Open Access
Add to BibSonomy
Abstract
Ge core fibers fixed on a SiO2 rod were annealed by a 360° axially symmetric distribution CO2 laser of the different power, respectively. The residual stress distribution, crystalline properties, and optical loss of the Ge cores were investigated. The non-uniform distribution of Raman frequencies at the Ge core cross-section were found after the fiber was annealed, which means that CO2 laser irradiation is one of the key factors determining the uniformity of the Ge core annealed by a CO2 laser. The bonding state between the Ge core and SiO2 cladding was analyzed according to the Raman mapping. Compared with the Si core fiber, there are fewer covalence bonds between the core and the SiO2 cladding in the Ge core fiber. For the Ge core fiber annealed at relatively high CO2 laser power, the tensile stresses transformed to compressive stresses in some areas of the Ge core cross-section, and the splitting of (111) plane X-ray diffraction peak appeared in the X-ray diffraction spectrum. The optical loss measured using a quantum cascade laser with the wavelength range from 4.7 μm to 4.9 μm shows the lowest optical loss of 2.05 dB/cm has been achieved in all samples.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Fabrication of semiconductor core fibers is motivated by the advantageous photonic and electronic properties of crystalline semiconductors for many applications [1,2]. Ge is a semiconductor material with high Raman gain and a low loss window from 2 to 14 μm [3], which manifests it to be a very intriguing core material for new infrared transmission optical fibers and fiber-based light sources [4–7]. The smaller energy difference between its indirect and direct band gap allows the indirect-to-direct band gap transition under strain, leading to more efficient light emission [8–10], and has been demonstrated as a direct band gap laser medium [11]. Furthermore, it promises to be a useful platform for photonics and optoelectronics in fiber geometry [10].
Many attempts to fabricate high quality Ge core fibers have been carried out by far. Ballato [5] reported that the glass-clad highly crystalline Ge core optical fiber was drawn by molten core drawing (MCD). Another study by the same group [7] indicated that the silica-clad optical Ge core fiber drawn using a molten core technique features low propagation loss of 0.7 dB/cm at 3.39 μm. Mehta [12] reported that Ge optical fibers fabricated using a high pressure chemical deposition (HPCVD) technique, exhibiting the optical loss of 4.8 dB/cm at 10.6 μm. Mcmillen [13] firstly studied the influence of tapering on the Ge core crystallography. Chaudhuri [14] studied that the incorporation of more hydrogen leads to further passivation of the Ge dangling bonds and a decrease in the number of scattering centers in the fiber core, lowering the optical loss. Moreover, Mustafa Ordu [15] studied on the mid-infrared transmission in Ge core borosilicate glass fibers; the loss of such fiber is 3.1-9.1dB/cm at 5.82-6.28μm wavelength range. However, the performance of Ge core fiber still does not meet the commercial application requirements.
Laser annealing is an effective way to increase the performance of semiconductor core fiber. At present, most of the reports on laser annealing are about Si core fiber [16–18], there are few reports on laser annealing of Ge core fiber. Ji [10] reported that the fabrication of the first small core diameter (5.6 μm) single-crystal Ge fibers, up to ≈9 mm long, and with optical loss down to 1.33 dB/cm at 2 μm by precisely scanning a continuous wave argon ion laser along the amorphous Ge core fibers deposited inside silica capillary fibers using HPCVD. It demonstrates that such single crystal Ge fibers with low optical loss can be obtained by laser annealing.
Different the furnace annealing with uniform temperature field, the temperature field distribution of the fiber during the process of laser annealing may be vice versa. The residual stress distribution and crystalline state could be affected by such thermal non-uniformity. Both residual stress distribution and the crystalline state are important characteristics of semiconductor core fiber, because they can affect the width of bandgap, and the optical propagation properties [11]. However, there are few reports about the residual stress distribution and crystalline state of the semiconductor core after annealed by laser, especially for Ge core fiber.
