A core chip of optofluidic variable optical attenuator (VOA) is reported. The chip, with a simple structure, utilizes microfluid and compressed air to regulate the optical attenuation, and it can be expanded to form a number of VOAs by using different microfluidic driving technologies. Three VOAs based on this chip and different driving technologies are introduced. The theoretical and experimental results show that the proposed chip possesses the advantages of large optical attenuation range (> 50dB) and low insertion loss (0.55 dB). Moreover it is a broadband optical device which can be operated in visible and near infrared wavelengths. The proposed chip provides a new method for seeking miniaturized VOAs with good performances, and it is promising to develop a number of different VOAs.
© 2016 Optical Society of America
Optical attenuators are used to control the optical signal intensity and are widely applied in optics, photonics and optical communication networks. They are used not only in free space, but also in the optical fiber systems . Variable optical attenuators (VOAs) are mainly applied in the optical telecommunication, particularly in the wavelength-division multiplexing (WDM) system. They are also applied in other fields. E.g. Pu Wei et al. proposed a current transformer with a VOA . With increasing popularity of the visible light communications, more and more VOAs are used in the visible waveband .
The attenuation range and the insertion loss are two important technical indexes of the VOAs. Most of the conventional VOAs are mechanical VOAs. They have good optical performances, but their volumes are too large to integrate and their structures are complicated. MEMS based VOAs have smaller volumes than the mechanical VOAs, but their structures are still complicated. There are also some other conventional VOAs which are based on liquid crystal, magneto-optic effect and electro-optic effect, but their insertion losses are over 1dB in general. In recent years, S. Rudra et al  reported a blue phase liquid crystal based VOA which has an attenuation range of 29dB from 1480nm to 1550nm. A MEMS based VOA proposed by Anartz Unamuno  has an attenuation range of over 47dB and an insertion loss of 1dB from 1525nm to 1565nm. Kah How koh and Bo Woon soon  proposed a micromirror based VOA which have an attenuation range of 40dB and an insertion loss of 1.8dB from 1510nm to 1610nm. Also, Kristjan Leosson et al  reported a compact plasmonic VOA with an attenuation range of 40dB and an insertion loss of 1dB from 1525nm to 1565nm.
Optofluidics is the combination of optics, optoelectronics, photonics and microfluidics. It can be applied in optical sensing, measurement, imaging, and communication and so on. There are only a few reports about the optofluidic VOAs. Syed A.R and Nabeel A.R  reported a liquid lens based VOA which used the driving technology of electrowetting on dielectric (EWOD), and this VOA had an attenuation range of 40dB and an insertion loss of 4.3dB from 1510nm to 1700nm. By using a waveguide and a microchannel filled with fluid mixture, Xionggui Tang and Rujian Li  proposed a VOA scheme with an attenuator range of 31dB from 1500nm to 1600nm. Recently Anna Dudu’s and Robert Blue  demonstrated an electrowetting based VOA which used a small liquid drop to regulate the attenuation, and this VOA had an attenuation range of 26dB from 1525nm to 1565nm.
Here a core chip of optofluidic VOA with a simple structure is proposed. It utilizes microfluid and air to regulate the optical attenuation. It is an extensible chip and can form a number of VOAs by using different microfluidic driving technologies. Moreover, the theoretical and experimental results show that the proposed chip possesses the advantages of large optical attenuation range (> 50dB) and low insertion loss (0.55dB). Also, it is a broadband optical device and can be widely applied in visible or NIR optical communications as well as other fields. In order to show the extendability of the proposed chip, three VOAs with this chip are introduced. They use the driving technologies based on air pressure, piezoelectric ceramics and electromagnetic pump, respectively. The proposed chip provides a new method for seeking miniaturized VOAs with good performances, and it is promising to develop a number of VOAs.
2. Structure and working principle
In this section, the structure and the working principle of the proposed chip are introduced, and three VOAs based on the proposed chip and different driving technologies are given.
2.1. Structure of the chip
The proposed optofluidic VOA chip adopts a sandwich structure, as shown in Fig. 1(a). The first layer is a cover. The second layer is a main working layer. Two V-shape grooves are carved on a homogeneous and transparent square medium, and their central axes lie on the same straight line. The incident and exit collimators are set in the V-shape grooves. There is a microchannel at the diagonal of the square medium, and the microchannel is connected with a liquid storage tank. Additionally there is some air and liquid sealed in the microchannel by a thin membrane. The third layer is a substrate. The V-shape grooves are beneficial to collimate the two collimators accurately. The liquid in the microchannel has the same or proximal refractive index as the square medium. The liquid storage tank is connected with a fluid driving pump. The microfluidic driving technologies can utilize electroosmosis , hot vapor [12, 13], electromagnetism , Marangoni convection , EWOD [16, 17], light [18, 19], etc.
