Open Access
Add to BibSonomyAbstract
For the detection of an exo-planet on which signs of life are possible, one needs to develop single mode fibers working in the mid-IR. Telluride glasses are known for their large transparency in the mid and far infrared region, but the drawback is their high crystallization tendency during the reshape process. In this paper, a new method of preform preparation is explored and used to develop small core (20 µm) tellurium based glass fibers. They transmit light from 3 to about 16 µm with minimum losses near 7.9 dB·m−1. Moreover, this fiber presents single mode guidance at 10.3 µm.
© 2016 Optical Society of America
1. Introduction
Chalcogenide glasses have been studied for several decades thanks to their excellent infrared transmission ability [1–6]. Selenide and telluride glasses, due to their higher atomic weight than sulfide glasses, are proper for the mid- and far-infrared applications respectively. Indeed, Te-based bulk glasses can transmit infrared signal up to 25 µm [7–10]. In recent years, the research on these glasses has been intensified thanks to the Darwin mission [11] of European Space Agency (ESA) and Terrestrial Planet Finder(TPF) [12] of National Aeronautics and Space Administration (NASA). Both projects have the same goal: characterizing the atmospheres of terrestrial planets, and looking for the chemical signatures of life.
To detect the signal from planets, the bright light from their parent stars have to be suppressed. Nulling interferometry [13] combines light from several telescopes that are phase shifted from each other, generating a destructive interference along the optical axis of the system to suppress the invasive signal from stars. In contrast, the light from planets is displaced from the axis by a small angle, creating a constructive interference. Using this technique, one can detect the very low signal from the planets. However, getting contrasted fringes from the interferometers requires that the wavefront of light reaching the interferometer are free of distortions. This can only be achieved by using single mode fibers, which are excellent wavefront filters [6]. Moreover, the main molecules indicating biological life are H2O, O3, and CO2 with their infrared signatures located at 6 µm, 9 µm, and 15 µm respectively. Therefore, the waveguide must be transparent at least in this spectral range.
Thus, the current ESA’s Darwin project asks for the development of low optical loss single mode waveguide working in the mid- and far-infrared spectral range. Selenide and telluride glass fibers are good candidate to transmit light up to 12 µm [14, 15] and 16 µm [16, 17] respectively. Indeed, a chalcogenide glass fiber based on the Te2As3Se5 (TAS) composition which is basically a selenide glass [6, 13, 18, 19] has been selected for the short wavelength band. For the longer wavelength beyond 12 μm, telluride single mode glass fibers have to be developed and are not yet available.
In order to get fibers without surface crystallization, initial glasses need to be very stable. Indeed, the temperature difference (∆T) between glass transition temperature (Tg) and crystallization temperature (Tx) must be larger than 120 °C [20]. Following this strategy, it has been shown that the introduction of a few percent of Se improve the resistance of Ge-Te glasses against crystallization [21, 22]. Very recently, a first multimode step index Te-Ge-Se glass fiber has been prepared using the classical rod in tube technique, transmitting light from 4 to 16 µm [23]. Compared to these first achievements, the core of the fiber needs to be much smaller in order to get single mode guiding. To obtain the desired core diameter, the commonly used rod-in-tube technique asks for several successive casting and drawing processes. Thus, this method is not proper for telluride glasses which are still not stable enough even when they contain a few percent of Se.
In this paper, a new glass preform preparation method is proposed to develop small core (20 µm) tellurium glass fibers which prevents from any crystallization. A Te76Ge21Se3/Te71Ge21Se8 fiber is firstly prepared to monitor the fiber drawing parameters thanks to the high core/clad refractive indices difference. Then, Te76Ge21Se3/ Te74Ge21Se5 glasses are selected to prepare a single mode fiber working far in the mid-infrared.
2. Experimental
2.1 Material
Three glass rods composed of tellurium, germanium and selenium elements have been synthesized for the preparation of step index fibers. The three different atomic compositions are Te76Ge21Se3 (TGS3), Te74Ge21Se5 (TGS5), and Te71Ge21Se8 (TGS8). As any tiny defect and impurity absorption in an optical fiber could be greatly amplified due to the long-distance propagation of light, one of the key operations in the glass preform synthesis is the starting elements purification. In this work, the glasses were synthesized by a chemical-distillation purification method.