In this study, by examining the CO2 laser annealed Ge core fiber drawn by MCD method, the residual stress distribution at the Ge core cross-section, crystalline state, and optical loss of the Ge core fibers against annealing laser power were investigated. Based on the previous study on Si core fibers [19], the bonding state between the Ge core and SiO2 cladding was analyzed. We aim to further improve the quality of Ge core fiber by laser annealing and provide a general reference for the laser processing of Ge core fiber devices in future.
2. Experiments and measurements
The Ge core fiber was fabricated using MCD method. Ge bulk crystal with purity of 99.99% was packed into a silica tube with an inner diameter of about 3 mm and outer diameter of 10 mm. The silica tube had been pre-drawn in order to seal one end of the tube, so that the Ge core which will be molten during the fiber drawing process was confined. The fiber was drawn using a drawing tower at approximately 2000 °C which is above the melting point of Ge, and the molten Ge was encapsulated by the viscous silica cladding. The fabricated fiber has a diameter of 200–220 μm, which yielded a core size of 30–60 μm.
Laser annealing was performed using a CO2 laser (SYNRAD Firestar series i401). The schematic diagram of fiber annealing apparatus is shown in Fig. 1(a). Laser beam with a diameter of 13 mm was extended to 26 mm by a beam expender, and then entered into a device composed by bronze mirrors (Fig. 1 (c)). The device is similar to the apparatus of the laser-heated pedestal growth (LHPG) method [20,21]. In order to enhance the mechanical stability of the fiber during the process of annealing, the Ge core fiber was fixed on a quartz rod with the diameter of 4 mm. The schematic diagram is shown in Fig. 1 (b) and (e). The laser beam irradiated onto the quartz rod in an approximate 360° axially symmetric distribution. The symmetric irradiance prevents uniform laser power relative to one-beam. The quartz rod was located in the centre of the 360° axially symmetric distribution laser as shown in Fig. 1 (d). Although the quartz rod partially blocked the laser irradiation area during annealing, it is still superior to the one-beam laser in terms of uniformity of the fiber temperature field. The quartz rod moved along the axial direction by a motorized translational stage during the annealing process at a velocity of 3 mm/s.
Fig. 1 (a) The schematic diagram of the fiber annealing apparatus, (b) the schematic diagram of the laser irradiating on rod, the arrow refers to the 3D illustration of bronze mirrors, (c) photographs of various parts of the bronze mirrors, (d) optical paths of CO2 laser in a cross-sectional diagram, (e) the fiber attached longitudinally to the side of the quartz rod, both ends of the fiber are fixed with heat-resistant tape.
The annealing laser power were precisely adjusted by the laser controller (SYNRAD UC-2000), and measured by a commercial optical power meter (Newport 1918-R model). The annealing laser power was set to 10%, 15%, 20%, and 25% of the peak power of the laser, respectively. The measured value was about 48 W, 67 W, 102 W, and 129 W, respectively, as shown in Fig. 2. The laser irradiated on the quartz rod, and produced a uniform irradiation field. The laser speckle, scattering and some other effect were not found.
The 360° axially symmetric laser beam irradiated on the quartz rod. The axis of the quartz rod was at the centre of the 360° axially symmetric distribution laser. Since the fiber direction is consistent with the axial direction of the rod, it absorbs part of the laser energy like a short edge of the rod does. And because the laser is uniformly symmetrically distributed, the short edge blocked by the fiber evenly separates the power. The ratio of the short edge to the circumference of the quartz can be calculated. We estimated roughly the laser power irradiated on the Ge core fiber according to the central angle of the quartz rod axis. The laser which irradiated on the fiber is in the central angle range covered by fiber, as shown in Fig. 3. This central angle is defined as 2θ.
The laser power irradiated on the Ge core fiber was estimated roughly as follow:
where Rfiber is the radius of the Ge core fiber, Rrod is the radius of the quartz rod, Ptotal is the total power of the annealing laser. If Rfiber is 110 μm, Rrod is 2 mm, the estimated laser power irradiated on the Ge core fiber is 0.78 W, 1.09 W, 1.66 W, and 2.1 W, respectively, as shown in Fig. 4.
