Micromachined MEMS Transmission Line Low Loss Phase shifter

Micromachined MEMS Transmission Line Low Loss Phase shifter

1. Janardhana1, S. Pamidighantam2, N. Chattoraj1, J. S. Roy3, L. Narasimha Charyulu4, Natarajan Chidambaram5 and C. Puttamadappa6

   1ECE Department, Birla Institute of Technology, Mesra, Ranchi, India

   2BITS Pilani, Hyderabad, India.,

   3School of Electronics Engineering, KIIT University, Bhubaneswar, Orissa, India

   4Visweswaraya Technological University, Belgaum, INDIA 560 018

   5Vellore Institute of Technology, INDIA

   6Sapthagiri College of Engineering, Bangalore, INDIA 560 057

   AbstractThe design, fabrication & measurement of surface micro machined true time delay distributed MEMS transmission line (DMTL) low-loss Phase shifter is described. The phase shifter is fabricated on a glass substrate consists of a high-impedance coplanar waveguide transmission line & a bunch of MEMS bridges at height go = 1.5 µm, from the center conductor. Numerical Simulation (using EM Simulator CST Microwave Studio) results along with wafer level measured results of the fabricated structure are presented. A phase shift of 120o/cm is measured with 0.39 dB/cm loss at 14GHz.

   1. INTRODUCTION

      The field of RF-Micro Electro Mechanical Systems has seen enormous growth in recent years due to its potential for high performance in defense and commercial applications. Many microwave circuits using MEMS devices have demonstrated outstanding RF performance and low DC power consumption. MEMS Phase shifters are gradually becoming passive devices of great value due to its inherent advantageous of small size, low signal attenuation & low DC

      power dissipation. Phase shifters loss directly impacts the signals dynamic range. MEMS Phase shifters designed with MEMS switches to alter the signal paths [1,2] are inherently digital & are not suitable if large number of phase states are required. Phase array antenna for example requires high resolution phase control, which would lead to a large and lossy digital phase system. The fabrication of analog adjustable components like DMTL phase shifters requires several equal, tunable capacitances loading a transmission line [2,3]. An analog phase shifter that has high impedance CPW transmission line with bridges across the line acting as tunable capacitors are periodically loaded, which in turn changes the phase velocity and consequently varying phase shift, upon altering the height of the bridge through electrostatic actuation. This type of configuration referred to as distributed MEMS transmission line (DMTL) phase shifters are well documented by Barker [4,5]. Greg McFeetors & Michal Okoniewski have enhanced tuning range of capacitors by using oxide layers on electrodes (CPW ground) & fixed bridge to form large capacitance [6]. The DMTL Phase shifter described in this paper is

      optimized for higher phase shift per dB loss. Phase shifters have many applications in instrumentation systems, wireless communication circuits and in phased array antennas for telecommunications and radar applications. RF MEMS phase shifter has advantage over traditional phase shifters based on ferrite material, PIN diodes & FET devices. MEMS phase shifters have low loss, consume less power, & have better performance.

   2. DESIGN

      The coplanar waveguide (CPW) transmission line is used as a foundation block for phase shifter. The planar attribute of CPW lines provides convenience for MEMS structure construction & on wafer characterization. Minimizing the transmission loss is one of the primary goals. Two major factors are the reflection and dissipation losses. To minimize the reflection loss, the characteristic impedance needs to match with that of the rest of the circuitry. The dissipation loss consists of three mechanisms: 1) The skin effect, 2) Metal conductive losses, & 3) Dielectric losses. By properly choosing a good conductor (gold), a micro-scaled MEMS device can have very low skin effect and conductive losses. The dielectric loss is inherently low due to the usage of glass substrate. A CPW line with input impedance Zo= 49 is used as feed lines with dimensions, width W=120 µm & gap G =15 µm Fig. 1. The DMTLs used has unloaded impedance of Z = 96 , with center conductor width W