During the chemical purification step, a batch of 25 grams of Te (6N), Ge (5N), and Se (5N) were weighed in the adequate proportions and melted with a small amount of Al (5N) at 750 °C for 10 hours. Al acted as a kind of oxygen absorber and can thoroughly react with the oxides via a redox reaction during the homogenization period. The amorphous material was obtained by quenching and annealing at 150 °C for 3 hours. During distillation at 1000 °C, due to the vapor pressure difference between Te, Se, Ge and Al2O3 [24, 25], the purified Te, Se and Ge vapor went through a filter and condensed in a reactional silica tube at room temperature. The generated alumina Al2O3 remained in the filter. The reaction tube was then sealed and homogenized at 750 °C for 10 hours. After quenching and annealing, a purified glass rod, with its diameter and length to be around 7 and 150 mm respectively, was obtained.
2.2 Fibers elaboration
2.2.1 Modified capillary method
As explained in the introduction, the classical rod-in-tube technique is not suitable for glasses having high tendency to crystallize. Capillary method which was originally developed for the fabrication of microstructured optical fibers [26] comes into our sight. To prepare small core telluride glass fiber, a modified capillary method is developed in this work (Fig. 1).
A cylindrical glass set-up with only one silica capillary (Ø = 520 µm) in the center was used to mold a rather fluid liquid telluride glass (TGS5 or TGS8). After quenching and annealing, the glass preform for clad was immersed in HF for 30 minutes to dissolute the silica capillary, and then thoroughly cleaned in deionized water. The fiber (Ø = 440 µm), drawn from a glass with a higher refractive index (TGS3), was then introduced in the hole of the jacket tube. By using this method, one step is enough to get the preform for small core step index fiber. Compared to traditional rod-in-tube method, this new technique prevents from drawing the rod several times and minimizes the crystallization tendency. Moreover, it enables to get a smooth interface between the core and the clad.
2.2.2 Fiber drawing process
The fabrication of the single index (TGS3) and step index (TGS3/TGS8 and TGS3/TGS5) fibers was carried out under a helium controlled atmosphere thanks to a home-made fiber tower.
To prepare a TGS3 single index fiber, the glass preform surface was firstly polished to minimize the surface defects and their potential influence on step index fiber. The polish steps were fully described in [16] and we invite interested people to read for more information. The fibers with their controlled diameter to be 440 μm were obtained from the polished preform by properly fixing preform feed speed and coordinating drum speed.
A combined preform was then obtained by inserting a TGS3 fresh fiber into the hole of the jacket tube (TGS5 or TGS8). A step index fiber with its core and clad to be 20μm and 310μm respectively was obtained by drawing this combined preform. During the fiber drawing process, spaces between the fibers and the jacket tubes were eliminated upon drawing thanks to a vacuum system. The fibers with their core and clad compositions to be TGS3/TGS8 and TGS3/TGS5 respectively were successfully drawn. The TGS3/TGS8 multimode was firstly prepared due to the high refractive index difference between core and clad in order to fix the appropriate fiber drawing parameters by checking cross-sectional shape of the core, interface quality, and composition diffusion between core and clad via scanning electron microscope (SEM) and optical microscope. Then the TGS3/TGS5 fiber was designed and fabricated for single mode propagation.
2.3 Optical characterization
Fiber attenuations were measured, using a Bruker Tensor 27 FT-IR spectrometer equipped with a MCT detector, by the cutback technique [26].
In order to characterize the mode profile of transmitted light through the TGS3/TGS5 fiber samples, 2D near field intensity distribution measurements were carried out. The setup used for measurement at 10.3 µm consisted of a tunable CO2 laser and a 2D-array micro bolometer IR CCD camera (Fig. 2).
