Highly efficient tunable optical filter based on liquid crystal micro-ring resonator with large free spectral range

Abstract

A highly efficient tunable optical filter of liquid crystal (LC) optical micro-ring resonator (MRR) was proposed. The 4-μm-radius ring consists of a silicon-on-insulator (SOI) asymmetric bent slot waveguide with a LC cladding. The geometry of the slot waveguide resulted in the strong electro-optic effect of the LC, and therefore induced an increase in effective refractive index by 0.0720 for the quasi-TE mode light in the slot-waveguide. The ultra-wide tuning range (56.0 nm) and large free spectral range (FSR) (~28.0 nm) of the optical filters enabled wavelength reconfigurable multiplexing devices with a drive voltage of only 5 V. The influences of parameters, such as the slot width, total width of Si rails and slot shift on the device’s performance, were analyzed and the optimal design was given. Moreover, the influence of fabrication tolerances and the loss of device were both investigated. Compared with state-of-the-art tunable MRRs, the proposed electrically tunable micro-ring resonator owns the excellent features of wider tuning ranges, larger FSRs and ultralow voltages.

Keywords

integrated optics devices / liquid crystals / micro-ring resonator / slot waveguide / wavelength tuning

1 Introduction

Low water peak single-mode fiber (LWP-SMF) develops very fast in recent years. Compared with the conventional SMF (G652A/B), the LWP fiber has significantly reduced optical attenuation at water absorption wavelength around 1383 nm. It provides expanded transmission window from 1310 to 1550 nm continuously and makes fiber useful for communication at O, E, and S bands. This optical fiber is designed and manufactured to meet the requirement in applications in optical fiber communication networks, cable TV, utilities, and special optical networks. It has been predicted that the LWP-SMF will be the main type of SMF products in the market in the future.

Deuterium (D₂) was popularly used in fiber production to produce fiber with low water peak at 1.38 μm. The reduction of hydroxyl (OH) absorption in optical fiber can be achieved by isotope exchange from OH to deuterloxyl (OD) [1]. There is almost no added loss by OD absorption in the entire window between 1.3 and 1.6 μm. It has been reported that the treatment of D₂ on optical fiber at high temperature (800°C) can effectively reduce the OH level in the glass fiber [2]. However, this method is not suitable for application in the production of optical fiber.

The fabrication of LWP fiber in the production includes two main steps. The first is to produce glass preform with low OH level inside. To achieve this target, many methods to reduce OH level in glass had been researched. The details of the OH-reducing process vary with the type of preform fabrication. In outside vapor deposition (OVD) and vapor-phase axial deposition (VAD) process, the important way to reduce water content is by drying the silica soot before it is consolidated. The porous preform consisting of silica soot particles is heated in an atmosphere of SOCl₂ or Cl₂. In modified chemical vapor deposition (MCVD) process, Cl₂ is used in deposition process and reacted with water to reduce OH level. The OH reduction in plasma activated chemical vapor deposition (PCVD) process is more complex. Because the transformation from silica soot to glass is so quick, it is not possible to dry the soot with Cl₂ in deposition process. Thus, in PCVD process, the reduction of OH focuses on humidity control of raw material, environment, and equipment [3].

Although OH can be incorporated into the glass fiber in many stages from the starting material to the fiber-drawing operation, the control of OH level in the drawing process is relatively simple. However, after drawing, the second important step for LWP fiber production is the D₂ treatment. For GeO₂-doped silica glass fiber, some structural defects always result in the fiber core during the fiber fabrication process. The defects include several different types of atomic defects and impurities present in the silica fiber. For this type of fiber, there is a significant risk that the optical loss could increase due to the chemical reaction between defects and molecular hydrogen present in or around optical cables [4]. The increase of optical loss for this reason is called hydrogen-induced loss (HIL).

The HIL generally includes two main parts: one part is the loss caused by diffusion of molecular H₂ in the glass fiber. It is reversible and can be eliminated by heat treatment. The other part is the loss resulting from chemical reaction between defects and H₂. It is irreversible and difficult to be eliminated [5]. However, by D₂ treatment (exposing fiber in D₂ environment) after drawing, the defects in the glass fiber can react with D₂ prior to H₂. The formation of OD and SiD is harmless to the normal operating wavelength windows. After that, the reactive silica defects in the glass fiber being passivated by deuterium reaction will no longer be able to cause additional HIL in the field.

