 µl of dichlorobenzene (DCB) through ultrasonic agitation. Meanwhile, in preparation for spin coating, a plane dielectric mirror with high reflectivity from 725–875 nm was treated with the adhesion promoter hexamethyldisilazane. The suspended carbon nanotubes were spin-coated onto the mirror surface at 3500 rpm and immediately baked at 95°C to evaporate the suspension solution. As reported in [24], carbon nanotubes show a tendency to clump together and form aggregates owing to van der Waals attractive forces between the tubes, which can introduce a source of non-saturable loss [25] and prevent exploitation of the uniquely high surface area of these materials. Clumping was visible after spin coating, and in Fig. 1 we show optical microscopy (Fig. 1(a)) and electron microscopy (Fig. 1(b)) images of the clumped nanotubes, which indicate that. even in the clumped regions. there were substantial gaps between the nanotubes that left the majority of their surface areas exposed. The atomic force microscope image in Fig. 1(c) shows that the local height of one such clump is ∼50 nm, and the residue left after DCB evaporation can be clearly seen as the concentric rings surrounding the clump. Laser modelocking was obtained when the laser was focused onto a single clump. The inset in Fig. 1(a) shows evidence of laser damage on one particular clump, which was the typical failure mechanism we observed.
Fig. 1. (a) Optical microscope images of the SWCNT-SA, showing clumping and (inset) a clump of nanotubes showing local optical damage. (b) Scanning electron microscope images of a single clump and (inset) detail of the individual carbon nanotubes in the dashed region. (c) Atomic force microscope image of the edge of one clump showing a surface height above the mirror of ∼50 nm. The concentric rings surrounding the clump are residue after evaporation of the DCB.
In Fig. 2 we present the mirror reflectivity curve after coating with the SWCNTs. The original mirror was a simple high reflectivity dielectric coating, centered at 800 nm and of a standard commercially available design. We measured the reflectance curve of the SWCNT-coated mirror (Fig. 2, blue), which shows a reflectivity of >99.9% over the laser operating bandwidth from 760–820 nm. This was gathered by placing the mirror between a broadband near-IR source and a spectrometer and comparing this reading with a reference measurement without the mirror. It should be noted that while this is not a measurement of reflectivity at the site of modelocking, the presence of nanotubes does little to alter the mean reflectivity across the mirror surface. For comparison we also show (Fig. 2, red) the reflectivity curve of a SESAM previously used to modelock the same Ti:sapphire laser, as reported in [26], which has a mean low-power reflectivity of 97.5% over a narrower wavelength range from 780–820 nm. This comparison allows us to make two observations about the potential value of the SWCNT-mirror. Firstly, the bandwidth of the underlying Bragg mirror is clearly wider than that of the SESAM, due to the different mirror composition available using dielectric rather than semiconductor Bragg layers. As reported by other groups [21], this wide bandwidth can support the generation of extremely short pulses, although in our laser the pulse duration is limited by the dispersion properties of the GTI mirror (see Fig. 8) and not by the modelocking element. The second benefit is the lower non-saturable loss compared with a semiconductor mirror, which contributes significantly to the total roundtrip cavity loss in a low-gain / low-loss system. Further discussion on this point is given in Section 5.
Fig. 2. Directly measured mirror reflectance spectrum for the SWCNT-coated mirror (blue) and comparison with a Ti:sapphire SESAM (red) used in a similar laser configuration published previously [26]. Inset: Detail of the SWCNT mirror reflectivity showing R > 99.9% across the laser operating wavelength of 760–820 nm.
3. Laser configuration and efficiency
The architecture of our system is shown in Fig. 3 and is similar to previously reported [26]. The pump diode was a 2W 462-nm M462 laser diode [27], which was initially collimated with a short-focal-length aspheric lens before being reshaped using two cylindrical lenses to reduce the fast axis diameter (see Fig. 3 for details). After the final pump lens of focal length 35 mm, this arrangement achieved a near-symmetric spot in the Ti:sapphire gain medium.