The principle of the measurement consists in the focusing and injection of an infrared interference light into the fiber core with the help of a ZnSe lens. The total fiber length is around 20 cm. In order to remove the cladding modes which are confined in the clad due to its much higher refractive index than the surrounding medium, a liquid GaSn alloy was applied on the external surface of the fiber. The signal at the output of the fiber was observed with a camera (FLIR ThermaCAM E300) equipped with an infrared detector operating in the 7 to13 μm range at room temperature. A germanium lens is placed near the output of the fiber in order to ensure a focalization on the camera sensor.
3. Results and discussion
In order to obtain a single-mode propagation, the normalized frequency V of the fiber, which is a classical parameter that enables to determine the number of modes propagated by a step-index fiber, should be less than 2.405. In this case, only one mode called fundamental mode can propagate. In practice, the V value of a specific fiber can be calculated from
where dcore is the core diameter, λ is the wavelength of the light passing through the fiber core. NA is the abbreviation of numerical aperture and can be calculated from the refractive index of core ncore and clad nclad.
According to previous works, the refractive index of Te-Ge-Se glasses can be modified in a controlled way by substituting Te by Se [23]. By a linear fitting at 10.3 μm, the relationship between Te-Ge-Se glasses refractive index nglass [23] and Se percentage CSe can be expressed by,
On this basis, to design a single-mode fiber, the TGS3 was firstly chosen to be the core composition as it is the best compromise between the optical transmission and the stability against crystallization [21, 23]. Then the TGS5 was selected as the clad composition. Their refractive indices are estimated to be 3.375 and 3.348 respectively. Taking the designed core diameter (20 μm) into account, the wavelength threshold value of single mode propagation is calculated to be around 11.2 μm. Thus, if this fiber is successfully prepared, it could be used for long wavelength, beyond 12 μm, in Darwin mission.
Note that according to previous experience on all solid microstructured fiber preparation by capillary method, inappropriate fiber drawing parameters may cause composition diffusion, tube hole shape change or interface space. Thus, several attempts are needed until the proper fiber drawing parameters are fixed. For the TGS3/TGS5 fiber, due to the similarity of ncore and nclad, the interface is almost not possible to be observed using energy-dispersive X-ray spectroscopy (EDS) analysis of SEM.
So, the TGS3/TGS8 multimode was firstly prepared. Indeed, for this step index fiber, the refractive index difference between core (ncore = 3.375) and clad (nclad = 3.307) should be high enough to enable a direct observation of the interface. Figure 3(a) displays the SEM and optical microscopy images of the fiber cross-section. To investigate the composition distribution, EDS of 20 points along a straight line through the center of the core were implemented. The actual composition evolution and the theoretical values (dotted line) are compared in Fig. 3 (b).
Fig. 3 (a) Electron microscopy images of TGS3/TGS8 fiber cross section (b) Measured and theoretical (dotted line) composition distributions comparison of TGS3/TGS8 fiber cross section.
As shown in Fig. 3(a), the final core diameter is equal to 20 µm. The interface between the core and the clad can be clearly observed using both SEM and optical microscope. The core kept a regular round shape and the space between TGS3 fiber and TGS8 preform was totally eliminated. Still, several defects can be observed in the core or in the clad, which could be attributed to the defects of the glass preform.
The EDS spatial scanning clearly points out the variation of the composition at the boundary between the core and clad. The measured concentrations of all the elements of core and clad are in good agreement with the initial composition of the glass. A transition zone with its width around 2 μm can be observed between core and clad. This composition fluctuation could be explained by several possible reasons including a limited diffusion between core and clad, the projected area limit of electron beam during EDS measurement, or the accuracy of the EDS measurement (about ± 1%). According to these first positive results, the set of parameters, such as temperature and vacuum between core and clad, used for fiber drawing, has been validated. Consequently, the first attempt of TGS3/TGS5 single mode optical preparation using capillary method was carried out following the same procedure. The targeted core and clad diameters are 20 μm and 310 μm respectively. As expected, for this new fiber, it was not possible to distinguish the core from the clad by direct SEM observation.
Then, the optical losses of this small core step index fiber were measured by the cut back method. The resulting attenuation curve is given on Fig. 4 and compared with the TGS3 single index fiber. Diamond knife was used to cleave the fibre during the measurement to obtain a good fiber cross section quality (Fig. 4 inset).