The fluid can be driven horizontally or vertically. In the horizontal fluid-drive chip, the liquid and air are distributed along the horizontal direction in the microchannel, as shown in Fig. 1(b). While in the vertical fluid-drive chip, the liquid storage tank is under the microchannel, where air and liquid are distributed vertically in the microchannel, as shown in Fig. 1(c).
2.2. Working principle of the Chip
If the fluid is driven horizontally, the working principle is shown in Fig. 2. As the pump drives liquid in the microchannel, air at the other end of the microchannel will be contracted or expanded. After the forces between liquid and air achieve a balance, liquid and air stop moving. Light on the solid-liquid interface can penetrate through the microchannel almost without loss because liquid has the same or approximate refractive index as the square medium, and then it is coupled into the exit collimator. While light on the solid-air interface will be reflected and lost. As a result, the light energy behind the microchannel is reduced. The air sealed in the microchannel acts as an “air shutter”. By driving the fluid to arrive at different balance locations, the optical attenuation is regulated. If the fluid is driven vertically, the working principle is similar to the one above.
The wavelength has less effect to the refractive index and the critical angle of the total reflection of light. If the chip is made of PMMA (Polymethyl methacrylate) with a refractive index of 1.491 at 650nm, the critical angle of the total reflection is between 41.5° and 42.5° within 400nm-1700 nm at the solid-air interface of the microchannel, while the incident angle of light is 45°, so the total reflection condition is met. Therefore the chip is a broadband optical device which can be operated in visible and NIR wavelengths.
2.3. Chip based VOAs
Here are three VOAs based on the proposed chip and different driving technologies, where two horizontal fluid-drive VOAs are shown in Fig. 3.
One VOA uses a pneumatic pump, as shown in Fig. 3(a). The liquid storage tank is connected with the pneumatic pump by an outlet on the cover. Meanwhile the outlet is also used to pour liquid. The pneumatic pump is set out of the chip. The fluid is driven by air pressure from the pump. The second VOA uses a driving setup with a piezoelectric ceramics, as shown in Fig. 3(b). The piezoelectric ceramics is bonded with the cover. The PDMS membrane is elastic. Based on the converse piezoelectric effect , the piezoelectric ceramics becomes longer when a positive voltage is applied. Then it pushes down the PDMS membrane and liquid in the liquid storage tank to drive the fluid in the microchannel. The piezoelectric ceramics length is controlled by the applied voltage. The two VOAs have simple structures and simple fabrication processes.
A vertical fluid-drive VOA is shown in Fig. 4. It uses a micro electromagnetic pump. The electromagnetic pump consists of a permanent magnet wafer and an electromagnet surrounded with electric wires, where the polarity of the magnet wafer is opposite from that of the electromagnet when the electric source is in a power-on state. In the power-on state, as the electromagnet and the magnet wafer repel each other, the magnet wafer moves up and pushes liquid into the microchannel up to a certain height. In a power-off state, as the electromagnet and the magnet wafer attract each other, adding the action of gravity, the magnet wafer falls and liquid rapidly falls into the liquid storage tank. The liquid height in the microchannel is controlled by the applied current on the electromagnet.
The optical field in the VOA chip is divided by the microchannel into two parts, the front optical field and the back optical field. The optical field in front of the microchannel can be expressed by Gaussian beam. The optical field behind the microchannel includes light penetrating through the microchannel and light diffracted by the liquid head face in the microchannel. Here Gauss beam propagation theory, Helmholtz equation and coupling theory are used in the theory. Mathematica software is also used for simulation.
3.1. Optical field analysis
Assuming that the VOA works in the single-mode optical field, the optical wavefunction of the optical field in front of the microchannel can be expressed by Gaussian laser beam as below,
Referring to Fig. 2, assume that the origin of the coordinates is the intersection point of the optical axis and the microchannel, x coordinate value of the liquid head face is ax. When diffraction is not considered, the position function M(x) of light at the microchannel is
At the liquid head face, an air-liquid interface in the microchannel, light is diffracted. So the light beam behind the microchannel becomes wider, and the optical energy into the exit collimator is further reduced. This loss originated from diffraction is called the mode-field mismatch loss.
The diffraction wave behind the microchannel is expressed by Helmholtz equation,
3.2. Light intensity distribution and attenuation
Assume that the optical source is a NIR laser with 1550nm wavelength, the distance between the two collimators is 20mm, the waist radius of Gaussian beam (ω0) is 0.18mm, and the liquid is driven along the horizontal direction. Figure 5 shows x and y directional distribution of optical intensity at the receiving face of the exit collimator, respectively.
