Gate current degradation in W-band InAlN/AlN/GaN HEMTs under Gate Stress

Transcription

1 Gate current degradation in W-band InAlN/AlN/GaN HEMTs under Gate Stress Yufei Wu and Jesús A. del Alamo Microsystems Technology Laboratories (MTL) Massachusetts Institute of Technology (MIT) Sponsor: National Reconnaissance Office April 04,

2 Purpose To understand degradation of gate leakage current in ultra-scaled InAlN/AlN/GaN HEMTs under positive gate stress 2

3 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 3

4 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 4

5 Benefits of GaN for RF Promising applications: Benefits of GaN for RF: Wide bandwidth High power density Excellent energy efficiency Small volume 5

6 InAlN as barrier material Al 0.2 Ga 0.8 N/GaN ln 0.17 Al 0.83 N/GaN Δ P 0 (cm -2 ) 6.5 x x P piezo (cm -2 ) 5.3 x P total (cm -2 ) 1.2 x x [J. Kuzmik, EDL 2001] High spontaneous polarization in InAlN high 2DEG density InAlN thickness scaling gate length scaling W- and V-band applications 6

7 InAlN as barrier material In 0.17 Al 0.83 N [M. A. Laurent, JAP 2014] In 0.17 Al 0.83 N lattice matched to GaN Potentially better reliability! 7

8 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 8

9 Devices (E-mode) InAlN/AlN/GaN HEMTs: E-mode L g = 40 nm L gs =L gd =1 µm W-band BFTEM of virgin device Gate metal 5 nm 1 nm passivation InAlN GaN AlN [Saunier, CSICS 2014] Thermal models available 9

10 FOMs for benign characterization, detrapping methodology FOMs: I Dmax : I D at V GS = 2 V and V DS = 4 V V Tsat : V GS extrapolated from I D at peak g m point at V DS = 4 V I Goff : I G at V GS = -2 V, V DS = 0.1 V I Doff : I D at V GS = -2 V, V DS = 0.1 V R D : at I G = 20 ma/mm R S : at I G = 20 ma/mm Detrapping & initialization: 100 C bake for 1 hour I D (ma/mm) V DS = 4 V virgin device V GS (V) 10

11 Impact of characterization and detrapping Impact of 200 successive characterizations After initialization After 200 characterizations ΔI Dmax /I Dmax (0) [%] ΔV Tlin [mv] R D /R D (0) R S /R S (0) After detrapping Nearly complete recovery after thermal detrapping step characterization suite is benign and detrapping step is effective 11

12 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 12

13 RT Positive-V G step-stress-recovery experiment Stress conditions: V GS,stress = V in 0.1 V steps, V DS = 0 V, RT Recovery: V DS = V GS = 0 V Stress time = recovery time = 150 s Characterization every 15 s I Gstress (ma/mm) V GS,stress (V) time (s) I Gstress (ma/mm) V GS,stress (V) time (s) 2.3 V I Gstress at constant V GS,stress for V GS,stress 2.3 V 13

14 Time evolution of I Goff and R D V GS,stress (V) 2.3 V V GS,stress (V) I Goff (ma/mm) stress recovery 1.7 V final detrapped R D (ohm.mm) stress recovery 2.3 V final detrapped time (s) time (s) Two mechanisms: From V GS,stress = 1.7 V: I Goff, trapping mechanism 1: new defects generated in AlN From V GS,stress = 2.3 V: o I Goff Mechanisms 2? o R D and R S 14

15 Before and after stress: permanent degradation I Dmax V Tsat I Doff I Doff I Dmax ΔV Tsat > 0 Mechanisms 2: consistent with gate sinking 15

16 High T Positive-V G step-stress experiment Stress conditions: V GS,stress = V in 0.1 V steps, V DS = 0 V, T stress = 150 C Stress time = 60 s Characterization every 15 s I Gstress (ma/mm) V GS,stress (V) time (s) I Gstress (ma/mm) 10 3 V GS,stress (V) V time (s) I Gstress at constant V GS,stress for V GS,stress 2.0 V 16