Fig. 4 Optical losses of TGS3/TGS5 fiber compared with TGS3 and Te2As3Se5 (TAS) [18] fibers. Inset is the optical microscope image of TGS3/TGS5 fiber cross section.
The both TGS3/TGS5 and TGS3 attenuation curves show a quite similar large signal transmission region extending from 3 to 16 µm. The optical losses are less than 10 dB/m between 8.5 µm and 11.5 µm. Note that when the light is guided inside the core by the total reflection at the core/clad interface, impurities and defects at the interface can cause extra losses. The optical losses of TGS3 and TGS3/TGS5 are quite similar. This result signify that the core/clad interface defects is not obvious, which is consistent with previous SEM observation. For the both curves, an absorption shoulder around 12.5 μm is visible, which is probably induced by the presence of Se into the glass composition. Compared with TGS3 fiber, the absorption shoulder of TGS3/TGS5 is higher. On the other hand, as expected, the main feature compared to selenide glass fibers [14, 18], deals with the low phonon energy associated with heavy tellurium atoms which yields multi-phonon cut-offs extending beyond 16 µm against 11 µm. The attenuation curve of a typical selenium based single mode fiber [18], Te2As3Se5 (TAS), is also shown in Fig. 4 as a comparison. Thus, such TGS3/TGS5 fibers show their potential to detect, for example, the broad absorption peak of CO2 located around 15 µm in the frame of the Darwin mission.
The second request from the ESA is to be able to filter the signal coming from the space. To achieve this goal, the tellurium glass fiber needs also to be a single mode optical guide. A single-mode propagation shows a fundamental mode bell-shaped spatial distribution similar to the Gaussian distribution [27]. To control the light intensity distribution in the core, a coherent light should be injected into the core of TGS3/TGS5 fiber. In this paper, a CO2 laser working at 10.3 µm was used for measurement.
Figure 5 shows the 2D and 3D near field profiles of the output light. After removing the confined cladding mode by coating an absorbing Ga/Sn alloy over the full fiber length, only the light propagating in the core is clearly observed. The output light exhibits a regular round shape and its intensity follows a Gaussian distribution (Fig. 5(a)), characteristic of single mode propagation. In particular, no large contribution from higher-order mode is visible. This is because the higher-order mode is not confined in the core or the losses are much greater than fundamental mode. According to the core and clad refractive index values together with the core diameter, the cut-off wavelength for single mode guiding should be around 11 µm. Nevertheless, the measurement carried out with a CO2 laser clearly demonstrates the single mode guiding from 10.3 µm, and so beyond this value. Thus, such a fiber could be used to filter the infrared signal from 10 µm until the limit of transmission of the core, 16 µm.
Fig. 5 Representation in 2D (a) and 3D (b) of near field intensity distribution of TGS3/TGS5 single mode fiber coated with a liquid GaSn alloy.
The counterpart of the broader transmission toward far infrared deals with the minimum of attenuation of the fiber which is higher (7.9 dB/m at 10.5 µm) compared to an equivalent selenide based glass fiber [18]. Indeed, for such telluride glass fiber, the residual optical absorption in the transmission range is due to the charge carriers’ concentration at room temperature inherent to the semiconducting behavior of the tellurium. Previous result [23] shows that the optical losses of Te-Ge-Se glass fiber can be greatly reduced at the temperature of liquid nitrogen (77 K). Note that Darwin mission only requires short fiber pieces of about 10 cm length [23] and the operational temperature is 40 K [13]. Thus, the measured optical losses at 15 µm, about 40 dB/m, that is to say 4 dB for 10 cm, measured at room temperature, meet the ESA request.
