SLIDE 1
- C. Boit
Nanoscale Debug & Diagnosis | TU Berlin page 1
Optical Backside Attack with Nanoscale Potential C. Boit TUB - - PowerPoint PPT Presentation
Optical Backside Attack with Nanoscale Potential C. Boit TUB - - PowerPoint PPT Presentation
Optical Backside Attack with Nanoscale Potential C. Boit TUB Technische Universitt Berlin From backside, all nodes act alike Optical interaction through frontside: each node has individual signature due to interconnect intransparency Access
SLIDE 2
SLIDE 3
Nanoscale Debug & Diagnosis | TU Berlin page 2
From backside, all nodes act alike
Backside point of view….
Bulk silicon Transistors Metallization Layer Passivation node of interest
SLIDE 4
Nanoscale Debug & Diagnosis | TU Berlin page 3
Optical Backside Circuit Analysis
CCD Modulation Laser Stimulated Electrical Signal Black Body
- r Photon
Emission Laser
- Photon Emission
- Rise and Fall Events of
Digital Signal Pattern
- Modulation of reflected
light by device operation: Contactless Probing
- Laser Stimulated
Current or Voltage Sources: Delay / Fault Injection
Bulk Si
SLIDE 5
Nanoscale Debug & Diagnosis | TU Berlin page 4
Coutesy IBM / Richard Ross
TRE in Ring Oscillator - Demonstrator …impossible with frontside detection
SLIDE 6
Nanoscale Debug & Diagnosis | TU Berlin page 5
Watching the Chip at Work
courtesy Richard Ross / IBM
…unthinkable with frontside detection
SLIDE 7
Nanoscale Debug & Diagnosis | TU Berlin page 6
n+ n+ p+ GND Vp
The reflected light modulation due to: Varying space charge layer – In the pinch off region – In the drain-junction region Varying Charge Density – In the channel region
Laser Beam
Reflectance Modulation Imaging
6 / 55
SLIDE 8
Nanoscale Debug & Diagnosis | TU Berlin page 7
7
Soref et al., IEEE J. of
- Quant. Elec., Vol. QE-
23, No.1, January 1987
α (cm-1)
100 101 102 103 104 105 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8
Photon Energy (eV)
Spectral Absorption in Silicon
Concentration of free electrons N [x 1018 cm-3 ] 0.32 6 24 40
undoped T = 300 K
Sufficient transmission through Si good for die thickness of
100µm 500µm 10µm 1µm
– Near Infra Red (λ≈1µm+) ideal for backside access
SLIDE 9
Nanoscale Debug & Diagnosis | TU Berlin page 8
nsilicon nair
SIL SIL
nSIL
NA R 2 λ =
Microscope Resolution …numerical aperture NA = sin δ2 …numerical aperture NA = n*sin δ2 nair =1 nSi = 3.5 Solid Immersion Lens (SIL) with Si-SIL: Resolution max 140nm Vendors claim: R ≈ 100-120nm
SLIDE 10
Nanoscale Debug & Diagnosis | TU Berlin page 9
ITRS: Tech Node Pitch Year 45nm 160nm 2007 32nm 112nm 2009 22nm 90nm 2011 14nm 70nm 2013
What is the required resolution ?
- NIR + Si SIL resolution
ca 100-120nm
- D&D requires to resolve pitch
- Pitch ca 3.5-8x min. feature size
- NIR good for > 20nm node
technologies
STI S/D G Node Pitch
But: there is some tolerance
SLIDE 11
Nanoscale Debug & Diagnosis | TU Berlin page 10
Discussion of required resolution by Intel @ ISTFA 2015
Table from v. Haartman et al, Proc. 41st ISTFA 2015, Nov. 1-5, Portland OR, pp 47-51
Diffraction-limited resolution δ:
Correlation factor F =
- Resol. / Pitch
F (1500)= 5.81 3,91 4,36 4,32 F(1064)= 4.12 2,77 3.08 3,07 2017!!! NIR is good for…
SLIDE 12
Nanoscale Debug & Diagnosis | TU Berlin page 11
FinFET age – what is required?
- NIR + Si SIL resolution ca 100-120nm
- D & D requires to resolve pitch
- Good for > 20nm node technologies
- For FinFETs, min 2x
improvement necessary
Image: https://www.synopsys.com/Company/Publications/DWTB/PublishingImages/dwtb-finfet-jan2013-fig2.JPG
ITRS 2013: Tech Node Pitch Year 16nm 80nm 2013 10nm 64nm 2015 7nm 50nm 2017 5nm 40nm 2019 3.5nm 32nm 2021 2.5nm 26nm 2023 1.8nm 20nm 2025
SLIDE 13
Nanoscale Debug & Diagnosis | TU Berlin page 12
R = λ/(2NA) NA = n x sinα
λ: Light wavelength (NIR: ≈ 1µm) NA: Numerical aperture n: Index of refraction (Air = 1, Si = 3.5) sinα: Aperture of Objective (<1)
Objective SIL Sample State of the art with Si-SIL: R ≈ 100-120nm
Abbe Criterion
..only one solution: reduction of λ Back to bulk Si – optical access with higher resolution?
SLIDE 14
Nanoscale Debug & Diagnosis | TU Berlin page 13
Optical D&D Techniques using visible light
http://pveducation.org/pvcdrom/materials/optical-properties-of-silicon
1µm — 1mm — 1nm —
What’s a good concept to start with?
- EOFM / LVP seems to be
technique to aim at (low power technologies)
- Lasers ~ 650nm available
- Absorption depth (AD)
( = 1 / abs. coefficient ) key for device prep
- Prep to ca 10µm thickness
within reach
Absorption depth in Silicon (AD)
SLIDE 15
Nanoscale Debug & Diagnosis | TU Berlin page 14
Material for SIL in visible regime - Bandgap
λ = 650nm E = 1.9 eV
http://www.tf.uni-kiel.de/matwis/amat/semitech_en/kap_2/backbone/r2_3_1.html Lattice Constant [Angstrom]
SLIDE 16
Nanoscale Debug & Diagnosis | TU Berlin page 15
Material for SIL in visible regime
Objective SIL Requirement:
- transparent to λ = 650nm
- n as close as possible to
nSi(NIR) = 3.5
- GaP identified for a while
as good candidate
- transparent to λ ≈ 550nm
- nGaP (650nm) ≈ 3.3
SLIDE 17
Nanoscale Debug & Diagnosis | TU Berlin page 16
No SIL SIL
SLIDE 18
Nanoscale Debug & Diagnosis | TU Berlin page 17
VIS LVP
– Average 100k
SLIDE 19
Nanoscale Debug & Diagnosis | TU Berlin page 18
d 380µm L L x LED PD α α
Backside Protection ?
– Connection from electrically active layers to backside far too expensive – Optical interaction ?
SLIDE 20
Nanoscale Debug & Diagnosis | TU Berlin page 19
- Optical interaction for IC Signal Tracking Very
Promising Even in FinFET Age
- Contactless Probing Still Almost Untouched
Opportunity in Attack World
- Backside Protection Necessary – Progress into More
Miniaturization Is No Valid Protection Concept
- But: Optical Interaction Not Only Good for Attacks -