Raman mapping was collected at the cross-section of Ge core at room temperature in the backscattering configuration using a LabRam HR Evolution Raman spectrometer from Horiba Jobin Yvon equipped with a Peltier-cooled CCD detector, an 1800 lines/mm grating, a He-Ne laser with an excitation wavelength of 633 nm, an optical microscope and a motorized X-Y stage. The incident laser light is focused onto the sample’s surface through a 50x objective. The Raman spectrum of a Ge bulk crystal was measured as a reference. The spectra were fitted by a Voigt function.
Micro-region XRD analysis was used to study the crystallization state of the Ge core. The cladding of the Ge core fiber was removed by hydrofluoric acid, and the side of the Ge core was measured, which used an 18 KW D/MAX2500V+/PC X-ray diffractometer with CuKα1 radiation (λ = 1.5418 Å) and a solid-state Ge detector. Diffraction patterns were collected in 0.02° steps from 20° to 80° in 2θ. The spot diameter of the X-ray is about 50μm. Because the Ge core diameter is small, in order to better collect the XRD spectrum, we have adopted the following method for every sample: Firstly, the cladding of the Ge core fiber was removed by Hydrofluoric acid, and the core was cut into several short cores. Then the cores were tightly arranged side by side without gaps between them, and investigated by XRD measurement.
Scanning electron microscope (SEM) and energy dispersive X-ray spectroscopy (EDS) characterization: The cross-section of the fiber annealed at the laser power of 20% was imaged using a HITACHI SU-1510 SEM. The electron microscope was operated at 10 mm working distance. EDS was used for elemental analysis to examine the distribution of elements across the core/clad interface. Elemental compositions were measured at several locations traversing the cross-section of the core, and the distribution of Ge, Si and O elements were examined.
The Optical loss of the Ge core fiber was measured using the cutback technique. Before the measurement, the fiber was mounted in a larger ceramic capillary and polished. The schematic diagrams are shown in Fig. 5 (a) and (b). The measurement was undertaken using a quantum cascade laser (Thorlabs QF4800CM1) with the wavelength range from 4.7 μm to 4.9 μm as light source. The spectrum of the light source (from spec sheet of THORLABS QF4800CM1) is shown in Fig. 5 (c). The laser was coupled into the fiber in free space using a black diamond-2 infrared aspheric collimating lens, and collected by an optical power detector (THORLABS S401), then measured by dynamometer. The measurement was performed on section of fiber that was 2-4 cm in length. A 2 mm section of fiber was removed for each cut-back measurement and this was repeated four times for each fiber to ensure the integrity of the measurement.
Fig. 5 (a) The schematic diagrams of the fiber optical loss measurement, (b) the detail of the device for placing fiber in the measurement, (c) the spectrum of the light source (from spec sheet of THORLABS QF4800CM1).
3. Experimental results and discussion
To determine the residual stress distribution and the crystalline state of the Ge core, Raman frequency maps and full width at half maximum (FWHM) values of Raman peaks maps were carried out at the cross-sections of the Ge core fibers. Raman spectrum of a point at the Ge core cross-section of as-drawn fiber is shown in Fig. 6(a). The peak position and the FWHM of the spectrum are obtained by selecting the fitting function through the instrument software (LabSpec 6). On Raman frequency maps, the different frequencies are represented by the different colors. For more obvious color contrast in the area of Ge core, the colors of SiO2 cladding are set to bright yellow. On FWHM values maps, the FWHM values of the SiO2 cladding are very large for the reason that the cladding is amorphous structure. They are set to deep blue on the FWHM value maps. It makes the color contrast in the area of Ge core are more obvious.
Fig. 6 (a) Raman spectrum of a point at Ge core cross-sections of as-drawn fiber, the arrow indicates the fitted peak position and fitted FWHM data. The inserted graph is an unfitted spectrum. (b) Raman frequency maps of the as-drawn Ge core fiber, (c) Raman frequency maps of the fiber annealed at the laser power of 10%, (d) Raman frequency maps of the fiber annealed at the laser power of 15%, (e) Raman frequency maps of the fiber annealed at the laser power of 20%, (f) Raman frequency maps of the fiber annealed at the laser power of 25%.
