PhD work

PhD work carried for 3 years at NTNU.

Transcript

1. Research experience
   PhD studies
   Trondheim 2010-2013
   (Unfinished)
   Detecting ultasonic
   vibrations using
   Si-integrated opical
   sensor.
   Oleg Żero
   

   View Slide

2. We have a Capacitive Ultrasonic Micromachined Transducer (CMUT) array.
   How do we make sonic-pixels for 30MHz detection?
   The pixels must be of surface area that is less than 10 x 10 µm2 .
   Si-integrated optical sensor
   Oleg Żero
   Problem outline.
   Why not using light?
   

   View Slide

3. Si-integrated optical sensor
   Oleg Żero
   My initital idea was to use gratings.
   Use Silicon-on-Insulator (SOI)
   technology for high index contast.
   Low loss at λ = 1.55 μm.
   Silicon will be harmless for human body
   (intravascular applications in mind).
   

   View Slide

4. Si-integrated optical sensor
   Oleg Żero
   My initital idea was to use gratings.
   Use Silicon-on-Insulator (SOI)
   technology for high index contast.
   Low loss at λ = 1.55 μm.
   Silicon will be harmless for human body
   (intravascular applications in mind).
   x
   y
   x
   y’
   x
   y
   z
   x
   y
   x
   y’
   Grating to couple the light out, which will reflect by a
   membrane, and interfere with the evanescently coupled
   beam.
   

   View Slide

5. Si-integrated optical sensor
   Oleg Żero
   I started looking for more compact architectures.
   Use Silicon-on-Insulator (SOI)
   technology for high index contast.
   Low loss at λ = 1.55 μm.
   Silicon will be harmless for human body
   (intravascular applications in mind).
   Grating-based mechanism was hardly scalable...
   

   View Slide

6. Si-integrated optical sensor
   Oleg Żero
   Then I realized plasmonics offers more confinement.
   COMSOL Multiphysics 4.2 a.
   Cantilever must be metal.
   Starting point is to model it as Perfect Electric
   Conductor (PEC)
   Sensitivity through effective index.
   Response through inter-modal interference.
   SiO
   2
   Si
   air
   continuity b. co.
   PEC
   PEC
   Model
   I started to search for optimal
   conditions for power transfer.
   

   View Slide

7. Si-integrated optical sensor
   Oleg Żero
   Then I realized plasmonics offers more confinement.
   Assuming ideal MZI, slab waveguide
   The relative MZI sensitivity would be
   ~4%/nm change in light intensity with
   respect to change in cantilever’s position!
   TE fundamental mode sensitivity TM fundamental mode sensitivity
   

   View Slide

8. Si-integrated optical sensor
   Oleg Żero
   That lead to MMI coupler as sensor.
   Again, breaking the symmetry turned out
   to promise high sensitivity.
   SiO
   2
   Si air
   continuity b. co.
   PEC
   PEC
   u
   

   View Slide

9. Si-integrated optical sensor
   Oleg Żero
   That lead to MMI coupler as sensor.
   I began to construct a new model,
   in order to describe the new system.
   SiO
   2
   Si air
   continuity b. co.
   PEC
   PEC
   u
   xy
   xy
   symmetric
   or
   asymmetric
   PML
   

   View Slide

10. Si-integrated optical sensor
    Oleg Żero
    That lead to MMI coupler as sensor.
    I began to construct a new model,
    in order to describe the new system.
    x
    y
    z
    xy
    xy
    xy
    That was it!
    It gave me the 2nd conference.
    

    View Slide

11. Si-integrated optical sensor
    Oleg Żero
    That lead to MMI coupler as sensor.
    MMI’s length was 4.1 μm. The width was 1.5 μm
    and the waveguide’s height was 200nm.
    The asymmetric case showed steep response: 2.4%/nm around g = 50 nm.
    Simple and compact.
    We wanted to fabricate it too!
    O. Zero et al, ‘Investigation of parallel coupling mechanisms in silicon
    integrated chip sensors’, Proc. of SPIE, Vol. 8226 (2012). ISBN:
    9780819489098.
    O. Zero et al., ‘Sensitivity limits of guided modes in silicon integrated
    waveguide-based sesnors’, Proc. SPIE 8561, Advanced Sensor Systems and
    Applications V, 85610P
    

    View Slide

12. Si-integrated optical sensor
    Oleg Żero
    We wanted to fabricate it too.
    Jens Høvik, ‘’Photonic crystal waveguides’’,
    MSc thesis NTNU 2012
    We wanted to
    • grow an a-Si:H based SOI
    • transfer waveguide patterns with EBL
    • transfer patterns with dry etch
    We lacked...
    • experience – no previous Si-photonics at NanoLab,
    • facility – no characterization equipment,
    • time...
    

    View Slide

13. Si-integrated optical sensor
    Oleg Żero
    At this point I realized, I have...
    • Spent 2.5 years out of 4 in the programme.
    • Have 2 conferences.
    • Spent 9 months just teaching students.
    • Spent 6 months in the clean room just getting started.
    • None of my older colleagues has finished the programme (even now!)
    

    View Slide