Fig. 3. Layout of the diode-pumped Ti:sapphire laser modelocked using a SWCNT-SA. Cylindrical lenses L1 (f = 150 mm) and L2 (f = −50 mm) were used to reduce the diameter of the fast-axis beam. The pump beam was focused into the Ti:sapphire crystal (Ti:S) using a doublet lens of focal length 35 mm (L5). Dispersion compensation was provided by a GTI mirror (M2), and an intracavity fused-silica 4° wedge (W) was used for output coupling with a loss of 0.2% per reflection. Mirrors M1 and M4 were plano-concave reflectors centered at 800 nm and with radii of curvature 250 mm and 500 mm respectively; mirror M3 was plane. The SWCNT-SA was spin-coated onto a plane mirror identical to M3. Insets: Mode radii in the cavity, showing spot sizes in the Ti:sapphire crystal of 42 µm and on the SWCNT-SA of 118 µm. No thermal lens was included in the calculation of the beam radii. Beam profiles from the laser are shown in (a) cw and (b) modelocked operation.
The 4-mm-thick Ti:sapphire crystal had an aperture of 5 × 6 mm2 with a highly reflecting (HR, R>99.8%) coating on its input face. The other face was cut at a 5° wedge and was anti-reflection (R<0.4%) coated. The Ti:sapphire material had an absorption coefficient of 4.1 cm−1 at 532 nm and a figure-of-merit of 100. The crystal itself acted as the first mirror in a linear resonator centered around 800 nm. This was then followed by a collimating HR mirror of radius 250 mm (M1), a dispersion compensating −550-fs2 per bounce Gires-Tournois interferometer (GTI) mirror (M2) and a plane mirror (M3). A fused-silica wedge (W) inserted into the cavity at Brewster’s angle acted as the output coupler and was followed by a 500-mm-radius focusing mirror (M4) and the SWCNT-SA mirror at the end of the cavity. The output coupling provided by the wedge was 0.4%, distributed across two outputs. The insets in Fig. 3 show the beam profiles of the laser in3 (a) cw and 3(b) modelocked operation; these are essentially identical, confirming that Kerr lensing is not responsible for the laser modelocking.
The laser output power is shown in Fig. 4 for cw and modelocked operation. While not self-starting, modelocking could be initiated at average powers of 1.24 W. Modelocking could then be sustained when the incident pump power was reduced to 1.01 W. The maximum output power obtained was 27 mW, with a pump power of 2.02 W. This performance can be compared with results from SESAM-modelocked diode-pumped Ti:sapphire systems optimised for lower power operation. Roth et al. [7] obtained sustainable modelocking for an incident pump power of 439 mW at 452 nm, while Sawai et al. [4] reported well-sustained modelocked operation with an incident pump power of 636 mW at 518 nm.
Fig. 4. Output powers from the laser modelocked using the SWCNT-SA (squares) and a commercial SESAM (circles), shown for CW (blue) and modelocked operation (red). With the SWCNT-SA, modelocking could be initiated at 1.2W pump power. Lines are fits from a rate-equation analysis used to infer the saturable loss of the SWCNT-SA (see Section 5). Error bars represent a ±0.5-mW measurement uncertainty.
Figure 5 shows the output of the laser measured using a fast photodiode. No Q-switching was observed from the laser in routine operation, and appropriate optimization provided continuous use indefinitely over many days. Damage to the SWCNT-SA occurred mainly as a result of Q-switching during alignment, but we also observed a slow degradation of almost 10% of output power in 2 weeks of SWCNT-SA performance, which is consistent with the speculated role of singlet oxygen in the failure mode of SWCNT modelocking elements [17], in which high-peak-power pulses convert atmospheric oxygen into this considerably more reactive form via multi-photon excitation. As proposed in [17], applying a thin over-coating of an oxygen-impermeable layer, e.g. an inorganic film, might therefore extend the lifetime of the SWCNT element.
We tested whether KLM was not responsible for pulse generation in the laser by substituting a standard dielectric high-reflectivity mirror for the SWCNT-coated mirror. No KLM operation could be obtained—even at full pump power—confirming that the cavity was not configured in a stability region suitable for this mode of operation, despite generating higher output powers.