The Raman frequency maps and the related statistical values at the Ge core cross-sections are shown in Fig. 6 and Table. 1. In Table. 1, Smean is the average value of the Raman frequencies; Srange is the range of the Raman frequencies; Sq is the standard deviation of the Raman frequencies, it reflects the uniformity of the Raman frequency distribution. The box diagram of the Raman frequency distribution is shown in Fig. 7. The cross (“×”) is outlier data, Most of the outliers are Raman frequencies of the areas near the Ge core boundary. The triangle (“△”) denotes the average value.
Table 1. The statistical values of Raman frequencies at the Ge core cross-sections
All the Ge cores and the Ge bulk crystal exhibit a strong peak around 301 cm−1. The Raman frequency of the Ge bulk crystal is 301.20 cm−1. For the as-drawn Ge core fiber, the Srange is 299.95-300.91 cm−1, the Smean is 300.50 cm−1; and the Sq is 0.098 cm−1. The Raman frequencies show that there are residual tensile stresses in the as-drawn Ge core. There is an annular thin layer which has low wavenumbers about 2 μm range near SiO2 cladding on the Raman frequency map of the as-drawn Ge core. The Raman frequencies of this thin layer are 0.2~0.6 cm−1 lower than the Raman frequencies of the interior area. This thin layer reflects the effect of SiO2 cladding on the Ge core during the fiber drawing process. The Smean of the Ge core annealed at the laser power of 10% shifts to high wavenumbers relative to that of the as-drawn Ge core. And the Smean increases with the further increasing of the laser power. The Smean of the Ge core annealed at the laser power of 25% is 301.20 cm−1. The phonon Raman shift occurs when the stress changes [22–24], because the phonon vibration frequency is related to the stress in the material. Hence Smean increasing and closing to the Raman frequency of Ge bulk crystal reflects the release of the residual tensile stresses in the Ge core after laser annealing. Raman frequency maps also clearly show that the residual tensile stresses in the Ge core decrease with the increasing of the laser power.
For all samples, Raman spectra of 3 different positions of the core side were taken through the cladding, which are shown in Fig. 8. The schematic diagram is as shown in Fig. 8 (a). Table 2 is the Raman frequencies obtained by spectral fitting of these points. It can be seen that the changing trends of Raman frequencies are consistent with those of Raman frequency maps.
Fig. 8 Raman spectra of 3 different positions of Ge core side. (a) The schematic diagram of the Raman measurement performed on the side of Ge core, (b) Raman spectra taken from position 1, (c) Raman spectra taken from position 2, (d) Raman spectra taken from position 3.
Table 2. Raman frequencies at the interface of Ge core side
The big differences in characteristics between core and cladding, and the relatively rapid cooling rate during fiber drawing introduced the high density of defects (such as dislocations) into Ge core, which caused relatively high residual stresses. These defects can be removed by CO2 laser annealing, and the reduction of the defects is more obvious in annealing of higher laser power, for the reason that Ge atoms gain more kinetic energy. However, there is an appropriate annealing laser power range for the decreasing of defects in Ge core. Exorbitant laser power may reduce the performance of the Ge core. For the fiber annealed at the laser power of 25%, more areas which have Raman frequencies higher than 301.2 cm−1 can be observed on the Raman frequency map. It means that more compressive stresses appeared in the Ge core annealed at this laser power. Previous study [25] shows the effect of the difference between crystalline Si density and liquid Si density on the residual stress in Si core during CO2 laser processing. Similar with Si, the density of liquid Ge is higher than the density of crystalline Ge. The density of liquid Ge near 1000 °C is close to 5.557 g/cm3, while that of crystalline Ge at 25 °C is close to 5.323 g/cm3. In our study, it most likely that the more obvious phase change occurred in Ge core during the annealing process at the laser power of 25%, the phase change during the annealing process and the relatively fast cooling rate after annealing caused the transition from tensile stress to compressive stress in some areas of Ge core.