As shown in Figs. 5(a) and 5(b), light intensity distribution is not symmetrical in x (horizontal) direction, but it is symmetrical in y (vertical) direction. This is because the fluid moves along horizontal direction. Figure 6 shows 3D light intensity distribution on the exit collimator. According to Fig. 5 and Fig. 6, when the location of the liquid head face is at ax = ω0, most light penetrates through the microchannel and reaches the exit collimator, hence, the light energy loss is very small. As the location of the liquid head face moves from ax = ω0 to ax = -ω0, more and more light is reflected and lost. As a result, the light intensity behind the microchannel becomes weaker and weaker. At ax = -ω0, all light is almost reflected and lost. As shown in Fig. 5(a) and Fig. 6, there are some fluctuations originated from diffraction in the light intensity distribution, moreover, the location of light intensity apex also gradually moves away from the central axis. It is evident that the optical field behind the microchannel is disturbed by diffraction and isn’t normal Gaussian distribution again.
The VOA attenuation is regulated by controlling the location of the liquid head face. The attenuation consists of two parts: 1) the attenuation brought by total internal reflection (IL1); 2) the mode-field-mismatch attenuation caused by diffraction (IL2). According to the mode-field-coupling theory, the total attenuation (IL) is
4. Experiments and discussion
The optical performances of the proposed VOA chip, such as the attenuation and the insertion loss, are researched by some principle experiments, where the specific driving technology is not temporarily considered.
The experiment set up is shown in Fig. 7. The chip is made of PMMA (Polymethyl methacrylate) with a refractive index of 1.491 at 650nm, and its dimension is 15 × 15 × 15mm3 (length × width × height), while the microchannel size is 10 × 1 × 4mm3 (length × width × depth). There are three fabrication methods for the PMMA chip. One is the injection molding method. The second is the hot embossing method. The third is the precision engraving technology based on a CNC engraving machine. The first two methods have higher costs than the third, so the third method is selected to fabricate the PMMA chip. Liquid is poured into the microchannel by the filling port, where air is in the upper layer while liquid is in the lower layer. The liquid is phenylmethylsilicone fluid with a refractive index of 1.490 at 650nm. This chip is set on a vertical axis lift with 0.005mm shifting precision. The light source uses a visible laser with 650nm wavelength. A stop is used to filter stray light. A collimator is behind the stop, and the laser beam behind it has a 1mm diameter. An optical sensor and a power meter are used to receive light and measure the output power. A screen is set in front of the optical sensor when observing the output spot of light. The whole set-up is laid on a quake proof platform.
The output light spot is captured by a CCD camera, as shown in Fig. 8. As the chip moves down following the vertical axis lift, the liquid head face in the microchannel descends, and the laser beam in front of the microchannel moves gradually from the solid-liquid interface to the solid-air interface. The experimental photos in Fig. 8 show that the output spot becomes gradually smaller and darker and finally becomes irregular at the edge as the liquid head face descends. The irregularity of the spot edge is caused by diffraction. The experiment results indicate that it is feasible to regulate optical attenuation using the proposed chip.
The output power is shown in Fig. 9(a), where the input optical power is 205μw. The measured output power decreases gradually as the liquid head face moves down. The output power changes slowly during the initial shift of the liquid head face, but drops quickly within the shift of 0.3mm-0.9mm. The output power becomes very weak over a 1mm shift. The nonlinear power variation is due to the nonuniform energy distribution of the laser beam itself (Gaussian distribution) as well as the actions of reflection and diffraction. As shown in Fig. 9(b), the measured attenuation increases gradually as the liquid head face moves down, and the attenuation range is over 50dB. Meanwhile, the experiment data are consistent with theoretical results. The experimental results show the proposed VOA chip has a larger attenuation range over the conventional VOAs.
Due to extra energy losses from absorption of the chip material and transverse/angle deviation of the two collimators, there is an insertion loss. The measured insertion loss is 0.55dB. It is lower than the insertion loss of the conventional VOAs.
In this paper, a core chip of optofluidic VOA is proposed and researched by theory and experiments. The proposed VOA chip uses microfluid and compressed air to regulate the optical attenuation. It is an extensible chip and can form a number of VOAs by using different microfluidic technologies. The theoretical and experimental results show the proposed chip is feasible. Moreover, it possesses the advantages of large optical attenuation range (>50dB) and low insertion loss (0.55 dB) . Meanwhile it is also a broadband optical device which can be operated in visible and NIR wavelengths. Based on the proposed chip, further work can research a VOA with a specific driving technology or develop VOAs which use other driving technologies.
This work is supported by National Natural Science Foundation of China (61574080, 61274121). The authors thank T. Stephan Cannon for his help with English, also thank the constructive discussions from Prof. Peili Li, Prof. Weihua Shi and Prof. Jin Wang.
