Silicon based Integrated Photonic Devices for optical Interconnects and Biosensing Applications Ray T. Chen Microelectronics Research Center Electrical and Computer Engineering Department, University of Texas at Austin Austin, TX, 78758, USA *
[email protected]
4/28/2015
This Research is supported by NSF AFOSR US Army DARPA DOE EPA NIH Omega Optics State of Texas 2
Outline 1. Introduction 2. Si Photonics for Optical Communications and Interconnects a. Free Space Optical Phased Array for Beam Steering b. On-chip Optical Interconnects with High Quasi-3D Interconnections c. On-chip 3D Optical Interconnects d. High-speed Si/EO Polymer PCW modulator 3. Si Photonics for Biosensors a. Early cancer detection b. Open sensing platforms 3
Pros and Cons of Silicon Photonics
Why Optical Interconnects
K. Bergman, Columbia 5
Silicon Photonics
6
Outline 1. Introduction 2. Si Photonics for Optical Communications and Interconnects a. Free Space Optical Phased Array for Beam Steering b. On-chip Optical Interconnects with High Quasi-3D Interconnections c. On-chip 3D Optical Interconnects d. High-speed Si/EO Polymer PCW modulator 3. Si Photonics for Biosensors a. Early cancer detection b. Open sensing platforms 7
Motivation for On-chip Optical Interconnection 100
Speed [ GHz ]
Limitation of Cu / low-K Interconnection - Crosstalk : EMI effects - Skin effects : Degradation of electrical signal - Dielectric loss increases as clock rate goes up - Power Dissipation Future System(2010 ~ ) - High Performance on-chip/off-chip - Substantial bandwidth - Scale-down of CMOS
[ITRS-2004]
10
On-chip local clock Chip-to-board(off-chip) speed
1 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020
Year
High Clock Speed & High Integration Density Optical Interconnection; Promising Solution for High Speed & High Density Interconnection Disadvantage of Optical Interconnection - Channel to channel separation is large - Interface problems between electric & optoelectronic components 8
Motivation Surveying and Mapping
Point-to-Point Communication
http://www.cablefreesolutions.com
http://ngom.usgs.gov/
Lightweight, compact, low power consumption, and large angle agile beam steering 9
OPA with Poly Overlay Input Grating Coupler
Cascaded 1x2 MMIs
Cr/Au Silicon Polysilicon
Wire Bonding Pads
16 Independent TO Phase shifters
Output gratings with polysilicon overlay
Steering in XY plane-TO phase tuning. Steering in XZ plane-wavelength tuning.
Thermo-Optic Phase Shifters
Pπ=20mW Optical Power (AU)
1
11
0.8 0.6 0.4 0.2 0 0
10 20 Electrical Power(mW)
30
Testing Setup Computer/ Labview
IR CCD Laser
16 channel voltage source
DUT
12
Completed Device
http://www.mrc.utexas.edu/people/faculty/ray-chen θ
ψ
2cm
13
Beam Profile Intensity (A. U.)
1
0.8 0.6 0.4 0.2 0
-1
0 1 or (degree)
Beam Widths=1.2° X 0.5° 14
Beam Steering at 1550nm
15
Beam Steering at 1550nm 1
2
FWHM (degree)
Intensity (A. U.)
0.8 0.6 0.4 0.2 0
-10
-5
0 (degree)
5
1.5
1
0.5
0
10
16
direction direction
-10
-5
0 (degree)
5
10
2D Beam Steering
17
Waveguide Crossing Direct crossing: ·1.7dB loss ·-10dB cross-talk
State of art crossing: ·0.2dB loss ·-40dB crosstalk ·6μm x 6μm space 1. 2. 3.
A. Biberman, et al, Rep. Prog. Phys. 75, 046402 (2012) W. Bogaerts, et al, Opt. Lett. 32, 2801 (2007) P. Sanchis, et al, Opt. Lett. 34, 2760 (2009)
Waveguide Crossing based on MMI
1.0
2.0
3.0
4.0
5.0
6.0
z/um
3-mode MMI (Only the two even modes are excited)
0.0
MMI crossing: ·0.2dB loss ·13μm x 13μm space
x/um
2
1
0
-1
-2
-3
1.