Form the values of Sq, the Raman frequency distribution of the fiber annealed at the laser power of 15% is the most uniform. However, different colors can be observed between the upper left and the lower right of Ge core cross-section on the Raman frequency map. Similar phenomenon is more obvious for the Ge core annealed at the laser power of 20%. It reflects the non-uniform distribution of the residual stresses in the Ge core after the fiber being annealed by laser. Although the laser of 360° axially symmetric distribution has better uniformity than the single beam laser (in our experiment, the quartz rod blocked one part of laser.), the Raman frequencies difference between the two sides of the Ge core cross-section are still being observed. The Raman frequency maps show that CO2 laser irradiation is one of the key factors determining the uniform of the Ge core after laser annealing.
After the Ge core being annealed at the laser power of 10%, the Raman frequencies of the annular thin layer near SiO2 cladding have blue shift. With the increasing of the laser power, the annular thin layer gradually becomes inconspicuous, even disappears in some areas. It means that the annealing of appropriate CO2 laser power effectively reduces the defect concentration in the areas near cladding, thereby, the residual stresses were decreased.
In our previous study on Si core fiber [19], the Raman frequencies of the Si core cross-section are ring form distribution, and the frequency step can be observed on the Raman frequency maps of Si core cross-section. The low Raman frequency annular thin layer about 3 μm range is near SiO2 cladding. Moreover, this annular thin layer still exists after the Si core being annealed. Different with the Si core fiber, the ring form distribution of Raman frequencies of the Ge core cross-section is unobserved, and the annular thin layer near cladding is unobvious relative to that of the Si core fiber. This annular thin layer becomes inconspicuous after the Ge core being annealed at the appropriate laser power. For the Si core fiber, the oxygen plays a strong role at the interface between Si and SiO2. It is most likely that Si-O is easily formed at the junction between Si core and SiO2 cladding during the fiber drawing process, which caused by the diffusion of oxygen. Therefore, more covalent bonds were formed between Si core and SiO2 cladding. The different characteristics between SiO2 cladding and Si core introduced the residual stresses into Si core through these covalent bonds. These residual stresses mainly distribute in the areas near cladding, and cannot be eliminated via annealing.
For the Ge core fiber, there are Ge, Si, O elements in the system, the system energy is calculated as follow [26,27]:
The formations of Ge-O-Ge and Si-O-Ge will dramatically increase the system energy as compared to the formation of Si-O-Si [27]. According to the lowest energy principle, the O atoms tend toward form covalence bonds with Si atoms, which is low energy state. The Ge atoms prefer to stay away from the oxide region, and the Ge/SiO2 interface has fewer Ge-O bonds, which means that there are fewer covalence bonds between Ge core and SiO2 cladding relative to that of Si core fiber. Therefore, the effect of SiO2 cladding on Ge core is unobvious relative to that of Si core, despite the thermal expansion coefficient of Ge is larger than the thermal expansion coefficient of Si. However, more Ge crystal nuclei were formed at the inner surface of SiO2 cladding than the inner of Ge core during the Ge core crystallization process in fiber drawing, which means that there are more defects and more residual stresses in the core areas near SiO2 cladding. These defects and residual stresses can be decreased by annealing. Therefore, the low Raman frequency annular thin layer near SiO2 cladding is sawn on the Raman frequency map of the as-drawn Ge core fiber, and becomes unobvious after the fiber being annealed at the appropriate laser power.FWHM value maps and related statistical values at the cross-section of Ge cores are shown in Fig. 9 and Table. 3. In Table. 3, Smean is the average value of the FWHM; Srange is the range of the FWHM; Sq is the standard deviation of the FWHM, it reflects the uniformity of the FWHM value distribution. The box diagram of the FWHM value distribution is shown in Fig. 10. The cross (“×”) is outlier data, the triangle (“△”) is the average value.
Fig. 9 Raman FWHM value maps of the Ge core fibers annealed at the different power of laser, (a) as-drawn, (b) annealed at the laser power of 10%, (c) annealed at the laser power of 15%, (d) annealed at the laser power of 20%, (e) annealed at the laser power of 25%.