4. Time- and frequency-domain performance
The pulse repetition frequency was 79.24 MHz, and the radio-frequency spectrum of this signal (Fig. 6(a)) shows a signal-to-noise ratio of around 80 dB, indicating excellent mode-locking stability. The lack of sidebands shows that no Q-switching behavior was present and that the laser was stably cw-modelocked. This claim is further supported by the relative intensity noise (RIN) data recorded from 1 Hz to 1 MHz and shown in Fig. 6(b), which indicate low-noise operation, with rms noise of 0.26%.
Fig. 6. (a) Radio-frequency (RF) spectrum of the pulse repetition frequency. No sidebands are present, indicating cw-modelocking with no evidence of Q-switching. (b) Relative intensity noise characterization from 1 Hz to 1 MHz, showing acoustic noise dominating in the kHz band. The contribution at 70 kHz is caused by the laser power supply unit. The cumulative RIN (equal to the rms noise) was 0.26%, supporting the RF data that indicate the absence of Q-switching.
The pulse autocorrelation and accompanying spectrum obtained are shown in Fig. 7. The pulse duration was estimated by adding 1100 fs2 of quadratic spectral phase to the measured spectrum (Fig. 7(b)), for which the calculated autocorrelation envelope (red dashed line) best matched the experiment autocorrelation (Fig. 7(a)). This procedure yielded a pulse with an intensity full-width half maximum duration of 79 fs and a transform-limited duration of 51 fs. As discussed in Section 2, the spin-coated SWCNTs were distributed non-uniformly on the mirror surface, forming small clumps of loosely overlapping nanotubes. As Fig. 7(b) clearly shows, the laser’s peak wavelength is 802 nm, which lies at the edge of the S22 absorption range for SWCNTs of the dimensions we have used [23]. The fact that modelocking is still possible supports the finding of other researchers that limited “bundling” has no significant impact on Mie scattering (unsaturable loss) but can accelerate the absorbed relaxation time and lead to a considerable broadening of the saturable absorption bandwidth [28].
Fig. 7. (a) Interferometric autocorrelation of the pulses obtained at a pump power of 1.21 W. A pulse duration of 79 fs was estimated by adding a quadratic spectral phase of 1100 fs2 to the spectral intensity, giving a calculated autocorrelation envelope (red) that matched the experimental measurement (blue). The corresponding transform-limited pulse duration was 51 fs. (b) Spectrum, centered at 802 nm and with a spectral bandwidth of 16.4 nm.
Fig. 8. Main contributions to the single-pass GDD of the laser resonator. The GDD of the GTI mirror is quoted from the vendor and the material GDDs are calculated from the Sellmeier equations for fused silica [29] and sapphire [30]. The total single pass GDD is −200fs2 at 802 nm.
In Fig. 8 we present the group-delay dispersion (GDD) of the laser resonator in single-pass, indicating that the main contributing elements to the net GDD are the Ti:sapphire crystal, the fused-silica wedge and the GTI mirror. Contributions from the high-reflectivity mirrors according to the manufacturer are small, since their design is a simple Bragg stack and the pulse bandwidth occupies only a limited region of the central bandwidth of these mirrors. In a similar laser system modelocked with a conventional SESAM [26], the pulse duration was also limited to around 57 fs. Figure 8 shows that this is determined primarily by the properties of the GTI mirror, which has a steep gradient in its GDD at wavelengths above ∼810 nm, also explaining the steeper long-wavelength edge of the laser spectrum. The measured pulse duration of 51 fs is therefore attributed to the cavity dispersion profile and not to the SWCNT properties.
5. Indirect estimation of the saturable loss of the SWCNT-SA
As can be seen in Fig. 4, more power was extracted from the laser when it was modelocked than when it was running in continuous-wave operation, which is to be expected when a saturable absorber is present. Output power data obtained during cw (Fig. 4, blue) and modelocked (Fig. 4, red) operation were fitted to a simple rate-equation model to estimate the saturable absorption of the SWCNT-SA (Fig. 4, square symbols). According to [31], the output-coupled power from a standing-wave laser can be written as:
To obtain the saturable loss we performed four simultaneous fits to the data in Fig. 4, using free parameters of the pump spot radii, values of ${\delta _0}$ for cw and modelocked operation, and the output coupling, ${\delta _e}$, yielding a saturable loss of 0.1%. We also applied this approach to data (Fig. 4, circles) obtained from our laser when modelocked using the commercial SESAM employed in [26], returning a saturable loss of 0.65%, which is consistent with the upper limit of ∼2% set by the measured reflectivity spectrum (Fig. 2). The analysis also returned a non-saturable loss of the SESAM which was 0.3% greater than that of the SWCNT-SA. The low saturable loss makes the SWCNT-SA most useful for modelocking low-gain systems operated with small output coupling, such as the laser we present here. The non-saturable loss of the SWCNT-SA is determined by the laser mirror and is around 0.1% (see Fig. 2), making the total loss of the SWCNT-SA very low.
