Packaging for Power Electronics
“Habilitation à Diriger des Recherches” Cyril BUTTAY
Laboratoire Ampère, Lyon, France
2015
1 / 45
Packaging for Power Electronics Habilitation Diriger des Recherches - - PowerPoint PPT Presentation
Packaging for Power Electronics Habilitation Diriger des Recherches Cyril B UTTAY Laboratoire Ampre, Lyon, France 2015 1 / 45 Outline Professional Record Background Contributions Perspectives Conclusion 1 / 45 Outline
1 / 45
1 / 45
1 / 45
2 / 45
Lab/Projects 59% Tutorials 6% Lectures 4% Administrative 31%
L1 27% L2 16% L3 24% M1 11% M2 21%
3 / 45
◮ 9 IEEE ◮ 6 Elsevier
4 / 45
2 9 2 1 2 1 1 2 1 2 2 1 3 2 1 4 2 1 5 Master Master Master Master License PhD PhD PhD PhD PhD
5 / 45
2009 2010 2011 2012 2013 2014 2015
EPAHT ETHAER BQR CIFRE Industry Project ECLIPSE ARC THOR Industry Project BQR
ACCITE SuMeCe
Supergrid Genome
◮ European: Euripides-Catrene (THOR) ◮ National: Agency for Research (ETHAER, ECLIPSE), Aerospace
◮ Local fundings: BQR, Carnot institute (SuMeCe) ◮ Direct funding by the industry (5 companies) 6 / 45
◮ Member of the laboratory board ◮ Installation and management of shared equipment: ◮ Packaging lab (≈ 300kC) ◮ Computer cluster (2009–2014)
◮ Reviewer for journals/conferences (20-30 publications/year) ◮ Reviewer for projects proposals (Cleansky, 7 days) ◮ Member of 3 selection panels (hiring of lecturers) ◮ Member of PhD jurys (10) ◮ Member of an evaluation committee (LN2, Sherbrooke) ◮ Management of the “3DPHI” platform on power integration
7 / 45
7 / 45
◮ Thermal management ◮ Electrical insulation ◮ Interconnects ◮ Mechanical/chemical protection
◮ Ceramics ◮ Metals ◮ Organics. . . 8 / 45
◮ Thermal management ◮ Electrical insulation ◮ Interconnects ◮ Mechanical/chemical protection
◮ Ceramics ◮ Metals ◮ Organics. . . 8 / 45
◮ Thermal management ◮ Electrical insulation ◮ Interconnects ◮ Mechanical/chemical protection
◮ Ceramics ◮ Metals ◮ Organics. . . 8 / 45
◮ Thermal management ◮ Electrical insulation ◮ Interconnects ◮ Mechanical/chemical protection
◮ Ceramics ◮ Metals ◮ Organics. . . 8 / 45
◮ Thermal management ◮ Electrical insulation ◮ Interconnects ◮ Mechanical/chemical protection
◮ Ceramics ◮ Metals ◮ Organics. . . 8 / 45
◮ Thermal management ◮ Electrical insulation ◮ Interconnects ◮ Mechanical/chemical protection
◮ Ceramics ◮ Metals ◮ Organics. . . 8 / 45
0°C 500°C 1000°C 1500°C 2000°C 2500°C 3000°C 10 V 100 V 1 kV 10 kV 100 kV 1 MV Junction temperature Breakdown voltage Silicon 3C−SiC 6H−SiC 4H−SiC 2H−GaN Diamond
Source: C. Raynaud et al. “Comparison of high voltage and high temperature performances of wide bandgap semiconductors for vertical power devices” Diamond and Related Materials, 2010, 19, 1-6
9 / 45
RGl Tl VDRl RGh Th VDRh VIn IOut
10 / 45
RGl Tl VDRl RGh Th VDRh VIn IOut CGDl CGSl CDSl CGDh CGSh CDSh LDC1 LDC3 CDC LCdc LDC2 LDC4 LDl LGl LGh LSl LSh COut CCM1 CCM2