4. Conclusion
A novel modified capillary preform technique by inserting a pre-drawn fiber into the hole of a molded glass tube has been developed. Using this method, the Te76Ge21Se3 (TGS3)/Te74Ge21Se5 (TGS5) step index fiber with its core diameter of 20 µm was designed and prepared for single mode propagation. This is, to our knowledge, the first successful preparation of small-core fiber from telluride glasses, which have a high tendency of crystallization. This step-index fiber exhibits a broad transmission from 3 µm to 16 µm with minimum losses around 7.9 dB·m−1. In addition, the nearfield intensity distribution measurement at 10.3 µm shows its single mode propagation, making it efficient for a wavefront filter operating between 10 µm and 16 µm for the Darwin mission. Note that, at this time, the single mode selenide fiber only covers the lower part of the wavelength range, enabling detection of H2O and O3. Thus, this tellurium based glass single mode fiber constitutes one of the key steps towards the realization of Darwin mission. It enables to detect simultaneously H2O, O3 and overall the broad band the CO2 at 15 µm, while filtering the invasive signal form the stars.
Acknowledgment
Financial supports from European Community's Seventh Framework Program through Marie-Curie Action: “Initial Training Networks” (GlaCERCo GA 264526) is gratefully acknowledged.
References and links
1. L. Calvez, H. Ma, J. Lucas, and X. Zhang, “Selenium-Based Glasses and Glass Ceramics Transmitting Light from the Visible to the Far-IR,” Adv. Mater. 19(1), 129–132 (2007). [CrossRef]
2. J. A. Savage, “Optical properties of chalcogenide glasses,” J. Non-Cryst. Solids 47(1), 101–115 (1982). [CrossRef]
3. X. Zhang, Y. Guimond, and Y. Bellec, “Production of complex chalcogenide glass optics by molding for thermal imaging,” J. Non-Cryst. Solids 326-327, 519–523 (2003). [CrossRef]
4. T. Kanamori, Y. Terunuma, S. Takahashi, and T. Miyashita, “Chalcogenide glass fibers for mid-infrared transmission,” J. Lightwave Technol. 2(5), 607–613 (1984). [CrossRef]
5. C. Vigreux, E. Barthélémy, L. Bastard, J.-E. Broquin, M. Barillot, S. Ménard, G. Parent, and A. Pradel, “Realization of single-mode telluride rib waveguides for mid-IR applications between 10 and 20 μm,” Opt. Lett. 36(15), 2922–2924 (2011). [CrossRef] [PubMed]
6. C. Vigreux-Bercovici, V. Ranieri, L. Labadie, J.-E. Broquin, P. Kern, and A. Pradel, “Waveguides based on Te2As3Se5 thick films for spatial interferometry,” J. Non-Cryst. Solids 352(23-25), 2416–2419 (2006). [CrossRef]
7. S. Danto, P. Houizot, C. Boussard-Plèdel, X. H. Zhang, F. Smektala, and J. Lucas, “A Family of Far-Infrared-Transmitting Glasses in the Ga–Ge–Te System for Space Applications,” Adv. Funct. Mater. 16(14), 1847–1852 (2006). [CrossRef]
8. A. A. Wilhelm, C. Boussard-Plédel, Q. Coulombier, J. Lucas, B. Bureau, and P. Lucas, “Development of Far-Infrared-Transmitting Te Based Glasses Suitable for Carbon Dioxide Detection and Space Optics,” Adv. Mater. 19(22), 3796–3800 (2007). [CrossRef]
9. Z. Yang and P. Lucas, “Tellurium Based Far Infrared Transmitting Glasses,” J. Am. Ceram. Soc. 92(12), 2920–2923 (2009). [CrossRef]
10. X. Zhang, H. Ma, C. Blanchetière, and J. Lucas, “Low loss optical fibres of the tellurium halide-based glasses, the TeX glasses,” J. Non-Cryst. Solids 161, 327–330 (1993). [CrossRef]
