Design of Clamped-Clamped Beam Resonator in Thick-Film Epitaxial Polysilicon Technology D. Galayko, A. Kaiser, B. Legrand, L. Buchaillot, D. Collard, C. Combi IEMN-ISEN UMR CNRS 8520 Lille, France ST MICROELECTRONICS Milano, Italy ESSDERC 2002, Florence, Italy
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 2
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 3
Introduction Motivation : Realization of a single-chip RF receiver RF filter IF filter Baseband processing RF IF Baseband LO1 LO2 Desirable performances for IF filters: Center frequency 100-400 MHz High selectivity (250 khz passband ) Integrable on silicon ESSDERC 2002, Florence, Italy Slide 4
Introduction Proposal: micromachined micromechanical filters Advantages: Manufactured on silicon integrable with electronic ciruits Small size Low power consumption High Q (up to 100.000 in vacuum) ESSDERC 2002, Florence, Italy Slide 5
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 6
Resonator design Principle of mechanical filtering: resonating element + input/output capacitive transudcers v i i i Capacitive transducers C(t) V P v i i i i o v o v i i i i o v o v o i o V P Resonating elements Operating principle: δv i δf i δx δc δq i o ESSDERC 2002, Florence, Italy Slide 7
Resonator design Different design issues Gap Resonator Input C 01 C X L X R X Output C 02 Equivalent small-signal model Electrode Anchor Clamped-clamped beam resonator in a thick-film technology 1 Tranduction factor: 2 gap 4 Impedance at resonance R X : gap Transducer s gap width: should be small 3D CoventorWare simulation of a beam L=40µm, W=2µm, H=15µm: resonance frequency Fo=9.8 MHz ESSDERC 2002, Florence, Italy Slide 8
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 9
Contact to substrats 15 µm Thick-film Thick-film technology technology presentation Device structure Metal Anchor/burried layer contact Trench Structural layer (epitaxial silicon 15 µm) Substrate Oxide Oxide Burried polysilicon ESSDERC 2002, Florence, Italy Slide 10
Thick-film technology Mask Dry etching Problem: impossibility to achieve a submicronic transducer s gap 15 µm 1.8 µm min 0.6 µm Underetching 3.0 µm Minimal trench width Minimal available gap: ~ 3.0 µm Maximal acceptable value: ~ 0.5 µm Desirable value: < 0.2 µm Conclusion: gap should be reduced! ESSDERC 2002, Florence, Italy Slide 11
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 12
Post-fabrication gap reduction Solution: moving the signal electrode close to the resonator Post-fabrication gap reduction method Spring Initial gap Signal electrodes Stoppers Resonator Vin Emot Eres Actual gap Motor electrodes Actual gap=initial gap gap-to-stoppers not sensitive to underetching ESSDERC 2002, Florence, Italy Slide 13
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 14
Gap reducing motor test Two devices are designed: identical resonators, but different actual gaps. Parameters of the designed resonators: clamped-clamped beam length 40 µm, width 2 µm actual gaps of 0.2 and 0.4 µm expected resonance frequency ~10 MHz 500 µm ESSDERC 2002, Florence, Italy Slide 15
Gap reducing motor test SEM picture of one of the fabricated devices with gap reducing motor Motor springs Input-output electrodes Motor electrodes Resonator Stoppers ESSDERC 2002, Florence, Italy Slide 16
Gap reducing motor test Test of gap reducing motor: comparison of devices with 0.2 and 0.4 µm gap Transmission, db gap=0.2 µm Vbias=11 V gap=0.4 µm Vbias=40 V Frequency, Hz Resonance frequency: ~9.8 MHz Equal impedances for 40 and 11 V bias voltages: R X =283 kω Motor activation voltage: 29 V (unbiased resonators) Without gap reducing (with gap of 3.0 µm), a 2500 V bias voltage needed ESSDERC 2002, Florence, Italy Slide 17