      =102 µm & gap G = 148 µm Fig. 1. The bridge has width w

      = 31.5 µm and gap gp = 10 µm, between each bridge, bridge thickness tb = 1 µm & seven bridges form one bunch Fig. 1. This line is loaded with 4 bunches of bridges spaced s = 518

      µm Fig. 3. The tunable capacitive loading by bridges Cb divide by bridge spacing s along with per-unit length capacitance Ct influences the line impedance (1), as well as the phase velocity vp (2). The bridge-variable MEMS capacitors have an electrostatic force as actuation principle. Nichrome DC bias pads are under the bridge. DC actuation does move the bridges by electrostatic actuation. The relationship between the actuation voltages Vp, spring constant k of the bridge, capacitor plate widths center

      conductor width W, bridge width w & height of the bridge go, is indicate in (3). Fig. 1 show the dc bias-actuation line physically separate from the transmission line resulting in a good isolation between DC and microwave lines [7]. The per unit length inductance Lt and Capacitance Ct are 546 nH & 58 pF respectively. The parallel plate bridge capacitance Cb

      = Cll + Cf =133 + 37 = 170 fF, for the bridge height go =

      1.5 µm & the loaded line impedance of DMTL line Zl = 37 . The load impedance Zl = 42 , for the bridge height go = 1 µm. Zl is extracted by noting the S11 peak of 15 dB (

      = 0.18) & knowing that the DMTL will act as a quarter wave transformer at this frequency. The substrate is glass(r

      = 4.82) with dimension 5124 µm x 2130 µm x 500 µm.

      s – Space between bridges

      Ct – Equivalent Transmission Line per Unit Length Capacitance.

      Lt – Equivalent Transmission Line per Unit Length Inductance.

      Cb – Bridge Capacitance

      W Width of the center conductor w Width of the bridge

      0 – Free space permittivity k- Bridge spring constant

      g0 Height between bridge and center conductor.

      vp phase velocity. Vp- Pull in Voltage.

      Figure 2. Layers of Electromagnetic Simulation Model

      Figure 3. DMTL Layout model.

      s Lt

      s Ct Cb

      p

      v s sLt sCt Cb

      0

      V 8k g3

      0

      p 27 Ww

      (1)

      (2)

      (3)

      Figure 4. DMTL Lumped Circuit equivalent unit section model of DMTL.

      Cb – Bridge Capacitance (between Center Conductor and Bridge)

      Lb – Bridge Inductance.

      Cp – Capacitance between bridges. (Due to bridge thickness tb and gap gp )

   3. SIMULATION

      Computer Simulation Technology AG, Microwave Studio electromagnetic simulation tool was used for simulating the model Fig. 1. The bridges are constructed with a conductor layer of gold with conductivity.4.09e+007 S/m, the CPW center conductor and ground are made of gold Fig. 2. The layout model is created using the graphical interface available in the simulation tool. The simulated and wafer level measured S – parameter results of upstate bridge position (go = 1.5 µm), without applying actuation voltage are presented below.

      Figure 1. Electromagnetic Simulation Model.

      Figure 9. Fabricated DMTL phase shifter

      Z0 eff 1 1

      Figure 5. Measured and Simulated S- Parameter Results of DMTL-

      c Z Z

      rad / m

      (4)

      go = 1.5µm

      lu

      Phase shift.

      Z0 Iput impedance.

      ld

      Zlu DMTL – Impedance when the bridge is in upper state- go 1.5 µm

      Zld DMTL Impedance when the bridge is in lower state go 1.0 µm

      eff Effective dielectric constant

      c Speed of light.

      Angular frequency

      Figure 6. Measured and Simulated S- Parameter Results of DMTL- go = 1.0µm

      Loss:

      The transmission line loss ( ) for the unloaded CPW line is found from a conformal mapping technique, and is given by Hoffmann [10].