11. L. Kaltenegger and M. Fridlund, “The Darwin mission: Search for extra-solar planets,” Adv. Space Res. 36(6), 1114–1122 (2005). [CrossRef]
12. C. A. Beichman, “Terrestrial Planet Finder: the search for life-bearing planets around other stars,” Proc. SPIE 3350, 719–723 (1998). [CrossRef]
13. P. Houizot, C. Boussard-Plédel, A. J. Faber, L. K. Cheng, B. Bureau, P. A. Van Nijnatten, W. L. M. Gielesen, J. Pereira do Carmo, and J. Lucas, “Infrared single mode chalcogenide glass fiber for space,” Opt. Express 15(19), 12529–12538 (2007). [CrossRef] [PubMed]
14. J. S. Sanghera and I. D. Aggarwal, “Active and passive chalcogenide glass optical fibers for IR applications: a review,” J. Non-Cryst. Solids 256–257, 6–16 (1999). [CrossRef]
15. D. Le Coq, K. Michel, J. Keirsse, C. Boussard-Plédel, G. Fonteneau, B. Bureau, J.-M. Le Quéré, O. Sire, and J. Lucas, “Infrared glass fibers for in-situ sensing, chemical and biochemical reactions,” C. R. Chim. 5(12), 907–913 (2002). [CrossRef]
16. S. Cui, C. Boussard-Plédel, J. Lucas, and B. Bureau, “Te-based glass fiber for far-infrared biochemical sensing up to 16 μm,” Opt. Express 22(18), 21253–21262 (2014). [CrossRef] [PubMed]
17. V. S. Shiryaev, A. P. Velmuzhov, M. F. Churbanov, A. D. Plekhovich, C. Boussard-Plédel, J. Troles, C. Conseil, and V. G. Plotnichenko, “Preparation and investigation of high purity Ge–Te–AgI glasses for optical application,” J. Non-Cryst. Solids 377, 1–7 (2013). [CrossRef]
18. C. Conseil, Q. Coulombier, C. Boussard-Plédel, J. Troles, L. Brilland, G. Renversez, D. Mechin, B. Bureau, J. L. Adam, and J. Lucas, “Chalcogenide step index and microstructured single mode fibers,” J. Non-Cryst. Solids 357(11-13), 2480–2483 (2011). [CrossRef]
19. V. S. Shiryaev, C. Boussard-Plédel, P. Houizot, T. Jouan, J. L. Adam, and J. Lucas, “Single-mode infrared fibers based on TeAsSe glass system,” Mat. Sci. Eng. B-Solid 127(2-3), 138–143 (2006). [CrossRef]
20. C. Conseil, V. S. Shiryaev, S. Cui, C. Boussard-Pledel, J. Troles, A. P. Velmuzhov, A. M. Potapov, A. I. Suchkov, M. F. Churbanov, and B. Bureau, “Preparation of High Purity Te-Rich Ge-Te-Se Fibers for 5-15 μm Infrared Range,” J. Lightwave Technol. 31, 1703–1707 (2013). [CrossRef]
21. S. Maurugeon, B. Bureau, C. Boussard-Plédel, A. J. Faber, X. H. Zhang, W. Geliesen, and J. Lucas, “Te-rich Ge–Te–Se glass for the CO2 infrared detection at 15μm,” J. Non-Cryst. Solids 355(37-42), 2074–2078 (2009). [CrossRef]
22. S. Maurugeon, B. Bureau, C. Boussard-Plédel, A. J. Faber, P. Lucas, X. H. Zhang, and J. Lucas, “Selenium modified GeTe4 based glasses optical fibers for far-infrared sensing,” Opt. Mater. 33(4), 660–663 (2011). [CrossRef]
23. S. Maurugeon, C. Boussard-Plédel, J. Troles, A. J. Faber, P. Lucas, X. H. Zhang, J. Lucas, and B. Bureau, “Telluride Glass Step Index Fiber for the far Infrared,” J. Lightwave Technol. 28, 3358–3363 (2010).
24. R. E. Cohen, D. Lide, and G. Trigg, Physicist's Desk Reference (Springer Science & Business Media, 2003)
25. Z. Z. Fang, Sintering of Advanced Materials (Elsevier, 2010).
26. Q. Coulombier, L. Brilland, P. Houizot, T. Chartier, T. N. N’guyen, F. Smektala, G. Renversez, A. Monteville, D. Méchin, T. Pain, H. Orain, J.-C. Sangleboeuf, and J. Trolès, “Casting method for producing low-loss chalcogenide microstructured optical fibers,” Opt. Express 18(9), 9107–9112 (2010). [CrossRef] [PubMed]
27. M. C. Teich and B. Saleh, Fundamentals of Photonics (Wiley 1991).
