Making Optical MEMS Sensors more compact using Organic Light Sources and Detectors

Thilo Sauter, Wilfried Hortschitz, Harald Steiner, Michael Stifter

Center for Integrated Sensor Systems Danube University Krems Wiener Neustadt, Austria

{thilo.sauter, wilfried.hortschitz, harald.steiner, michael.stifter}@donau-uni.ac.at

Hsin Chiao1, Hsiao Wen Zan1, Hsin-Fei Meng2, Paul Chao3

1Department of Photonics, 2Institute of Physics, 3Department of Electrical and Computer Engineering,

National Chiao Tung University Hsinchu, Taiwan

{hsiaowen, meng, pchao}@ mail.nctu.edu.tw

Abstract—Vibration and displacement sensors need to be compact for many applications in automation or consumer electronics, and microelectromechanical structures are a convenient way to implement such sensors. For these MEMS devices, optical readout methods have proven to be superior to capacitive or piezoresistive strategies in terms of sensitivity as well as noise and interference immunity, however the integration of light sources and detectors is not easily possible. This paper presents an approach to use organic optoelectronic devices for the readout. OLED and OPD (organic photo detector) are structured on the glass substrate and cover encapsulating the MEMS devices, allowing for a tightly integrated sensor based on vertical light flux modulation by a horizontally moving proof mass. The paper describes the principle sensor structure as well as the fabrication of suitable organic devices. First test results show that the approach is feasible.

Keywords—MEMS, vibration sensor, displacement sensor, OLED, OPD, smart sensor

I. INTRODUCTION Measuring vibrations and displacements is crucial for

many automation and control applications, for health monitoring of machine parts (such as rotating elements) or building structures, but also for environmental surveillance including seismic monitoring. In several of these applications, miniaturization is important [1, 2]. Thanks to silicon micromachining technology, small sized and cheap inertial sensor devices emerged on the market, that are suited for everyday consumer applications like automotive traction control or camera jitter compensation. In addition to size, also robustness is an issue, as many of these sensors are used in harsh environments. “Harsh” in this context often means exposure to strong fields, but also to mechanical stress and shocks that can be significant even in consumer devices such as smart phones. The design of micro-electromechanical (MEMS) devices is therefore always a tradeoff between robustness and sensitivity, even more so when considering the readout principle used to extract the measurement signal from the mechanical structure.

The classical readout strategies for MEMS displacement and vibration sensors are piezoresistive or –electric and capacitive. However, they have limitations. Piezoelectric transducers are complex in terms of fabrication processes, whereas piezoresistive devices as all resistors suffer from high 1/f noise, which limits resolution, and are also temperature-dependent [3, 4]. For high-precision applications, capacitive readout concepts are therefore usually the method of choice, where a set of fixed and movable electrodes form a capacitor whose capacitance is measured to determine the static displacement or vibration amplitude [5, 6, 7]. However, they have limitations for practical use, too: To achieve high sensitivity and resolution, distances between the electrodes have to be small, which limits the measurement range. Especially for vibration sensors, the spring suspension must be sufficiently stiff in order to avoid a detrimental pull-in effect due to electrostatic forces when the gap becomes too small [8, 9], which in turn prohibits low mechanical resonance frequencies that might be interesting for certain applications.

Optical readout principles were proposed as seemingly ideal solution [10-14]. They are contactless, require no electrical leads on the movable parts of the sensor, are free from electrostatic force-feedback, and they also allow for galvanic insulation between the actual transducer and the electronics required for generation and detection of the optical signal modulated by the MEMS resonator. Nevertheless, they have drawbacks in that the light source and detector can hardly be integrated with the MEMS device because of the different semiconductor materials involved. Hence, the miniaturization is limited when using solid-state optoelectronic devices. This paper presents an idea to overcome this deficiency by the use of organic optoelectronics that can be integrated much tighter with the MEMS structures. In the subsequent sections we describe the starting points for the optical MEMS vibration and displacement sensor, for the technology needed to produce the organic devices, and how to combine both to a sensor device that is much more compact as its solid-state counterparts.

II. OPTICAL MEMS SENSOR The transduction principle applied in the sensor is simple

light intensity modulation derived from displacement or inertia. To that end, two corresponding stencil masks are placed in front of each other as shown in Fig. 1. One mask is formed by openings in a spring-suspended mass that can laterally move in-plane. The second mask is fixed with respect to the frame holding the movable mass and is typically implemented as an opaque grid deposited on a glass substrate. It would also be possible to coat the photo detector directly with a structured metal layer and implement the second mask this way. Using this setup, by illuminating the structure from the top side, the transmitted light intensity can be measured with the detector placed at the bottom.