6. Conclusions
By fabricating a saturable absorber from spin-coated SWCNTs with diameters from 1.2–1.7 nm we have demonstrated the first diode-pumped Ti:sapphire laser to be modelocked using neither KLM nor a SESAM device. Modelocked operation has been achieved at pump powers as low as 1.01 W while still maintaining a high level of stability. Pulses with durations of 51 fs were supported by the SWCNT saturable absorber, in which low-density entanglement of the nanotubes is thought to broaden the S22 absorption feature responsible for saturable absorption near 800 nm. At these low entanglement densities, the contribution to non-saturable losses is likely to be lower than conventional SESAMs, leading to an overall higher cavity finesse.
In conclusion, we consider here the potential advantages of a SWCNT modelocking element in the context of diode-pumped Ti:sapphire lasers. Firstly, the poorer pump beam quality in diode-pumping (M2∼5–10) makes Kerr-lens modelocking more difficult and incurs a loss of power due to the need to detune the cavity from the stability region providing the best cw power [5]. Historically many diode-pumped Ti:sapphire lasers therefore used saturable absorber elements, but the limited commercial availability has meant that these have either been proprietary devices [5,6,33] or sources from a single commercial supplier [4,9,26,34]. Given that direct diode pumping could facilitate low-cost, low-power ‘commodity’ Ti:sapphire femtosecond oscillators, the lower non-saturable loss exhibited by SWCNT saturable-absorber elements is potentially important in such lasers. SWCNT saturable absorbers could also simplify such Ti:sapphire laser designs, for example by providing a dual modelocking and output coupling element by depositing SWCNTs onto a pre-existing dielectric output coupler. Similarly, a considerably wider operating bandwidth can be achieved using a SWCNT saturable absorber element than with a typical SESAM by depositing SWCNTs onto a broadband dielectric reflector. Although not a focus of this paper, SWCNTs should therefore enable extremely short pulses to be generated from a low-power diode-pumped Ti:sapphire laser. Finally, like all previously reported CNT modelocking devices, a susceptibility to optical damage was observed, implying that an appropriate over-coating process of the SWCNT may be needed to improve long-term damage resistance.
Funding
Engineering and Physical Sciences Research Council (EP/N002547/1).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. P. Roth, A. J. Maclean, D. Burns, and A. J. Kemp, “Directly diode-laser-pumped Ti:sapphire laser,” Opt. Lett. 34(21), 3334–3336 (2009). [CrossRef]
2. M. Murayama, Y. Nakayama, K. Yamazaki, Y. Hoshina, H. Watanabe, N. Fuutagawa, H. Kawanishi, T. Uemura, and H. Narui, “Watt-class green (530 nm) and blue (465 nm) laser diodes,” Phys. Status Solidi A 215(10), 1700513 (2018). [CrossRef]
3. P. F. Moulton, J. G. Cederberg, K. T. Stevens, G. Foundos, M. Koselja, and J. Preclikova, “Optimized InGaN-diode pumping of Ti:sapphire crystals,” Opt. Mater. Express 9(5), 2131 (2019). [CrossRef]