References and links
1. N. Dávila, E. Merced, and N. Sepulveda, “Electronically variable optical attenuator enabled by self-sensing in vanadium dioxide,” IEEE Photonics Technol. Lett. 26(10), 1011–1014 (2014). [CrossRef]
2. P. Wei, C. Cheng, X. F. Wang, X. Shan, and X. Sun, “A high-performance hybrid current transformer based on a fast variable optical attenuator,” IEEE Trans. Power Deliv. 29(6), 2656–2663 (2014). [CrossRef]
3. S. Rudra, R. V. Hoof, and J. D. Coster, “A 2D MEMS grating based CMOS compatible poly-SiGe variable optical attenuator,” Microelectron. Eng. 105, 8–12 (2013). [CrossRef]
4. G. Zhu, B. Y. Wei, L. Y. Shi, X. W. Lin, W. Hu, Z. D. Huang, and Y. Q. Lu, “A fast response variable optical attenuator based on blue phase liquid crystal,” Opt. Express 21(5), 5332–5337 (2013). [CrossRef] [PubMed]
5. A. Unamuno, R. Blue, and D. Uttamchandani, “Modeling and characterization of a vernier latching MEMS variable optical attenuator,” J. Microelectromech. Syst. 22(5), 1229–1241 (2013). [CrossRef]
6. K. H. Koh, B. W. Soon, J. M. Tsai, A. J. Danner, and C. Lee, “Study of hybrid driven micromirrors for 3-D variable optical attenuator applications,” Opt. Express 20(19), 21598–21611 (2012). [CrossRef] [PubMed]
8. S. A. Reza and N. A. Riza, “A liquid lens-based on broadband variable fiber optical attenuator,” Opt. Commun. 282(7), 1298–1303 (2009). [CrossRef]
9. X. Tang, R. J. Li, J. K. Liao, H. Li, J. Li, and Y. Liu, “A scheme for variable optofluidic attenuator: Design and simulation the corresponding,” Opt. Commun. 305, 175–179 (2013). [CrossRef]
10. A. Dudus, R. Blue, and M. Zagnoni, “Modeling and characterization of an electrowetting-based single-mode fiber variable optical attenuator,” IEEE J. Sel. Top. Quantum Electron. 21(4), 4500209 (2015).
11. J. Hrdlička, P. Červenka, M. Přibyl, and D. Šnita, “Mathematical modeling of AC electroosmosis in microfluidic and nanofluidic chips using equilibrium and non-equilibrium approaches,” J. Appl. Electrochem. 40(5), 967–980 (2010). [CrossRef]
12. J. J. Sha and L. Y. Hou, “Research on the driving-methods of microfluidic system,” Micronanoelectron. Tech. 43(12), 586–591 (2006).
14. J. Wan, Y. Zuo, Z. B. Wang, F. Yan, L. Ge, and Z. C. Liang, “Magnetohydrodynamic microfluidic drive of ionic liquids,” J. Microelectromech. Syst. 23(6), 1463–1470 (2014). [CrossRef]
15. A. S. Basu and Y. B. Gianchandani, “Virtual microfluidic traps, filters, channels and pumps using Marangoni flows,” J. Micromech. Microeng. 18(11), 115031 (2008). [CrossRef]
16. H. H. Shen, L. Y. Chung, and D. J. Yao, “Improving the dielectric properties of an electrowetting-on-dielectric microfluidic device with a low-pressure chemical vapor deposited Si3N4 dielectric layer,” Biomicrofluidics 9(2), 022403 (2015). [CrossRef] [PubMed]
17. M. R. Javed, S. Chen, H.-K. Kim, L. Wei, J. Czernin, C.-J. C. Kim, R. Michael van Dam, and P. Y. Keng, “Efficient radiosynthesis of 3′-deoxy-3′-18F-fluorothymidine using electrowetting-on-dielectric digital microfluidic chip,” J. Nucl. Med. 55(2), 321–328 (2014). [CrossRef] [PubMed]
18. D. Yang, M. Piech, N. S. Bell, D. Gust, S. Vail, A. A. Garcia, J. Schneider, C. D. Park, M. A. Hayes, and S. T. Picraux, “Photon control of liquid motion on reversibly photoresponsive surfaces,” Langmuir 23(21), 10864–10872 (2007). [CrossRef] [PubMed]
20. A. Mazzalai, D. Balma, N. Chidambaram, R. Matloub, and P. Muralt, “Characterization and fatigue of the converse piezoelectric effect in PZT films for MEMS applications,” J. Microelectromech. Syst. 24(4), 831–838 (2015). [CrossRef]