10 / 45
0 J 500 uJ 1 mJ 0 H 20 nH 40 nH 60 nH 80 nH 100 nH 120 nH Switching energy (J) Drain inductance (H) Simulations at I0 = 100 A Energy stored in the drain inductance
11 / 45
11 / 45
11 / 45
12 / 45
12 / 45
12 / 45
12 / 45
2 4 6 8 10 12
Forward voltage [V]
2 4 6 8 10 12
Forward current [A]
30 ◦ C 70 ◦ C 110 ◦ C 150 ◦ C 190 ◦ C 230 ◦ C 270 ◦ C 300 ◦ C
13 / 45
2 4 6 8 10 12
Forward voltage [V]
2 4 6 8 10 12
Forward current [A]
30 ◦ C 70 ◦ C 110 ◦ C 150 ◦ C 190 ◦ C 230 ◦ C 270 ◦ C 300 ◦ C
50 50 100 150 200 250 300
Junction temperature [C]
20 40 60 80 100 120 140
Dissipated power [W]
2.0 A 4.0 A 6.0 A 8.0 A 10.0 A
13 / 45
14 / 45
14 / 45
14 / 45
14 / 45
14 / 45
Always stable 14 / 45
Always stable Always unstable 14 / 45
Always stable Always unstable Becomming unstable with ambient temperature rise 14 / 45
15 / 45
15 / 45
Buttay et al. “Thermal Stability of Silicon Carbide Power JFETs”, IEEE Trans on Electron Devices, 2014
current changed from 3.65 to 3.7 A Run-away
16 / 45
Strengh/Hardness Homologous Temperature
0.4 0.6 1
Properties little affected by temperature Creep range Unable to bear engineering loads Solders operate in this region
Source:
❤tt♣✿✴✴✇✇✇✳❛♠✐✳❛❝✳✉❦✴❝♦✉rs❡s✴t♦♣✐❝s✴✵✶✻✹❴❤♦♠t✴
◮ Sintering (solid state, process below melting point) ◮ Diffusion soldering/TLPB (creation of a high melting point alloy) 17 / 45
Strengh/Hardness Homologous Temperature
0.4 0.6 1
Properties little affected by temperature Creep range Unable to bear engineering loads Solders operate in this region
Source:
❤tt♣✿✴✴✇✇✇✳❛♠✐✳❛❝✳✉❦✴❝♦✉rs❡s✴t♦♣✐❝s✴✵✶✻✹❴❤♦♠t✴
◮ Sintering (solid state, process below melting point) ◮ Diffusion soldering/TLPB (creation of a high melting point alloy) 17 / 45
Strengh/Hardness Homologous Temperature
0.4 0.6 1
Properties little affected by temperature Creep range Unable to bear engineering loads Solders operate in this region
Source:
❤tt♣✿✴✴✇✇✇✳❛♠✐✳❛❝✳✉❦✴❝♦✉rs❡s✴t♦♣✐❝s✴✵✶✻✹❴❤♦♠t✴
◮ Sintering (solid state, process below melting point) ◮ Diffusion soldering/TLPB (creation of a high melting point alloy) 17 / 45
Strengh/Hardness Homologous Temperature
0.4 0.6 1
Properties little affected by temperature Creep range Unable to bear engineering loads Solders operate in this region
Source:
❤tt♣✿✴✴✇✇✇✳❛♠✐✳❛❝✳✉❦✴❝♦✉rs❡s✴t♦♣✐❝s✴✵✶✻✹❴❤♦♠t✴
◮ Sintering (solid state, process below melting point) ◮ Diffusion soldering/TLPB (creation of a high melting point alloy) 17 / 45
Strengh/Hardness Homologous Temperature
0.4 0.6 1
Properties little affected by temperature Creep range Unable to bear engineering loads Solders operate in this region
Source:
❤tt♣✿✴✴✇✇✇✳❛♠✐✳❛❝✳✉❦✴❝♦✉rs❡s✴t♦♣✐❝s✴✵✶✻✹❴❤♦♠t✴
◮ Sintering (solid state, process below melting point) ◮ Diffusion soldering/TLPB (creation of a high melting point alloy) 17 / 45
◮ Sintering pressure ◮ Surface roughness ◮ Thickness of stencil ◮ Substrate finish. . .