      8.686.102 Rs eff

      X

      4 0 SK k K k'1 k

      2

      (5)

      Where,

      X= 2S

      4W 1 k

      4S1 k

      Figure 7. Simulated Phase Shift Results of DMTL- g = 1.5µm &

      W ln

      t1 k

      2 ln

      t1 k

      dB / cm

      o

      go = 1.0µm

      S = CPW line width (W+2G)

      G- CPW line gap between center conductor and ground. K(k) & K(k) = Complete elliptical integral of first kind t = thickness of the conductor

      Where t is the metal thickness, Rs is the surface resistance

      given by

      Rs

      f 0 ,

      and is the conductivity of

      Figure 8. Measured Phase Shift Results of DMTL- go = 1.5µm &

      the metal. The CPW line is of thickness t = 3 m of gold with a conductivity of approximately 4.09 x 107 Siemens / met.

      0 -free space permeability.

      go = 1.0µm

   4. FABRICATION

   5. MEASUREMENTS

      Wafer level measurements were taken using Cascade

      Surface micromachining technology process is used, unlike bulk micromachining where a substrate is selectively etched to produce structures, surface micromachining is based on the deposition and etching of different structure layers [8,9]. Corning glass substrate is used.

      Microtech probe station & Agilent PNA 8362 B Network analyzer. The simulated and measured S-parameter results are in good agreement Fig. 5. The simulated and Measured S-Parameter results for go = 1 µm has mismatch due to the variation in height of the various bridges along fabricated structure Fig. 6. At 14 GHz. the calculated phase shift is 90o/cm, using (4), the simulated phase shift is 97o/cm. & the

      experimental / measured phase shift is 120o/cm. The discrepancy may be due to non-uniform height of the bridge along the structure. The loss at 14 GHz. S21 is 0.2 dB (Measured). The bridge has length of 762µm, the bridge inductance Lb = 235 pF, thickness of the bridge tb = 1µm, the gap gp between bridges is 10 µm & the parallel plate capacitance due to this gap between bridges is 5.31 aF (5.31e-18). The fringe capacitance is 25% of parallel plate capacitance [5]; hence the total capacitance Cp =6.63 aF. The effect of self resonance of the bridge inductor in parallel with capacitance is the cause of dip in S11 at 18.7 GHz. (Simulated) & at 15.7 GHz. (Measured) – Fig. 5.

      Figure 10: Measured, Simulated & Calculated Loss versus Frequency for Unloaded CPW line & Loaded DMTL on Glass with center conductor 102

      µm wide (Zo = 96 )

      The Fig. 10 shows the line loss of CPW unloaded line calculated from (5), simulated line loss of DMTL and measured line loss of DMTL at go = 1.5 µm & 1.0 µm. The measured line loss at 1 µm over a frequency range is better than the measured line loss at 1.5 µm since the line impedance match is better. Zl = 37 at go = 1.5 µm & Zl = 42 at go = 1.0 µm, the calculated line loss is less compared to the simulated loss as it does not take care of the standing waves on transmission line. The standing waves increase the loss of the line. The rapid increase in line loss at 14 GHz for measure line loss at go = 1.5 µm, at 15 GHz for simulated line loss & measured line loss at go = 1.0 µm is due to self resonance of bridge inductor at 14 GHz & 15 GHz.

      TABLE I. DMTL Comparison

      Device	Loss	Phase Shift / cm	Phase Shift / dB	Frequency
      Barker[4]	1.7 dB/cm	120° / cm	70° / dB	@ 40 GHz
      W.Palei [11]	2.0 dB/cm	150° / cm	75° / dB	@ 20 GHz
      This work	0.39 dB/cm	120° / cm	307° / dB	@ 14 GHz

   6. CONCLUSION

The measured results indicate a phase shift 120o /cm, 0.39 dB/cm losses at 14GHz with Figure of Merit of 307° / dB – Phase Shift / dB. This is the lowest loss for maximum phase shift reported in the literature. Work is underway to push the self resonance frequency beyond 30 GHz, by variation of bridge inductance, so that the reflections from the self resonance have less effect on the return loss in band up to 20 GHz.

ACKNOWLEDGMENT

The authors would like to thank Bharat Electronics, Bangalore, India, for their cooperation, by providing their facility for fabrication & measurements. We would also thank Prof. K C Gupta, university of Colorado for his invaluable guidance.