For the D₂ treatment process, the determination of treatment time and concentration of D₂ in the environment is important. The improper design on these parameters will result in two opposite results: one is the insufficiency of reaction between D₂ and defects in the fiber. The fiber will have problem of HIL in later use. The other is the excessive diffusion of D₂ in the glass fiber. The attenuation of fiber will increase obviously after D₂ treatment.

This paper will try to investigate the reaction between D₂ and defects in glass fiber to get proper set point for the D₂ treatment process.

2 Experiment

2.1 Fabrication of preform

Preforms for fiber drawing were pure silica glass doped with GeO₂ and F. The preform is fabricated by PCVD process. The deposition speed of glass in the PCVD process is about 0.5-2.0 g/min in the core. Most of the cladding of the fiber is pure silica glass. The amount of dopants in the core is about 1.5 mol% GeO₂ and 0.7 mol% F.

2.2 Drawing process of fiber

Optical fiber was drawn on drawing tower with the height of 25 m in total. In experiment, the drawing speed was 1300-1500 m/min, and the drawing tension was kept at 100-150 g. Inert gases, Ar and He, were used in the electricity-powered drawing furnace.

2.3 Process of D₂ treatment

After measurement on geometry and optical parameters, the fiber was packed with plastic film and put into box filled with N₂/D₂ gases. The device is schematically described in Fig. 1. The temperature in the box was kept constant at 21°C. The concentration of D₂ in the box can be monitored and controlled by a computer. During the experiment, the concentration of D₂ changed in the range of 1.0%-2.0%.

Fig.1 Schematic structure of D2 treatment system

Full size|PPT slide

2.4 Measuring of fiber after D₂ treatment

After D₂ treatment, the fiber was kept in natural environment for about 2 h. After that, the fiber was measured on attenuation and tested on hydrogen aging. The difference on attenuation of fiber before and after D₂ treatment is defined as the excess loss resulting from the D₂ treatment.

3 Results and discussion

3.1 Effect of D₂ treatment time on HIL

The fiber was treated with 1.2% D₂ for 12, 16, 24, and 40 h. The same experiment also has been done on 1.5% D₂ for 12, 24, and 40 h. The fiber which has not been D₂ treated was selected as the reference fiber. The HILs at 1383 and 1530 nm of these fibers were shown in Figs. 2(a) and 2(b), respectively.

Fig.2 HIL of fibers samples. (a) HIL at 1383 nm; (b) HIL at 1530 nm

Full size|PPT slide

Figure 2 shows that the time of D₂ treatment has obvious effect on hydrogen aging resistance of fiber. If the treatment time is less than 24 h, the HIL at 1383 nm will be higher than 0.035 dB/km. It means that the fiber cannot be qualified on the property of the hydrogen aging resistance as LWP fiber. Thus, the time of D₂ treatment should be more than 24 h; 40 h will be better for safety. Long-time D₂ treatment will consume too much of D₂, which is very expensive. However, the time of D₂ treatment seems to have no important effect on HIL at 1530 nm. It proves that the formation of SiH during hydrogen aging test is very little.

3.2 Effect of D₂ concentration on excess loss

Based on experiment results discussed above, the effect of D₂ concentration on excess loss of fiber after D₂ treatment has been investigated. The fiber was treated with different D₂ concentration of 0.8%, 1.0%, 1.2%, 1.5%, and 2.0%. The time of D₂ treatment was kept for 40 h. After the D₂ treatment, the fiber samples treated with 0.8% and 1.0% D₂ cannot be qualified for the hydrogen aging test. The HIL at 1383 nm is high even with long treatment time of 40 h.

The excess loss of fiber samples at 1383 nm after D₂ treatment is shown in Fig. 3.

Fig.3 Excess loss at 1383 nm of fiber after D2 treatment

Full size|PPT slide

In Fig. 3, when the concentration of D₂ is more than 1.5%, the excess loss at 1383 nm after D₂ treatment will become obvious. For fiber production, the excess loss should be controlled at a low level to ensure that the fiber will be kept as LWP fiber after D₂ treatment. So the concentration of 1.2% is the ideal choice for production.

3.3 Mechanism of excess loss resulting from D₂ treatment

The excess loss of fiber after the D₂ treatment was thought to result from D₂ molecules in the core of the fiber. When fiber was heated to about 50°C for 24-48 h in the air, most of the excess losses can be reduced because the D₂ molecules can be released from the fiber by diffusion. This means that there is no strong chemical bond between D₂ molecules and silica structure. However, only some weak bond formed between the D₂ molecules and the defects in the silica. The combinations between D₂ molecules and defects may form through a special “bridge”.