Table 3. The statistical values of Raman peak FWHM at Ge core cross-sections
The FWHM of the Ge bulk crystal is 2.54 cm−1. For all Ge core fibers, the Smean is in the range of 2.89-3.08 cm−1, no obvious changing trend was observed for the Ge cores annealed at the different laser power. There is no significant crystalline state change which can cause the obvious change of Raman peak FWHM after the CO2 laser annealing. The areas which have large FWHM values are near the cladding, and form a narrow ring area on the FWHM value maps, for the reason that the effect of SiO2 cladding on the Ge core. In spite of fewer covalence bonds were formed between Ge core and SiO2 cladding relative to that of Si core fiber.
The XRD spectra of the Ge cores are shown in Fig. 11. For the as-drawn Ge core fiber, the Ge core fibers annealed at the laser power of 10%, and the laser power of 15%, the diffraction peaks of the (220) planes were observed, as shown by the black line, blue line, and yellow line in Fig. 11. Slight shifts in the 2θ values between the peak positions of these samples and this indexed in the Joint Committee on Powder Diffraction Standards (JCPDS) database(45.31°), which is attribute to the residual tensile stresses in the Ge core. For these 3 samples, the changing trend of diffraction peak positions reflects that the residual tensile stresses decreased with the increasing of the laser power. After annealing at the laser power of 20%, the diffraction peak (72°) of (331) planes was observed, and the 2θ value is very close to the JCPDS database (72.80°), which proves that the laser annealing at this condition decreased the residual stresses introduced by defects effectively. Further increasing the laser power to 25%, the diffraction peak of (111) planes and the diffraction peak of (311) planes are generated, the 2θ value of the (331) planes diffraction peak (56.11°) is larger than JCPDS database (53.68°), which reflects the existence of the compressive stresses in the Ge core.
For the fiber annealed at the laser power of 25%, there is a splitting of the X-ray diffraction peak of (111) planes. Twinning could cause the splitting of the X-ray diffraction peak. The (111) planes of the diamond structure crystal is prone to form stacking faults in the recrystallization annealing, which is an important cause of the twinning formation. The XRD result shows that this laser power is more likely to introduce structure defects into the Ge core relative to the laser power of 20%. The XRD measurement results are in good agreement with the Raman mapping results in reflecting the residual stresses in the Ge core.
The SEM image and the Elemental analysis of the cross-section of the Ge core fiber annealed at the laser power of 20% are shown in Fig. 12. The core size is about 30 μm. The core is circular and there is a strong contrast between the Ge core and the SiO2 cladding. The Ge concentration of the core is almost 100% with negligible O and Si, and the interface is remarkably well-defined. The Ge concentration in cladding is almost 0. There are no obvious discontinuities at the core/cladding interface or obvious cracks or signs of bubbles in the core.
Fig. 12 The cross-section image and the elemental analysis of the Ge core fiber annealed at the laser power of 20%.
The optical losses of the Ge core fibers are shown in Fig. 13. The relatively high optical loss of SiO2 cladding at the wavelength range from 4.7 μm to 4.9 μm ensured that the light transmitted in the Ge core. For the as-drawn Ge core fiber, the measured optical loss is dB/cm. The main reasons for the high losses are most likely attributed to the defects of the Ge core interface and the grain boundaries in the Ge core. After annealing at the laser power of 10%, the optical loss decreased to dB/cm. The optical losses of the fibers decrease with the increasing of the laser power until the laser power of 20%. The optical loss of the fiber annealed at the laser power of 20% is dB/cm, which is the lowest optical loss in all samples. When the laser power reached to 25%, the optical loss increased to dB /cm. According to the results of Raman mapping and the XRD spectra, the laser annealing under this laser power condition introduced more structure defects into the Ge core than that of the laser power of 20%. The structure defects increase the optical loss of the fiber.
4. Conclusions
In summary, Ge core fiber was fabricated using MCD method, and then annealed by CO2 laser. The residual stress distribution at the Ge core cross-section, the structural properties, and the optical losses of Ge cores annealed at the different laser power were mainly investigated. CO2 laser annealing of appropriate laser power could decrease the residual stresses and the optical loss of the Ge core fiber effectively. The residual stress distribution at the Ge core cross-section is relative to the way of laser irradiation. The obvious bonding phenomenon between Ge core and SiO2 cladding has not been found.