H. Chen, IEEE Photon. Tech. Lett. 18, 2260, 2006
Loss Mechanism in 3-mode MMIs Single mode
Multi mode
250nm
air
nc nf
WMMI=1.2 µm Single mode Taper
Multi mode
SiO2
TE fraction versus nc for the mode in the single-mode waveguide
Loss Mechanism in MMI Crossing
Tapered MMI W/O Crossing Multi mode Slab
Transmission (dB)
0 -0.05 -0.1 -0.15 -0.2 1
MMI W/ Crossing MMI W/O Crossing 1.5
2
2.5
nc
TE fraction versus nc for the 2nd mode in the multimode waveguide
Tapered MMI W/ Crossing
3
Coupling Loss at Sharp Transitions (a)
(b)
(c)
Multi mode
Slab
cm: modal field excitation coefficient
Cascaded MMI Crossing Array (b)
(a)
(b)
(a) WMMI WMMI
W
Lt Lin
(c) W
Lt
Lin L
Lin
s
Lt
nc
Ls W
WMMI
Ls
WMMI
nsi
nsi Crossing pitch: WMMI+ Ls
Lin
Lt
W
2
4
6
Simulated MMI Crossing Performance
8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
z/um
nc=1, 20 crossings
-1.5
-2.0
-2.5
0
0.5
0.0
-0.5
-1.0
·Simulated loss: 0.11dB/crossing
1.5
1.0
0.5
0.0
-0.5
-1.0
m 2.0 -2.0 1.5 -2.5 1.0 /u.5 x-1
0.02.04.06.08.010.0 12.0 14.0 16.0 18.0 20.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 36.0 38.0 40.0 42.0 44.0 46.0 48.0 50.0 52.0 54.0 56.0 58.0 60.0
nc=2.5, 20 crossings
z/um
·Crossing pitch: 2.24µm
·Simulated loss: 0.008dB/crossing ·Crossing pitch: 3.08µm
Index-engineering by Subwavelength Nanostructure (a)
nf nc
(b)
250nm
nc nf SiO2 nc=nsub=2.5 Λsub=200 nm Wsub=50 nm (c)
Λsub 200 nm
Wsub
air
Fabricated Waveguide Array Single-mode waveguide array
101 x 101 array Pitch: 3.08µm
SEM Images of the MMI Crossings
a)
Device Characterizations (b)
101 MMI Crossings 0.019dB/crossing@1550nm
Cross-talk
0.14dB/crossing@1550nm 0.14dB/crossing@1550nm
1-to-32 H-tree Optical Distribution Outputs
Input
Two grating couplers for input and output Five 1-to-2 Y-splitters 1.1 cm long single-mode waveguide
H-tree Characterization Top-down IR image
Intensity from each output
1mm Uniformity: 0.72dB Excess loss: 2.2dB
Loss and Bandwidth Measurements
31
SWN based Inter-layer Grating Coupler
Subwavelength Gratings and Coupling Efficiencies
33
Photonic Crystals and Defects Periodic variation of refractive index (Artificial)
Periodic atomic structure (Natural)
Photonic Bandgap
ω
Electronic Bandgap
Photonic Crystals
Semiconductors
Introduction of defects
k Trapped Carriers
Schrődinger Equation
{
Electronic Bandgap
hν }
Photonic Bandgap
k
1000 Electronic Dispersion
Photonic Dispersion
Eigenvalue Problem
Introduction of defects PC resonant cavities and waveguides
Maxwell’s Equations
High speed RF Modulator • Silicon doping to broaden RF bandwidth • Silicon doping
0.5μm 5μm 250nm
RC time delay ↓ Bandwidth ↑ Prons: E-field inside slot ↑ Power consumption ↓ Extinction ratio ↑
; Cons: Optical loss ↑
4μm W
Au
d=300nm Sw=320nm
z n-type doping concentration in silicon : 35
Si
a=425nm y
1×1017 cm−3
x
SiO2 1×1020 cm−3
Applying gate voltage on silicon handle Equivalent circuit
V
Ed Ep
Si SiO2
Benefits
R
R
Ed Ep R
• • R
To increase bandwidth To reduce power consumption
Vgate
Si
Working principle The structure is similar to the well-know MOS structure. When a positive voltage is applied across the oxide, the energy bands in the top silicon are bent, and a high-mobility electron accumulation layer is formed at the Si/SiO2 interface. The conductivity of top silicon is proportional to the mobility and the free-electron density, the limiting frequency increases with increasing gate voltage.