17 High T Positive-V G step-stress experiment V GS,stress (V) V GS,stress (V) V 12 I Goff (ma/mm) V final detrapped R D (ohm.mm) final detrapped 2.0 V time (s) time (s) From V GS,stress = 1.4 V: I Goff From V GS,stress = 2.0 V: I Goff, R D, and R S Lower threshold for degradation than at RT Both mechanisms thermally enhanced 17

18 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 18

19 RT Constant-V G stress experiment Stress conditions: V GS,stress = 2 V, V DS = 0 V, RT Characterization every 15 s I Goff becomes noisy at t stress ~ 500 s; increases afterwards consistent with trap generation in AlN layer (mechanism 1) R D changes little throughout experiment I Gstress keeps decreasing no Schottky barrier degradation 19

20 Before and after stress: permanent degradation I Doff I Doff No significant I Dmax degradation No significant subthreshold characteristics degradation 20

21 Summary so far Two degradation mechanisms identified: 1. Under mild gate stress: o Observation: I Goff, trapping, thermally enhanced o Proposed mechanism 1: high electric field induced defect generation in AlN interlayer 2. Under harsh gate stress: o Observation: I Goff, R D and R S, ΔV T > 0, I Dmax, thermally enhance o Proposed mechanism 2: self-heating induced Schottky gate degradation, or gate-sinking 21

22 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 22

23 Thermal stress experiment Impact of thermal stress: 1 min. RTA in N 2 I D (ma/mm) I Doff before stress after 400 C after 500 C V Tsat I Dmax V GS (V) V DS = 4 V Permanent degradation: Prominent I Dmax with T Positive V Tsat shift I Doff Same signature as that of degradation mechanism 2 consistent with gate sinking 23

24 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 24

25 Virgin device: Thermionic Field Emission fitting I G (ma/mm) T: -50 C to 200 C experiment TFE fitting V GS (V) ln[i s ktcosh(e 00 /kt)] φ b = 0.95 ev /E 0 (ev -1 ) T dependence well explained by Thermionic Field Emission (TFE) theory Extracted effective Schottky barrier height (φ b ): 0.95 ev 25

26 After harsh gate stress: Poole-Frenkel Emission fitting Stress conditions: V GS,stress = V in 0.1 V steps, V DS = 0 V ln( I G /V G ) T: -50 C to 200 C experiment P-F fitting y-int. [ln(i G /V G ) vs. V G 0.5 ] φ t = 0.11 ev V 0.5 G (V) 0.5 1/kT (ev) -1 T dependence well explained by Poole-Frenkel Emission (P-F) theory Extracted effective trap energy level (φ t ): 0.11 ev 26

27 After mild gate stress: Poole-Frenkel Emission fitting Stress conditions: V GS,stress = 2 V, V DS = 0 V ln( I G /V G ) T: -50 C to 200 C experiment P-F fitting V G 0.5 (V) 0.5 y-int. [ln(i G /V G ) vs. V G 0.5 ] φ t = 0.36 ev /kT (ev) -1 T dependence well explained by Poole-Frenkel Emission (P-F) theory Extracted effective trap energy level (φ t ): 0.36 ev Close to donor level of N vacancy in AlN of 0.5 ev [T. L. Tansley, PRB 1992] 27

28 Outline 1. Motivation 2. Devices and experimental approach 3. Gate stress experiments Harsh step-stress (-recovery) experiments Mild constant gate stress experiment 4. Thermal stress 5. Gate current: dominant charge transport mechanisms 6. Conclusions 28

29 Conclusions Identified two degradation mechanisms: 1. Under mild gate stress: o Observation: I Goff, trapping, thermally enhanced o Proposed mechanism 1: high electric field induced defect generation in AlN interlayer 2. Under harsh gate stress: o Observation: I Goff, R D and R S, ΔV T > 0, I Dmax, thermally enhanced o Proposed mechanism 2: self-heating induced Schottky gate degradation, or gate-sinking Transport model for I G in low forward regime: Virgin device: TFE with φ b = 0.95 ev After degradation mechanism 1: P-F with φ t = 0.36 ev After degradation mechanism 2: P-F with φ t = 0.11 ev 29