REFERENCES

1. B.Lakshminarayana and T.Weller, MEMS Phase Shifters Using Cascaded Slow-wave Structures for Improved Impedance Matching and/or Phase Shift, in Proc IMS04 Conf., Fort Worth. TX Jun. 2004. pp 725-728.

2. Rebeiz, G. M., G. L. T. Tan, and J. S. Hayden, \RF MEMS phase shifters: Design and applications," IEEE Microwave Magazine,Vol. 3,72-81, Jun. 2002.

3. A.S.Nagra and R. A. York, Distributed analog phase shifter with low insertion loss, IEEE Trans. Microwave Theory and Tech., Vol. 47, pp. 1705-1711. Sept. 1999.

4. N. Scott Barker and Gabriel M. Rebeiz, Distributed MEMS True- Time Delay Phase shifters and Wide Band Switches in IEEE Transactions on Microwave Theory and Techniques, VOL.46 NO.11, November 1998.

5. Rebeiz, G. M., RF MEMS Theory, Design and Technology, Wiley- Interscience, New York, 2003.

6. Greg McFeetors, Michal Okoniewski, Distributed MEMS Analog Phase Shifter With Enhanced Tuning, IEEE Microwave AND WIRELESS COMPONENTS VOL.16, NO.1, January 2006.

7. D. Girbau and A. Lazaro, Extended tuning range RF mems variable capacitors using electrostatic and electro thermal actuators, SPIE Proceedings Photonics West Microm and Microfab, pp. 59-70, Jan. 2004.

8. K. Srinivasa Reddy, Madhusudhan Nadgir, G. S. Sharma, Ranjay Laha, K. Natarajan and Sayanu Pamidighantam, Novel corrugated surface micro machined beams for mixed mode MEMS varactor application, Proceedings of IMAPS, National Conference, Dec 2006, Hyderabad.

9. Laha, R., B. Sriram, A. T. Kalghatki, K. Natarajan, and S. Pamidighantam, \Design, development and characterization of surface micromachined passive components for radar applications," Proceedings of International Radar Symposium (IRSI),Bangalore, India, Dec. 2005.

10. R. K. Hoffmann, " Handbook of Microwave Integrated Circuits ", Artech House, Inc., Norwood, MA, 1987.

11. W. Palei, A.B. Yu and A.Q Liu , Distributed MEMS Transmission line (DMTL) for true time delay (TTD) phase shifter applications , www.icee-con.org/papers/2004/415.pdf.

12. V.Janardhana, R.G.Kulkarni, J.S Roy Feasibility Study of MEMS Distributed Phase Shifters for Phased Array, ISM December 2006.

13. Janardhana, V., J. S. Roy, S. Pamidighantam, and R. G. Kulkarni," Analysis of surface micromachined RF MEMS phase shifters," IEEE Aerospace & Electronic Systems Magazine, Vol. 23, No. 5,32{35, May 2008.

14. Jahanbakht,M.,M. Naser-Moghadasi,andA. A. Lot¯ Neyestanak,\Low actuation voltage Ka-band fractal MEMS switch," Progress In Electromagnetics Research C, Vol. 5, 83{92, 2008.

15. Topalli, K., M. Unlu, H. I. Atasoy, S. Demir, O. Aydin Civi, and T. Akin, \Empirical formulation of bridge inductance in inductivelytuned RF MEMS shunt switches," Progress In Electromagnetics Research, Vol. 97, 343-356, 2009.

16. Raedi, Y., S. Nikmehr, and A. Poorziad, \A novel bandwidth enhancement technique for X-band RF MEMS actuated reconfigurable reflect array," Progress In Electromagnetics Research, Vol. 111, 179- 196, 2011.

17. V.Janardhana, Sayanu Pamidighantam, R.G.Kulkarni, J.S.Roy "Analysis of MEMS Distributed Phase Shifters," ISMOT, Rome Italy, December. 17-21, pp.527- 532, 2007.