•
L. Alloatti, D. Korn, R. Palmer, D. Hillerkuss, J. Li, A. Barklund, R. Dinu, J. Wieland, M. Fournier, J. Fedeli, H. Yu, W. Bogaerts, P. Dumon, R. Baets, C. Koos, W. Freude, and J. Leuthold , “ 42.7 Gbit/s electro-optic modulator in silicon technology”, Optics Express, Vol. 19, Issue 12, pp. 11841-11851 (2011)
36
Fabrication • Fabricated device
d = 300nm Sw = 320nm a = 425nm 5 μm Air PCs
Slot
PCs EO polymer
Silicon
1 μm
Box 37
200 μm
Characterization • Testing setup
Testing results: small Vπ × L Silicon organic hybrid modulators
39
Testing results: broad RF bandwidth • Measured EO response vs modulation frequency •
Small signal modulation test
Normalized EO response (dB)
Experimental setup VNA
0
Laser
-5 -10 -15
PD -20
MSA -25
0
5
10
15
20
Frequency (GHz)
25
VNA: vector network analyzer PD: photo detector MSA: microwave spectrum analyzer
•
3-dB device bandwidth: ~ 11GHz 40
Testing results: low power consumption • Measured energy consumption per bit, for 100% modulation depth
1 Wbit CV 2 2 4 At 10Gbit/s, Wbit = ¼*39fF*4.08V2*2 = 0.324 pJ/bit At 40Gbit/s, Wbit = ¼*39fF*27.06V2*2 = 14.278 pJ/bit
• C = 39fF from simulation. • Small slot capacitance is due to the large slot width. 41
Testing results: Improved bandwidth with Vg • Measured modulation index under different gate voltage Vg=0V Vg=75V Vg=150V Vg=0V Vg=300V Vg=75V Vg=-75V Vg=150V Vg=300V Vg=-75V
1
SiO2
R
R
Ed Ep R
Vgate
R
Si
0.1 0.1
Inversion Depletion Accumulation 0.01 0.01
0 0
5 5
10 10
15
20
25
30
35
(GHz) 15 Frequency 20 25 30 35
40 40
Frequency (GHz) 3-dB bandwidth (GHz)
V
Ed Ep
Si
15 14
45 45
50 50
Modulation index (Rad)
(Rad) (Rad) indexindex Modulation Modulation
1
0.1
10GHz 20GHz 30GHz 40GHz
13
0.01
12
-300
11 0
50
100
150
200
Gate voltage (V)
250
300
42
-200
-100
0
100
Gate voltage (V)
200
300
Outline 1. Introduction 2. Si Photonics for Optical Communications and Interconnects a. Free Space Optical Phased Array for Beam Steering b. On-chip Optical Interconnects with High Quasi-3D Interconnections c. On-chip 3D Optical Interconnects d. High-speed Si/EO Polymer PCW modulator 3. Si Photonics for Biosensors a. Early cancer detection b. Open sensing platforms 43
Highly Multiplexed Early Cancer Detection using On-Chip Silicon Photonic Devices
Detection Device
Team 1 Highly Sensitive Multiplexed Silicon On-Chip System
Silicon Chip with Selected Biomarkers Team 2 Identify biomarkers to be use with high sensitivity and specificity
Clinical Samples
Team 3 Medical Doctors Provide clinical sera for detection
Motivation Human Health
45
46
Procedure for the Diagnostic Facility
Light Out
• -Diagnostic Facility drops the patient blood / saliva/ serum / any fluid on to the chip • -Again measure the output light spectrum from the waveguide • Identify differences in the transmission spectrum before and after the blood / serum/ saliva/ any fluid was introduced.