Funding
National Natural Science Foundation of China (NSFC) (61505103, 61475096).
References
1. S. Chaudhuri, J. R. Sparks, X. Ji, M. Krishnamurthi, L. Shen, N. Healy, A. C. Peacock, V. Gopalan, and J. V. Badding, “Crystalline Silicon Optical Fibers with Low Optical Loss,” ACS Photonics 3(3), 378–384 (2016). [CrossRef]
2. N. Healy, S. Mailis, N. M. Bulgakova, P. J. A. Sazio, T. D. Day, J. R. Sparks, H. Y. Cheng, J. V. Badding, and A. C. Peacock, “Extreme electronic bandgap modification in laser-crystallized silicon optical fibres,” Nat. Mater. 13(12), 1122–1127 (2014). [CrossRef] [PubMed]
3. R. Soref, “Mid-infrared photonics in silicon and germanium,” Nat. Photonics 4(8), 495–497 (2010). [CrossRef]
4. J. Parker, D. Feldman, and M. Ashkin, “Raman scattering by silicon and germanium,” Phys. Rev. 155(3), 712–714 (1967). [CrossRef]
5. J. Ballato, T. Hawkins, P. Foy, B. Yazgan-Kokuoz, R. Stolen, C. McMillen, N. K. Hon, B. Jalali, and R. Rice, “Glass-clad single-crystal germanium optical fiber,” Opt. Express 17(10), 8029–8035 (2009). [CrossRef] [PubMed]
6. J. Ballato, T. Hawkins, P. Foy, B. Yazgan-Kokuoz, C. McMillen, L. Burka, S. Morris, R. Stolen, and R. Rice, “Advancements in semiconductor core optical fiber,” Opt. Fiber Technol. 16(6), 399–408 (2010). [CrossRef]
7. J. Ballato, T. Hawkins, P. Foy, S. Morris, N. K. Hon, B. Jalali, and R. Rice, “Silica-clad crystalline germanium core optical fibers,” Opt. Lett. 36(5), 687–688 (2011). [CrossRef] [PubMed]
8. J. R. Jain, A. Hryciw, T. M. Baer, D. A. B. Miller, M. L. Brongersma, and R. T. Howe, “A micromachining-based technology for enhancing germanium light emission via tensile strain,” Nat. Photonics 6(6), 398–405 (2012). [CrossRef]
9. M. J. Süess, R. Geiger, R. A. Minamisawa, G. Schiefler, J. Frigerio, D. Chrastina, G. Isella, R. Spolenak, J. Faist, and H. Sigg, “Analysis of enhanced light emission from highly strained germanium microbridges,” Nat. Photonics 7(6), 466–472 (2013). [CrossRef]
10. X. Ji, R. L. Page, S. Chaudhuri, W. Liu, S. Yu, S. E. Mohney, J. V. Badding, and V. Gopalan, “Single-Crystal Germanium Core Optoelectronic Fibers,” Adv. Opt. Mater. 5(1), 100592 (2016).
11. J. Liu, X. Sun, R. Camacho-Aguilera, L. C. Kimerling, and J. Michel, “Ge-on-Si laser operating at room temperature,” Opt. Lett. 35(5), 679–681 (2010). [CrossRef] [PubMed]
12. P. Mehta, M. Krishnamurthi, N. Healy, N. F. Baril, J. R. Sparks, P. J. A. Sazio, V. Gopalan, J. V. Badding, and A. C. Peacock, “Mid-infrared transmission properties of amorphous germanium optical fibers,” Opt. Lett. 35(5), 679–681 (2010). [PubMed]
13. C. McMillen, G. Brambill, S. Morris, T. Hawkins, P. Foy, N. Broderick, E. Koukharenko, R. Rice, and J. Ballato, “On crystallographic orientation in crystal core optical fibers II: Effects of tapering,” Opt. Mater. 35(2), 93–96 (2012). [CrossRef]
14. S. Chaudhuri, S. Li, N. Healy, A. Peacock, and J. Badding, “Hydrogenated Amorphous Germanium Optical Fiber,” Adv. Photonics 10(9), 189–191 (2015).