4. S. Sawai, A. Hosaka, H. Kawauchi, K. Hirosawa, and F. Kannari, “Demonstration of a Ti:sapphire mode-locked laser pumped directly with a green diode laser,” Appl. Phys. Express 7(2), 022702 (2014). [CrossRef]
5. K. Gürel, V. J. Wittwer, M. Hoffmann, C. J. Saraceno, S. Hakobyan, B. Resan, A. Rohrbacher, K. Weingarten, S. Schilt, and T. Südemeyer, “Green-diode-pumped femtosecond Ti:sapphire laser with up to 450 mW average power,” Opt. Lett. 23(23), 30043–30047 (2015). [CrossRef]
6. A. Muti, A. Kocabas, and A. Sennaroglu, “5-nJ Femtosecond Ti3+:sapphire laser pumped with a single 1 W green diode,” Laser Phys. Lett. 15(5), 055302 (2018). [CrossRef]
7. P. Roth, A. J. Maclean, D. Burns, and A. J. Kemp, “Direct diode-laser pumping of a mode-locked Ti:sapphire laser,” Opt. Lett. 36(2), 304–306 (2011). [CrossRef]
8. C. G. Durfee, T. Storz, J. Garlick, S. Hill, J. A. Squier, M. Kirchner, G. Taft, K. Shea, H. Kapteyn, M. Murnane, and S. Backus, “Direct diode-pumped Kerr-lens mode-locked Ti:sapphire laser,” Opt. Express 20(13), 13677–13683 (2012). [CrossRef]
9. J. C. E. Coyle, A. J. Kemp, J.-M. Hopkins, and A. A. Lagatsky, “Ultrafast diode-pumped Ti:sapphire laser with broad tunability,” Opt. Express 26(6), 6826 (2018). [CrossRef]
10. M. Lai, “Self-starting, self-mode-locked Ti:sapphire laser,” Opt. Lett. 19(10), 722–724 (1994). [CrossRef]
11. S. Backus, M. Kirchner, C. Durfee, M. Murnane, and H. Kapteyn, “Direct diode-pumped Kerr Lens 13 fs Ti:sapphire ultrafast oscillator using a single blue laser diode,” Opt. Express 25(11), 12469–12477 (2017). [CrossRef]
12. E. P. Ippen, “Principles of passive mode locking,” Appl. Phys. B 58(3), 159–170 (1994). [CrossRef]
13. U. Keller, K. J. Weingarten, F. X. Kärtner, D. Kopf, B. Braun, I. D. Jung, R. Fluck, C. Hönninger, N. Matuschek, and J. Aus der Au, “Semiconductor saturable absorber mirrors (SESAM’s) for femtosecond to nanosecond pulse generation in solid-state lasers,” IEEE. J. Sel. Top. Quantum Electron 2(3), 435–453 (1996). [CrossRef]
14. S. Y. Set, H. Yaguchi, Y. Tanaka, and M. Jablonski, “Laser mode locking using a saturable absorber incorporating carbon nanotubes,” J. Lightwave Technol. 22(1), 51–56 (2004). [CrossRef]
15. D. V. Khudyakov, A. S. Lobach, and V. A. Nadtochenko, “Passive mode locking in a Ti:sapphire laser using a single-walled carbon nanotube saturable absorber at a wavelength of 810 nm,” Opt. Lett. 35(16), 2675–2677 (2010). [CrossRef]
16. I. H. Baek, S. Y. Choi, H. W. Lee, W. B. Cho, V. Petrov, A. Agnesi, V. Pasiskevicius, D. Yeom, K. Kim, and F. Rotermund, “Single-walled carbon nanotube saturable absorber assisted high-power mode-locking of a Ti:sapphire laser,” Opt. Express 19(8), 7833–7838 (2011). [CrossRef]
17. T. R. Schibli, K. Minoshima, H. Kataura, E. Itoga, N. Minami, S. Kazaoui, K. Miyashita, M. Tokumoto, and Y. Sakakibara, “Ultrashort pulse-generation by saturable absorber mirrors based on polymer-embedded carbon nanotubes,” Opt. Express 13(20), 8025–8031 (2005). [CrossRef]
18. A. Agnesi, A. Greborio, F. Pirzio, G. Reali, S. Y. Choi, F. Rotermund, U. Griebner, and V. Petrov, “99fs Nd:Glass laser mode-locked with carbon nanotube saturable absorber mirror,” Appl. Phys. Express 3(11), 112702 (2010). [CrossRef]