Light In
47
Photonic Crystal for Early Cancer Detection Slow Light
Trapped Light Sub-Micron scale Guiding and Confinement of Light 48
Biochip Preparation and Detection Procedure Probe-Target Reaction Probe Immobilization Surface Functionalization
Piranha acid sol, At 60 ⁰C, 1 hour
I. sacrificial oxidation II. 48HF acid for 5 min
APTES
And blocking with 1% BSA
49
Intensity (a.u.)
Principle of Operation
Intensity (a.u.)
Wavelength (nm)
Wavelength (nm)
50
Intensity (a.u.)
Principle of Operation
Intensity (a.u.)
Wavelength (nm)
Wavelength (nm)
51
Intensity (a.u.)
Principle of Operation
Intensity (a.u.)
Wavelength (nm)
Wavelength (nm)
52
Scheme for Virus Detection
53
DNA Detection
54
1. Breast Cancer Biomarker Detection in Serum
Baseline
Probe-EGFR Antibody
Target-EGFR Antigen
EGFR: Epidermal Growth Factor Receptor is the cellsurface receptor
Breast Cancer Biomarker Detection in Cell lysate
2 NRP2b Biomarker Detection for Lung Cancer ---Medical University of South Carolina 1.4
L13 nanoholes L13
L13H L13
1.2
0.475 1.0
Resonance shift (nm)
0.425
0.3
0.321 0.228
0.2
0.1
0.256
0.0650
0.59
0.82
0.80
0.77
0.8
0.65
0.6 0.4 0.2
0.00 0.0
0.0350
0.0
0.00 APTES
L13H L13
1.10 0.8
0.390
0.80 0.68
0.60 0.6
0.57
0.56
0.62
0.51 0.4
0.2
0.0 0.0
BSA
sample
secondary Ab
before
probe
probe
bsa er sample ary Ab second biomark
before
probe
probe
bsa er sample ary Ab second biomark
Test step
Test step
1.0 0.70
0.69
0.69
0.56
0.6 0.5
0.38
0.4
0.40
0.40
0.29
0.3 0.2 0.1
0.00 0.0
Resonance shift (nm)
L13H L13
0.7
0.87
L13H L13
0.8
0.8
0.74
0.80
0.65 0.59
0.6
0.49
0.56
0.54
0.4 0.2 0.0
0.00
-0.1
before
0.73
-0.2 probe
Test steps
Resonance shift (nm)
Resonance shift (nm)
0.4
1.0
1.26
1.24
Resonance shift (nm)
0.5
probe
probe
Ab ple bsa er sam secondary biomark
Test step
before
probe
probe
bsaer sample ndary Ab rk a seco m io b
Test step
3 Therapeutic Drug Detection
Chip-Integrated Microarray for High Throughput Highly Sensitive Highly Specific Cancer Detection
Prototype system demonstrated at Baylor College of Medicine in May 2013 Translation to portable platforms possible 60
Formation of Microfluidic Channels
61
Silicon Nanophotonic Biosensor Chip for Lung Cancer Detection
Figure of merits of our cancer detection chip in reference to all existing results [1-12] [1] J. Waswa, J. Irudayaraj, C. Deb Roy, “Direct detection of E-Coli O157: H7 in selected food systems by a surface plasmon resonance biosensor”, LWT-Food Science and Technology 40 (2), 187 (2007). [2] M.G. Scullion, et al., “Slotted photonic crystal cavities with integrated microfluidics for biosensing applications”, Biosens. Bioelectron. 27, 101-105 (2011). [3] S. Mandal, D. Erickson, “Nanoscale optofluidic sensor arrays”, Opt. Exp.16(3), 1623 (2008). [4] S. Pal, et al., “Silicon photonic crystal nanocavity-coupled waveguides for error-corrected optical biosensing”, Biosens. Bioelectron. 26, 4024 (2011). [5] C.F. Carlborg, et. al, “A packaged optical slot-waveguide ring resonator sensor array for multiplex label-free assays in labs-on-chips”, Lab on a Chip 10, 281 (2010). [6] K. De Vos, et al., “Silicon-on-insulator microring resonator for sensitive and label-free biosensing”, Opt. Exp. 15 (12), 7610 (2007). [7] C.A. Barrios, “Optical slot-waveguide based biochemical sensors”, Sensors 9, 4751 (2009). [8] S. Zlatanovic, et al., “Photonic crystal microcavity sensor for ultracompact monitoring of reaction kinetics and protein concentration”, Sens. and Actuators B 141, 13-19 (2009). [9] H. Li, X. Fan, “Characterization of sensing capability of optofluidic ring resonator biosensors”, Appl. Phys. Lett. 97, 011105 (2010). [10] B.T. Cunningham, et al, “Label-free assays on the BIND system”, J. Biomol. Screen. 