15. M. Ordu, J. Guo, B. Tai, J. Bird, S. Ramachandran, and S. Basu, “Mid-infrared transmission through germanium-core borosilicate glass-clad semiconductor fibers,” Opt. Mater. Express 7(9), 3107–3115 (2017). [CrossRef]
16. N. Healy, S. Mailis, T. D. Day, P. J. Sazio, J. V. Badding, and A. C. Peacock, “Laser annealing of amorphous silicon core optical fibers,” in Advanced Photonics Congress (2012), Paper STu1D.1 (Optical Society of America, 2012), p. STu1D.1. [CrossRef]
17. N. Healy, M. Fokine, Y. Franz, T. Hawkins, M. Jones, J. Ballato, A. C. Peacock, and U. J. Gibson, “CO2 laser-induced directional recrystallization to produce single crystal silicon-core optical fibers with low loss,” Adv. Opt. Mater. 4(7), 1004–1008 (2016). [CrossRef]
18. X. Ji, S. Lei, S.-Y. Yu, H. Y. Cheng, W. Liu, N. Poilvert, Y. Xiong, I. Dabo, S. E. Mohney, J. V. Badding, and V. Gopalan, “Single-crystal silicon optical fiber by direct laser crystallization,” ACS Photonics 4(1), 85–92 (2017). [CrossRef]
19. Z. Zhao, F. Xue, Y. Mao, N. Chen, and T. Wang, “Effects of annealing on the residual stresses distribution and the structural properties of Si core fiber,” Opt. Fiber Technol. 41, 193–199 (2018). [CrossRef]
20. M. M. Fejer, J. L. Nightingale, G. A. Magel, and R. L. Byer, “Laser‐heated miniature pedestal growth apparatus for single‐crystal optical fibers,” Rev. Sci. Instrum. 55(11), 1791–1796 (1984). [CrossRef]
21. K. Y. Huang, K. Y. Hsu, D. Y. Jheng, W. J. Zhuo, P. Y. Chen, P. S. Yeh, and S. L. Huang, “Low-loss propagation in Cr4+:YAG double-clad crystal fiber fabricated by sapphire tube assisted CDLHPG technique,” Opt. Express 16(16), 12264–12271 (2008). [CrossRef] [PubMed]
22. N. Fukata, M. Mitome, T. Sekiguchi, Y. Bando, M. Kirkham, J. I. Hong, Z. L. Wang, and R. L. Snyder, “Characterization of impurity doping and stress in Si/Ge and Ge/Si core-shell nanowires,” ACS Nano 6(10), 8887–8895 (2012). [CrossRef] [PubMed]
23. C. E. Finlayson, A. Amezcua-Correa, P. J. A. Sazio, N. F. Baril, and J. V. Badding, “Electrical and Raman characterization of silicon and germanium-filled microstructured optical fibers,” Appl. Phys. Lett. 90(13), 132100 (2007). [CrossRef]
24. Z. Zhao, X. Cheng, F. Xue, T. He, and T. Wang, “Effect of Annealing Temperature on the Stress and Structural Properties of Germanium Core Fibre,” J. Cryst. Growth 473(1), 1–5 (2017).
25. M. Fokine, A. Theodosiou, S. Song, T. Hawkins, J. Ballato, K. Kalli, and U. J. Gibson, “Laser structuring, stress modification and Bragg grating inscription in silicon-core glass fibers,” Opt. Mater. Express 7(5), 1589–1597 (2017). [CrossRef]
26. W. Windl, T. Liang, S. Lopatin, and G. Duscher, “Investigation of Nanostructured Germanium/Silicon Dioxide Interfaces,” J. Comput. Theor. Nanosci. 1(3), 286–295 (2004). [CrossRef]
27. W. Windl, T. Liang, S. Lopatin, and G. Duscher, “Characterization and Modeling of Atomically Sharp Perfect Si:Ge/SiO2 Interfaces,” ECS Trans. 3(7), 539–549 (2006).