Fig. 1. Cross section of the mechanical structure of a MEMS vibration sensor with optical readout [13].

The transduction principle is relatively simple: The light impinging on the photo detector is modulated by the in-plane relative movements of the two arrays of apertures. The concept is easily understood using the simple ray optics approach sketched in Fig 1. We aim at sensor configurations facilitating highly sensitive devices. Therefore, the aperture arrays are aligned such that at the resting position, the light flux is blocked. This way, small lateral displacements already create a significant relative signal change at the photo detector. If the apertures have a rectangular shape as in Fig. 2, the light flux shows a linear dependency on the displacement, at least within the periodicity of the pattern. Nevertheless, the transduction principle permits also more complex aperture patterns with non-rectangular shape that may be used to extend the dynamic range or to implement dynamic compression already on the mechanical level [15].

The spring-suspended mass system of the sensor shown in Fig. 2 is fabricated using a deep reactive ion etching (DRIE) in the device layer of a silicon-on-insulator (SOI) wafer. Monocrystalline Si has excellent mechanical properties, in particular high stability, making it well suited as material for implementing springs. After the DRIE process, the structure is released from the handle layer by wet etching, and the remaining sacrificial SiO2 layer is removed. The stationary aperture array is fabricated as chromium layer evaporated on a Pyrex glass wafer and structured by lift-off techniques.

Fig. 2. Technology prototype of the two aperture arrays [12].

The two wafers are connected by a custom SU-8 bonding process which creates a 10-µm spacer between the two masks. For the optical readout, a standard surface-mounted device (SMD) LED and a photo detector are attached at the top and bottom of the sandwich structure, respectively.

III. IMPROVED SENSOR CONCEPT The optical readout principle was tested in a prototype

implementation and proved to be feasible and sensitive [13]. Nevertheless, the sensor concept shown in Fig. 1 has some drawbacks. First of all, even if small SMD optoelectronic devices are used, the overall device is quite bulky, as shown in Fig. 3.

Fig. 3. Exploded view of the sensor with MEMS aperture array and SMD-type LED and photodetector. Actual implementations have a larger MEMS structure (about double length and width).

This problem could be alleviated by using bare die devices without housing; however this entails a new difficulty: The active area of an LED is relatively small, and it can be

considered a point source for the light emission. In devices with housing, the package is often implemented as a concave, almost parabolic reflector that parallelizes the light flux. Using bare LED dies reduces the overall sensor size, but the ray optics sketched in Fig. 1 no longer hold, particularly not if the aperture arrays are made larger to increase sensitivity. Especially towards the rim of the aperture array, the precision of the optical readout will degrade. In principle, the Si holes could be considered as optical wave guides, but due to the DRIE process there is a certain residual roughness in the walls of the holes that will lead to partial diffraction of the impinging light, thereby reducing the sensitivity.

In order to address both problems, planar light sources and detectors are desirable. The idea put forward in this paper is to use organic devices for this purpose, as sketched in Fig. 4. The organic LEDs (OLEDs) and photo detectors (OPDs) can be applied directly on the surface of the glass carrier and cover of the MEMS structure. The benefits of such a structure are obvious: While the basic design is similar to the one shown in Fig. 1, additional bonding processes between devices are not necessary. Using OLEDs/OPDs, the sensor can be made much more compact and better integrated, moreover the light distribution will be more homogeneous.

Fig. 4. Schematical cross section of a MEMS structure with integrated organic optoelectronic devices.

The problem to be solved is primarily the fabrication of the organic devices on the glass substrates. OLEDs and OPDs, particularly when processed in a solution, are very well suited for fabricating thin, large-area devices. For the application considered here, they must be also structured to yield the desired patch size.

IV. OPTOELECTRONIC DEVICE TECHNOLOGY The schematic diagram of a solution-processed green

OLED is shown in Fig. 5(a). The glass substrate is first patterned with a transparent ITO electrode. The PEDOT:PSS polymer is used as conducting layer to interconnect to the active layer: TCTA is used as hole transporting layer, TPBi as electron transporting layer to form a structure similar to a pn junction in semiconductors. To achieve the green emission, Ir(mppy)3 is used embedded in a CBP polymer. The top electrode is made of Al. The OLED structure is produced by blade coating technology which can be used for multi-layer solution coating without dissolution problems [16, 17]. A good efficiency as high as 25 cd/A can be obtained. Fig. 5(b)

shows the energy band diagram and the materials of the green OLED. A photograph of the device under operation is shown in Fig. 5(c).