18 / 45
◮ Sintering pressure ◮ Surface roughness ◮ Thickness of stencil ◮ Substrate finish. . .
10 20 30 40 50 60 70 A’ B C D E F J N O T Shear strenght [MPa] Series name 18 / 45
◮ Sintering pressure ◮ Surface roughness ◮ Thickness of stencil ◮ Substrate finish. . .
Ag SiC Cu
10 20 30 40 50 60 70 A’ B C D E F J N O T Shear strenght [MPa] Series name 18 / 45
◮ Oxygen is necessary ◮ Bonding on copper (oxide) ◮ Standard Ni/Au finish not ideal ◮ Confirmed by several teams ◮ weak bonds at Ag/Au interface ◮ Bond strength lower ◮ Porosity higher ◮ Can be used to attach fragile components 19 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
49.0 48.8 48.6 48.4 time [µs] 50 50 100 150 200 250 Vout [V] 0.2 0.0 0.2 time [µs]
49.0 48.8 48.6 48.4 time [µs] 1 1 2 3 4 5 Iout [A] 0.2 0.0 0.2 time [µs]
20 / 45
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 1E-3 1E-2
1/t (h
Electric Field (V/mm)
Without parylene Parylene SCS HT
T = 300°C Stop parameter
21 / 45
500 1000 1500 2000 2500 3000 3500 4000 4500 5000 1E-3 1E-2
1/t (h
Electric Field (V/mm)
Without parylene Parylene SCS HT
T = 300°C Stop parameter
21 / 45
◮ pressureless processes are a good model 22 / 45
22 / 45
23 / 45
23 / 45
23 / 45
24 / 45
24 / 45
24 / 45
24 / 45
25 / 45
25 / 45
25 / 45
25 / 45
25 / 45
25 / 45
25 / 45
25 / 45
25 / 45
25 / 45
25 / 45
26 / 45
10.1 10.2 10.3 time [µs] 100 200 300 400 500 600 700 Vout [V] 11.7 11.8 11.9 time [µs]
27 / 45
Vbus OUT GND JH JL
28 / 45
0.2 mm
0,3 mm
0.16 mm
0,15 mm
0.15 mm
0.3 mm
29 / 45
0.9 1.0 1.1 1.2 time [µs] 50 50 100 150 200 Vout [V] 49.9 50.0 50.1 50.2 time [µs]
200°C
30 / 45
31 / 45
◮ Copper posts growth on die (electroplating) ◮ Original die/DBC assembly technology: SnCu diffusion bonding
32 / 45
◮ All configuration but one yield to bonding ◮ Tensile strenght 106 to 261 MPa (365 MPa for bulk copper)
◮ Some investigations performed, much more needed 33 / 45
◮ All configuration but one yield to bonding ◮ Tensile strenght 106 to 261 MPa (365 MPa for bulk copper)
◮ Some investigations performed, much more needed 33 / 45
◮ All configuration but one yield to bonding ◮ Tensile strenght 106 to 261 MPa (365 MPa for bulk copper)
◮ Some investigations performed, much more needed 33 / 45
◮ All configuration but one yield to bonding ◮ Tensile strenght 106 to 261 MPa (365 MPa for bulk copper)
◮ Some investigations performed, much more needed 33 / 45
5 10 15 20 25 5 10 15 20 25 30 60 90 120 150 180 210 240
34 / 45
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Forward Voltage [V] 50 100 150 200 Forward Current [A]
To 247 Sandwich package
◮ Due to seed layer/die topside interface ◮ Would not happen with suitable dies
◮ Tens of sample assembled, with good yield
200 300 400 500 600 700 800 Sputter time [s] 2000 4000 6000 8000 10000 12000 14000 16000 Intensity
02 Ti Al2 AlO2 TiO CuO Cu2
35 / 45
◮ Solder ◮ Silver sintering ◮ Direct Cu/Cu bonding (Micro-posts)