19. K. H. Fong, K. Kikuchi, C. S. Goh, S. Y. Set, R. Grange, M. Haiml, A. Schlatter, and U. Keller, “Solid-state Er:Yb:glass laser mode-locked by using single-walled carbon nanotube thin film,” Opt. Lett. 32(1), 38–40 (2007). [CrossRef]
20. C. Li, W. Cai, J. Liu, L. Su, D. Jiang, F. Ma, Q. Zhang, J. Xu, and Y. Wang, “Single-walled carbon nanotube saturable absorber for a diode-pumped passively mode-locked Nd,Y:SrF2 laser,” Opt Comm. 372, 76–79 (2016). [CrossRef]
21. G. Tanısalı, I. Baylam, M. Tascı, J. E. Bae, F. Rotermund, U. Demırbas, and A. Sennaroglu, “21 fs Cr:LiSAF laser mode locked with a single-walled carbon nanotube saturable absorber,” Opt. Lett. 44(19), 4662–4665 (2019). [CrossRef]
22. “Technical Data Sheet: IsoNanotubes-M (Metallic SWNT), IsoNanotubes-S (Semiconducting SWNTs) and PureTubes (Ultra High Purity SWNTs)”, NanoIntegris, 3765 La Vérendrye, Boisbriand, QC J7H 1R8, Canada.
23. H. Kataura, Y. Kumazawa, Y. Maniwa, I. Umezu, S. Suzuki, Y. Ohtsuka, and Y. Achiba, “Optical properties of single-wall carbon nanotubes,” Synth. Met. 103(1-3), 2555–2558 (1999). [CrossRef]
24. C. Saltiel, S. Manickavasagam, M. Pinar Mengüc, and R. Andrews, “Light-scattering and dispersion behavior of multiwalled carbon nanotubes,” J. Opt. Soc. Am. A 22(8), 1546–1554 (2005). [CrossRef]
25. A. H. Quarterman, “Femtosecond pulse generation in surface-emitting semiconductor lasers,” University of Southampton, School of Physics & Astronomy, Doctoral Thesis, 139pp. (2009).
26. P. Castro-Marin, T. Mitchell, J. Sun, and D. T. Reid, “Characterization of a carrier-envelope-offset-stabilized blue- and green-diode-pumped Ti:sapphire frequency comb,” Opt. Lett. 44(21), 5270–5273 (2019). [CrossRef]
27. The original manufacturer of the M462 laser diode is not known, however the diode itself is easily found from third-party suppliers using only this designation, e.g. https://sites.google.com/site/dtrslasershop/.
28. F. Rotermund, W. B. Cho, S. Y. Choi, I. H. Baek, J. H. Yim, S. Lee, A. Schmidt, G. Steinmeyer, U. Griebner, D.-I. Yeom, K. Kim, and V. Petrov, “Mode-locking of a solid-state lasers by single-walled carbon-nanotube based saturable absorbers,” Quantum Electron. 42(8), 663–670 (2012). [CrossRef]
29. I. H. Malitson, “Interspecimen Comparison of the Refractive Index of Fused Silica,” J. Opt. Soc. Am. 55(10), 1205 (1965). [CrossRef]
30. I. H. Malitson, “Refraction and Dispersion of Synthetic Sapphire,” J. Opt. Soc. Am. 52(12), 1377–1379 (1962). [CrossRef]
31. A. E. Siegman, Lasers (University Science Books, 1986).
32. P. F. Moulton, “Spectroscopic and laser characteristics of Ti:Al2O3,” J. Opt. Soc. Am. B 3(1), 125 (1986). [CrossRef]
33. A. Rohrbacher, O. E. Olarte, V. Villamaina, P. Loza-Alvarez, and B. Resan, “Multiphoton imaging with blue-diode-pumped SESAM-modelocked Ti:sapphire oscillator generating 5 nJ 82 fs pulses,” Opt. Express 25(9), 10677 (2017). [CrossRef]
34. N. Sugiyama, H. Tanaka, and F. Kannari, “Mode-locked Ti:sapphire laser oscillators pumped by wavelength-multiplexed laser diodes,” Jpn. J. Appl. Phys. 57(5), 052701 (2018). [CrossRef]