Fig. 5. (a) layer structure, (b) enenegy band diagram, and (c) photograph of the solution-processed green OLED.

The structure and band diagram of the solution-processed OPD are shown in Fig. 6. A conventional donor-acceptor blended system using p-type P3HT and n-type PCBM is used as the active layer [18, 19]. The detail process steps are shown in Fig. 7. First, PEDOT:PSS is coated on an ITO patterned glass substrate. Then, blended P3HT:PCBM is coated and annealed at 140°C. Aluminium is evaporated to form the top electrode. To facilitate the alignment between OPD and MEMS structure, it is beneficial to remove the organic layers outside the active region. Hence, oxygen plasma etching is used to etch through the organic layer to form a final device as shown in Fig. 7(e).

Fig. 6. (a) schematic layer diagram and (b) energy band diagram of the solution-processed OPD.

Fig. 7. Process steps to fabricate OPD on glass substrate.

glass

glass

OLED

OPD

Cr mask

SoI wafer

MEMS

Fig. 8 shows the incident photon-to-electron conversion efficiency (IPCE, also known as external quantum efficiency, EQE) of the produced OPD. After oxygen plasma etching, a 10% drop of the IPCE (at 450-600 nm) may be due to the oxygen doping in P3HT. Further examination to explain the mechanism is required in future work. The proposed OPD with 50% IPCE is reasonably sensitive. With the feasibility of OLEDs and OPDs on simple glass substrates, in future work, the OPD and the OLED can successfully be integrated with the MEMS structure to realize the targeted compact displacement sensor.

Fig. 8. The incident photon-to-electron conversion efficiency (IPCE) of the OPDs with and without post-plasma-etching.

V. CONCLUSIONS The combination of MEMS structures and organic

optoelectronic devices is very promising to achieve highly integrated, compact displacement and vibration sensors. First results show that the approach is feasible. Next steps in this research will be the integration of the different fabrication technologies. Possible further improvements can be made by an even more advanced structuring of the light source. It seems possible that the OLED can be structured in the same shape (but negative structure) as the fixed Cr mask currently evaporated on the top cover. This will lead to a smaller active area of the OLED, thus less power consumption, and the fixed mask can be omitted.

Open technological challenges are the lifetime of the devices and the related passivation technology. Organic semiconductors are sensitive to oxidation and humidity. Glass sealing with nitrogen atmosphere is usually employed to encapsulate OLEDs and OPDs. Future research is needed for thin film passivation methods to make the integrated device even more compact.

REFERENCES [1] C. Brigante, A. Basile, N. Abbate, A. Faulisi, and S. Sessa, “Towards

miniaturization of a MEMS based wearable motion capture system,” IEEE Trans. Ind. Electron., vol. 58, no. 8, pp. 3234–3241, Aug. 2011.

[2] M.Kraft, “Micromachined inertial sensors state of the art and a look into the future,” IMC Measurement and Control, vol. 33, no. 6, pp. 164-168, 2000.

[3] A. Barlian, W.-T. Park, J. Mallon, A. Rastegar, and B. Pruitt, “Review: Semiconductor piezoresistance for microsystems,” Proc. IEEE, vol. 97, no. 3, pp. 513–552, Mar. 2009.

[4] S. Tadigadapa and K. Mateti, “Piezoelectric MEMS sensors: State-of-the-art and perspectives,” Meas. Sci. Technol., vol. 20, no. 9, p. 092001, Sep. 2009.

[5] B.V. Amini and F. Ayazi, “Micro gravity capacitive silicon on insulator accelerometers,” Journal of micromechanics and microengineering, vol. 15, pp. 2113-2120, 2005.

[6] J.Bernstein, R.Miller, W.Kelley and P.Ward, “Low Noise MEMS vibration sensor for geophysical applications,” Journal of Microelectromechanical systems, vol. 8, 4, pp. 433-438, 1999.

[7] N. Barbour and G. Schmidt, “Inertial sensor technology trends,” IEEE Sensors J., vol. 1, no. 4, pp. 332–339, Dec. 2001.

[8] A. Fargas-Marques and A. Shkel, “On electrostatic actuation beyond snapping condition,” in Proc. IEEE Sensors, p. 4, Nov. 2005.