Gap reducing motor test Limitation of the gap reduction method: non-ortagonality of trench walls Designed gap width Slope of etched trench walls: 89 Effective gap value is increased 15 µm Designed gap width + 0.5 µm Minimal achievable effective value (for 0.1 µm designed gap): ~ 0.25 µm ESSDERC 2002, Florence, Italy Slide 18
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 19
Test of resonators Dimensions of the designed clamped-clamped beam resonators and test results N*** 1 2 3 5 4 6 7 Length, µm 30 40 50 80 50 80 80 Width, µm 1.8 1.8 1.8 1.8 4.8 2.8 4.8 Res. Frequency, MHz 16.4 9.8 6.23 2.30 12.8 3.65 5.65 Q-factor 530 2400 4390 16900 924 4000 3645 ESSDERC 2002, Florence, Italy Slide 20
Test of resonators Plot of the quality factor versus resonance frequency for all tested resonators Experimental data * Q 50000 fo 3 2 Fitted plot Thick beams ESSDERC 2002, Florence, Italy Slide 21
Test of resonators Quality factor 16000 14000 12000 10000 8000 6000 4000 Quality factor versus air pressure L=40 µm, W=1.8µm, fo=9.8 MHz L=80 µm W=1.8 µm, fo=2.3 MHz 2000 0 10-4 10-2 10 0 10 2 10 4 Pressure, Torr Atmospheric pressure ESSDERC 2002, Florence, Italy Slide 22
Outline Introduction Resonator design Thick-film technology Post-fabrication gap reduction Gap reducing motor test Test of resonators Conclusions ESSDERC 2002, Florence, Italy Slide 23
Conclusions Design and test of clamped-clamped beam resonators in thickfilm epipoly technology; Original technique for post-fabrication gap reduction ; Empirical law established for qualty factor evolution with resonance frequency of resonators ; Highest reached resonance frequency with reasonable quality factor : 10 MHz (Q=2400) ESSDERC 2002, Florence, Italy Slide 24
Conclusions Design and test of clamped-clamped beam resonators in thickfilm epipoly technology; Original technique for post-fabrication gap reduction ; Empirical law established for qualty factor evolution with resonance frequency of resonators ; Highest reached resonance frequency with reasonable quality factor : 10 MHz (Q=2400) ESSDERC 2002, Florence, Italy Slide 25
Conclusions Design and test of clamped-clamped beam resonators in thickfilm epipoly technology; Original technique for post-fabrication gap reduction ; Empirical law established for qualty factor evolution with resonance frequency of resonators ; Highest reached resonance frequency with reasonable quality factor : 10 MHz (Q=2400) ESSDERC 2002, Florence, Italy Slide 26
Conclusions Design and test of clamped-clamped beam resonators in thickfilm epipoly technology; Original technique for post-fabrication gap reduction ; Empirical law established for qualty factor evolution with resonance frequency of resonators ; Highest reached resonance frequency with reasonable quality factor : 10 MHz (Q=2400) ESSDERC 2002, Florence, Italy Slide 27
Conclusions Perspectives Packaging of resonators and filters under vacuum: wafer-level encapsulation; High-order filter design: coupled-resonator architectures v i v 0 V P1 V P3 V P2 ESSDERC 2002, Florence, Italy Slide 28
Conclusions Perspectives Packaging of resonators and filters under vacuum: wafer-level encapsulation; High-order filter design: coupled-resonator architectures v i v 0 V P1 V P3 V P2 ESSDERC 2002, Florence, Italy Slide 29
Conclusions Perspectives Packagig of resonators and filters under vacuum: wafer-level encapsulation; High-order filter design: coupled-resonators architectures v i v 0 V P1 V P3 V P2 Acknowlegments This work was supported by the EU under contract number IST-1999-10945 ESSDERC 2002, Florence, Italy Slide 30
Thank you for your attention ESSDERC 2002, Florence, Italy Slide 31