Because the excess loss is related to defects in the glass fiber, some ways to reduce defects in the glass fiber would be helpful on reduction of excess loss.

3.4 Effect of fiber RI profile design

To reduce the drawing-induced defects in the glass fiber, the viscosity match between the core and the cladding was optimized. For conventional SMF, the viscosity of the cladding is much higher than that of the core. So, in the drawing process, there will be defects formed at the boundary between the core and the cladding. By adding a buffer layer between the core and the cladding, the viscosity match between the core and the cladding can be effectively optimized. The refractive index (RI) profile and viscosity profile of the glass fiber are shown in Fig. 4, where Δ in Fig. 4(a) represents the percentage of change on RI compared with that of pure silica.

Fig.4 RI and viscosity profile of optimized SMF. (a) RI profile; (b) viscosity profile

Full size|PPT slide

With this optimized viscosity match between the core and the cladding, the excess loss at 1383 nm of fiber after the D₂ treatment with 2.0% D₂ for 40 h can be reduced to less than 0.005 dB/km.

3.5 Optimization on D₂ treatment process

Based on experiment results discussed above, the proper D₂ treatment process for production can be decided. To ensure that the fiber has resistance to hydrogen aging while not resulting in high excess loss after D₂ treatment, and also to save the cost of D₂ treatment process, the concentration and time of the D₂ treatment should be set as 1.2% and 40 h, respectively.

4 conclusion

The results of the experiment prove that both the D₂ treatment time and concentration have an important effect on the fiber’s properties after D₂ treatment. For the D₂ treatment time, 40 h is suitable for fiber production to ensure that the fiber has resistance to hydrogen aging, because longer D₂ treatment time will cause too much consumption of D₂. The concentration of D₂ with 1.2% is enough for common LWP-SMF and will not result in obvious excess loss after D₂ treatment. The excess loss is thought to be related to combinations between the D₂ molecules and defects in the glass fiber, and the excess loss can be reduced by optimization on viscosity between the core and the cladding of the glass fiber.

References

[1] Selvaraja S K, Jaenen P, Bogaerts W, Van Thourhout D, Dumon P, Baets R. Fabrication of photonic wire and crystal circuits in silicon-on-insulator using 193-nm optical lithography. Journal of Lightwave Technology, 2009, 27(18): 4076–4083

[2] Dumon P, Bogaerts W, Wiaux V, Wouters J, Beckx S, Van Campenhout J, Taillaert D, Luysaert B, Bienstman P, Van Thourhout D, Baets R. Low-loss SOI photonic wires and ring resonators fabricated with deep UV lithography. IEEE Photonics Technology Letters, 2004, 16(5): 1328–1330

[3] Basch E B, Egorov R, Gringeri S, Elby S. Architectural tradeoffs for reconfigurable dense wavelength-division multiplexing systems. IEEE Journal of Selected Topics in Quantum Electronics, 2006, 12(4): 615–626

[4] Ng H, Wang M R, Li D, Wang X, Martinez J, Panepucci R R, Pathak K. 1×4 wavelength reconfigurable photonic switch using thermally tuned microring resonators fabricated on silicon substrate. IEEE Photonics Technology Letters, 2007, 19(9): 704–706

[5] Yalcin A, Popat K C, Aldridge J C, Desai T A, Hryniewicz J, Chbouki N, Little B E, King O, Van V, Chu S, Gill D, Anthes-Washburn M, Unlu M S, Goldberg B B. Optical sensing of biomolecules using microring resonators. IEEE Journal of Selected Topics in Quantum Electronics, 2006, 12(1): 148–155

[6] Dong P, Qian W, Liang H, Shafiiha R, Feng N, Feng D, Zheng X, Krishnamoorthy A V, Asghari M. Low power and compact reconfigurable multiplexing devices based on silicon microring resonators. Optics Express, 2010, 18(10): 9852–9858

[7] Guarino A, Poberaj G, Rezzonico D, Degl’innocenti R, Günter P. Electro-optically tunable microring resonators in lithium niobate. Nature Photonics, 2007, 1(7): 407–410