◮ Solid/liquid interface ◮ Homogeneous compressing force ◮ No issue with flatness
◮ Dies topside finish ◮ Mechanical relief structures ◮ Intrinsic thermo-mechanical reliability ◮ Need for further investigation 36 / 45
◮ Solder ◮ Silver sintering ◮ Direct Cu/Cu bonding (Micro-posts)
◮ Solid/liquid interface ◮ Homogeneous compressing force ◮ No issue with flatness
◮ Dies topside finish ◮ Mechanical relief structures ◮ Intrinsic thermo-mechanical reliability ◮ Need for further investigation 36 / 45
◮ Solder ◮ Silver sintering ◮ Direct Cu/Cu bonding (Micro-posts)
◮ Solid/liquid interface ◮ Homogeneous compressing force ◮ No issue with flatness
◮ Dies topside finish ◮ Mechanical relief structures ◮ Intrinsic thermo-mechanical reliability ◮ Need for further investigation 36 / 45
36 / 45
36 / 45
◮ Thermo-mechanical stress in assemblies (esp. “Sandwiches”) ◮ Thermal resistance ◮ Parasitic inductance/capacitance
◮ Die stacking (Chip-On-Chip, G2ELab) for low EMI ◮ PCB Embedding 37 / 45
◮ Thermo-mechanical stress in assemblies (esp. “Sandwiches”) ◮ Thermal resistance ◮ Parasitic inductance/capacitance
◮ Die stacking (Chip-On-Chip, G2ELab) for low EMI ◮ PCB Embedding
J.-L. Marchesini et al., “Realization and Characterization of an IGBT Module Based on the Power Chip-on-Chip 3D Concept”, ECCE 2014
boards, 2012
37 / 45
38 / 45
38 / 45
38 / 45
38 / 45
38 / 45
38 / 45
38 / 45
38 / 45
38 / 45
38 / 45
Source: Pressureless Sintering of Microscale Silver Paste for 300° C Applications, Fang Yu et al., IEEE trans. on CMPT, vol. 5, No.9, p 1258–1264
◮ Assessment of migration phenomenon ◮ Next step: effect of atmosphere (oxygen content) ◮ Use of pressureless sintered silver as a material model ◮ Behaviour in high temperature storage conditions ◮ Mechanisms are accelerated and highlighted 39 / 45
◮ High temperature ageing of PCBs ◮ Thermo-mechanical analysis of structures ◮ Manufacturing of integrated inductors
40 / 45
40 / 45
Source: 10 kV SiC MOSFET from Wolfspeed
◮ Ensure low inductance ◮ Large creepage distance (> 8 cm !)
41 / 45
Source Gunturi, S. et al. “Innovative Metal System for IGBT Press Pack Modules”, ISPSD, 2003, 4
◮ Melting of silicon ◮ Alloying with surrounding metals ◮ Formation of conductive area
42 / 45
Source Gunturi, S. et al. “Innovative Metal System for IGBT Press Pack Modules”, ISPSD, 2003, 4
◮ Melting of silicon ◮ Alloying with surrounding metals ◮ Formation of conductive area
42 / 45
Source Gunturi, S. et al. “Innovative Metal System for IGBT Press Pack Modules”, ISPSD, 2003, 4
◮ Melting of silicon ◮ Alloying with surrounding metals ◮ Formation of conductive area
42 / 45
42 / 45
◮ From electrical engineering to packaging ◮ 11 PhD students supervised (5 theses defended)
◮ High temperature packaging ◮ High voltage packaging ◮ Integration for WBG devices
◮ Strong support from the industry ◮ Many scientific challenges 43 / 45
44 / 45
45 / 45