9, 481 (2004). [11] M. Iqbal, et al., “Label-Free Biosensor Arrays based on silicon ring resonators and high-speed optical scanning instrumentation”, IEEE J. Sel. Top. Quant. Electron. 16(3), 654 (2010). [12] Y. Zou, S. Chakravarty, W-C. Lai, R.T. Chen, “High yield silicon photonic crystal microcavity biosensors with 100fM detection limit”, Proc. of the SPIE 8570, 857008 (2013) and S. Chakravarty, Y. Zou, W-C. Lai, R.T. Chen, “Slow light engineering for high Q high sensitivity photonic crystal microcavity biosenors in silicon”, Biosensors and Bioelectronics 38(1), 170 (2012).
SINECAD
Center Overview
63
Breast Cancer Biomarker Detection in Serum
Baseline
Probe-EGFR Antibody
Target-EGFR Antigen
EGFR: Epidermal Growth Factor Receptor is the cellsurface receptor
2013 American Cancer Society Annual Meeting
Current Status of biomarker study Today, there is not a single platform approved by the FDA capable of testing a panel of breast cancer biomarkers. The detection methods suffer with Low diagnostic sensitivity and specificity Poor multiplexibility
Figure 3. 16 channel multiplexing microfluidic channel module. Top piece is 16 channel microfluidic chip made of PDMS. Bottom piece is plainarization chip made of PDMS to embed photonic crystal chip to expand the plain surface for sealing top microfluidic channels. Optical fibers are brought in contact with photonic crystal chip to detect wavelength shift. Individual pieces are shown in the right picture.
69
Low Cost of Ownership Chip-Integrated Microarray for High Throughput Highly Sensitive Highly Specific Cancer Detection Omega Optics Inc., Austin, TX Slowing Light for Sensitive Diagnostics
Competitive Advantage: Price of consumables < 50% Biacore 4000 (GE) Bench-Top System Price <50% Biacore 4000 (GE) Biacore acquired for $436M in 2006
Patent Position: 7 issued , 4 pending; more in pipeline, 6 from UT. Team: Omega Optics, University of Texas, Austin; MD Anderson Cancer Center (Breast Cancer); Medical Univ. of S. Carolina (Lung Cancer). Contact: Dr. Ray Chen, CTO,
[email protected], 512-825-4480
70
Fabrication and testing of 1x16 splitter
Xu et al. IEEE Photonics Technology Letters, Vol. 25, No.16, pp.1601-1604 (2013)
64 Highly Multiplexed Early Cancer Detection Chip
72
Open Sensor Systems on Silicon PCW Chip
What has been demonstrated On Semiconductor Nanomembranes
Early Cancer Detection (NIH)* Air-pollution (Methane gas) Sensor(EPA)* Water-pollution (Xylene) Sensor (NSF)* Sensor for National Security (Chemical Warfare Simulant TEP Triethylphosphate, funded by US Army)* EM-Wave Sensor (Missile Defense Agency)*
*: Peer-reviewed Journal publications
available or to be available
Integrated Sample Preparation and Sensors on a Chip with User-Friendly Machine-Human Interface Fast Plasma Separation on Chip
Light Out
Sensor Arrays on Chip
Sample Preparation on Chip
Light In
Silicon Nanophotonic Devices for Early Cancer Detection
Selective CTC Capturing
Thank You