[9] T. Chen, L. Chen, L. Sun, and X. Li, “Design and fabrication of a four-arm-structure MEMS gripper,” IEEE Trans. Ind. Electron., vol. 56, no. 4, pp. 996–1004, Apr. 2009.

[10] U. Krishnamoorthy, R. Olsson, G. Bogart, M. Baker, D. Carr, T. Swiler, and P. Clews, “In plane MEMS based nano g accelerometer with sub wavelength optical resonant sensor,” Sens. Actuators A, Phys., vol. 145/146, pp. 283–290, Jul./Aug. 2008.

[11] N. Hall, M. Okandan, R. Littrell, D. Serkland, G. Keeler, K. Peterson, B. Bicen, C. Garcia, and F. Degertekin, “Micromachined accelerometers with optical interferometric read-out and integrated electrostatic actuation,” J. Microelectromech. Syst., vol. 17, no. 1, pp. 37–44, Feb. 2008.

[12] W. Hortschitz, H. Steiner, M. Sachse, M. Stifter, F. Kohl, J. Schalko, A. Jachimowicz, F. Keplinger, and T. Sauter, “An Optical In-Plane MEMS Vibration Sensor,” IEEE Sensors Journal, vol. 11, no. 11, pp. 2805-2812, 2011.

[13] W. Hortschitz, H. Steiner, M. Sachse, M. Stifter, F. Kohl, J. Schalko, A. Jachimowicz, F. Keplinger, and T. Sauter, “Robust Precision Position Detection With an Optical MEMS Hybrid Device,” IEEE Trans. on Industrial Electronics, vol.59, no.12, pp. 4855-4862, 2012.

[14] O. Bochobza-Degani, R. Yechieli, S. Bar-Lev, U. Yehuda, and Y. Nemirovsky, “Characterization of a novel micromachined accelerometer with enhanced-MIDOS,” 12th Int. Conf. Solid-State Sensors, Actuators, Microsystems, TRANSDUCERS, Jun. 2003, vol. 2, pp. 1395–1398.

[15] W. Hortschitz, F. Kohl, H. Steiner, M. Sachse, J. Schalko, and F. Keplinger, “Design of linear and nonlinear hybrid optical MEMS displacement sensors,” SPIE—Smart Sensors, Actuators, and MEMS V, 2011, vol. 8066, pp. 806 60P-1–806 60P-9.

[16] Yu-Fan Chang, Yu-Chian Chiu, Han-Cheng Yeh, Hao-Wen Chang, Chun-Yu Chen, Hsin-Fei Meng, Hao-Wu Lin, Heh-Lung Huang, Teng-Chih Chao, Mei-Rurng Tseng, Hsiao-Wen Zan, and Sheng-Fu Horng, “Unmodified small-molecule organic light-emitting diodes by blade coating,” Organic Electronics, vol. 13, p. 2149, 2012.

[17] Chun-Yu Chen, Hao-Wen Chang, Yu-Fan Chang, Bo-Jie Chang, Yuan-Sheng Lin, Pei-Siou Jian, Han-Cheng Yeh, Hung-Ta Chien, En-Chen Chen, Yu-Chiang Chao, Hsin-Fei Meng, Hsiao-Wen Zan, Hao-Wu Lin, Sheng-Fu Horng, Yen-Ju Cheng,Feng-Wen Yen, I-Feng Lin, Hsiu-Yuan Yang, Kuo-Jui Huang, and Mei-Rurng Tseng, “Continuous blade coating for multi-layer large-area organic light-emitting diode and solar cell,” J. Appl. Phys.,vol. 110, 094501, 2011.

[18] En-Chen Chen, Cheng-Yang Shih, Ming-Zhi Dai, Han-Cheng Yeh, Yu-Chiang Chao,Hsin-Fei Meng, Hsiao-Wen Zan, Wei-Ren Liu, Yi-Chen Chiu, Yao-Tsung Yeh, Chien-Jen Sun, Sheng-Fu Horng, and Chain-Shu Hsu, “Polymer infrared proximity sensor array,” IEEE Transactions Electron Devices, vol.58, no.4, pp. 1215-1220, 2011.

[19] En-Chen Chen, Shin-Rong Tseng, Yu-Chiang Chao, Hsin-Fei Meng, Chih-Feng Wang, Wen-Chang Chen, Chian-Shu Hsu, and Sheng-Fu Horng, “Polymer infrared photo-detector with high sensitivity up to 1100 nm,” Synthetic Metals, vol. 161, p. 618, 2011.