[8] Liu A, Liao L, Rubin D, Nguyen H, Ciftcioglu B, Chetrit Y, Izhaky N, Paniccia M. High-speed optical modulation based on carrier depletion in a silicon waveguide. Optics Express, 2007, 15(2): 660–668

[9] Dong P, Qian W, Liang H, Shafiiha R, Feng D, Li G, Cunningham J E, Krishnamoorthy A V, Asghari M. Thermally tunable silicon racetrack resonators with ultralow tuning power. Optics Express, 2010, 18(19): 20298–20304

[10] Nishihara H, Haruna M, Suhara T. Optical Integrated Circuit. New York: McGraw-Hill1985, 36–44

[11] Soref R A, Bennett B R. Electrooptical effects in silicon. IEEE Journal of Quantum Electronics, 1987, 23(1): 123–129

[12] Maune B, Lawson R, Gunn C, Scherer A, Dalton L. Electrically tunable ring resonators incorporating nematic liquid crystals as cladding layers. Applied Physics Letters, 2003, 83(23): 4689–4691

[13] Falco A D, Assanto G. Tunable wavelength-selective add–drop in liquid crystals on a silicon microresonator. Optics Communications, 2007, 279(1): 210–213

[14] De Cort W, Beeckman J, James R, Fernandez F A, Baets R, Neyts K. Tuning silicon-on-insulator ring resonators with in-plane switching liquid crystals. Journal of the Optical Society of America B, Optical Physics, 2011, 28(1): 79–85

[15] De Cort W, Beeckman J, Claes T, Neyts K, Baets R. Wide tuning of silicon-on-insulator ring resonators with a liquid crystal cladding. Optics Letters, 2011, 36(19): 3876–3878

[16] Almeida V R, Xu Q, Barrios C A, Lipson M. Guiding and confining light in void nanostructure. Optics Letters, 2004, 29(11): 1209–1211

[17] Anderson P A, Schmidt B S, Lipson M. High confinement in silicon slot waveguides with sharp bends. Optics Express, 2006, 14(20): 9197–9202

[18] Kargar A, Chao C. Design and optimization of waveguide sensitivity in slot microring sensors. Journal of the Optical Society of America A, Optics, Image Science, and Vision, 2011, 28(4): 596–603

[19] Pfeifle J, Alloatti L, Freude W, Leuthold J, Koos C. Silicon-organic hybrid phase shifter based on a slot waveguide with a liquid-crystal cladding. Optics Express, 2012, 20(14): 15359–15376

[20] Barrios C A, Lipson M. Electrically driven silicon resonant light emitting device based on slot-waveguide. Optics Express, 2005, 13(25): 10092–10101

[21] Gould M, Baehr-Jones T, Ding R, Huang S, Luo J, Jen A K, Fedeli J M, Fournier M, Hochberg M. Silicon-polymer hybrid slot waveguide ring-resonator modulator. Optics Express, 2011, 19(5): 3952–3961

[22] Desmet H, Neyts K, Baets R. Liquid crystal orientation on patterns etched in silicon on insulator. Integrated Optics, Silicon Photonics, and Photonic Integrated Circuits, 2006, 6183: 61831Z

[23] Ge Z, Wu T X, Lu R, Zhu X, Hong Q, Wu S. Comprehensive three-dimensional dynamic modeling of liquid crystal devices using finite element method. Journal of Display Technology, 2005, 1(2): 194–206

[24] Fallahkhair A B, Li K S, Murphy T E. Vector finite difference modesolver for anisotropic dielectric waveguides. Journal of Lightwave Technology, 2008, 26(11): 1423–1431

[25] Dell’Olio F, Passaro V M. Optical sensing by optimized silicon slot waveguides. Optics Express, 2007, 15(8): 4977–4993

[26] Donisi D, Bellini B, Beccherelli R, Asquini R, Gilardi G, Trotta M, d’ Alessandro A. A switchable liquid-crystal optical channel waveguide on silicon. IEEE Journal of Quantum Electronics, 2010, 46(5): 762–768

Acknowledgements

The authors would like to thank Shuchang Yao and Yong Mei for useful discussion. This work was supported by the National High Technology Research and Development Program of China (No. SS2015AA010104), the Major Project of Science and Technology Innovation Program of Hubei Province of China (No. 2014AAA006), and the National Natural Science Foundation of China (Grant No. 61107051).

RIGHTS & PERMISSIONS

2014 Higher Education Press and Springer-Verlag Berlin Heidelberg