Field-induced strain degradation of AlGaN/GaN ... - Semantic Scholar

APPLIED PHYSICS LETTERS 97, 223502 共2010兲

Field-induced strain degradation of AlGaN/GaN high electron mobility transistors on a nanometer scale Chung-Han Lin,1,a兲 D. R. Doutt,2 U. K. Mishra,3 T. A. Merz,2 and L. J. Brillson1,2,4 1

Department of Electrical and Computer Engineering, The Ohio State University, Columbus, Ohio 43210, USA 2 Department of Physics, The Ohio State University, Columbus, Ohio 43210, USA 3 Department of Electrical and Computer Engineering and Materials, University of California, Santa Barbara, California 93106, USA 4 Center for Materials Research, The Ohio State University, Columbus, Ohio 43210, USA

共Received 7 October 2010; accepted 5 November 2010; published online 29 November 2010兲 Nanoscale Kelvin probe force microscopy and depth-resolved cathodoluminescence spectroscopy reveal an electronic defect evolution inside operating AlGaN/GaN high electron mobility transistors with degradation under electric-field-induced stress. Off-state electrical stress results in micron-scale areas within the extrinsic drain expanding and decreasing in electric potential, midgap defects increasing by orders-of-magnitude at the AlGaN layer, and local Fermi levels lowering as gate-drain voltages increase above a characteristic stress threshold. The pronounced onset of defect formation, Fermi level movement, and transistor degradation at the threshold gate-drain voltage of J. A. del Alamo and J. Joh 关Microelectron. Reliab. 49, 1200 共2009兲兴 is consistent with crystal deformation and supports the inverse piezoelectric model of high electron mobility transistor degradation. © 2010 American Institute of Physics. 关doi:10.1063/1.3521392兴 AlGaN/GaN high electron mobility transistors 共HEMTs兲 are leading devices for high frequency, high power electronics due to their high current density, breakdown voltage, and temperature capabilities.1,2 High current and voltage are challenges to device reliability,3–11 yet physical degradation mechanisms of AlGaN/GaN HEMTs are still being explored. Early work showed that they experience large temperature10 and mechanical strain effects4,6–8,11 during the device operation. These effects combine with piezoelectric strain and occur in highly localized regions of the transistor so that device failure is hard to predict. Here we use a depth-resolved catholuminescence spectroscopy 共DRCLS兲共Ref. 12兲 and Kelvin probe force microscopy 共KPFM兲 to measure the degradation of AlGaN/GaN HEMTs during an off-state stress. These nanoscale techniques measure both where and how electric-field-induced stress degrades the state-of-the-art AlGaN/GaN HEMTs, the electrically active defects produced at degradation sites, and the threshold stress above which these defects begin to rapidly grow. Significantly, the local nature of our probes allows us to predict where the device failure will occur. Unintentionally doped 共UID兲 GaN layers grown by metal-organic chemical vapor deposition on sapphire substrates followed by a 0.7 nm AlN interfacial layer, 40 nm Al0.22Ga0.78N layer, 10 nm silicon-doped 共7 ⫻ 1018 cm−3兲 graded 共0 ⱕ x ⱕ 0.22兲 AlxGa1−xN, and a 250 nm UID GaN cap were used. The UID GaN layer is etched in order to form drain, source, and gate electrode. For convenience, regions between gate and drain and between gate and source are defined as “extrinsic drain” and “extrinsic source.” For the same transistor measured before stress, we apply an off-state stress 共10⬍ VDS ⬍ 30 V兲 in steps 共with gate bias VG = −6 V to suppress ID兲 with an Agilent 共Santa Clara. CA兲 4145B analyzer. A key failure indicator is a gate leakage a兲

Electronic mail: [email protected].

0003-6951/2010/97共22兲/223502/3/$30.00

current IG-off, which is taken at VDS = 0.5 V, VGS = −6 V. To obtain nanometer-scale optical spectra at 12K, we used a JEOL 共Tokyo, Japan兲 JAMP-7800F ultrahigh vacuum scanning electron microscope 共SEM兲. From Monte Carlo simulations,13 we used the beam voltage EB = 5 keV to probe near the two dimensional electron gas 共2DEG兲 heterointerface. Experimental details are available elsewhere.12 A Park Systems 共Suwon, Republic of Korea兲 XE-70 AFM/KPFM provided simultaneous topography and potential maps.14 Figure 1共a兲 shows the dc-IV and IG-off before and after an off-state stress. IG-off increases by 2.6⫻ with VDG, sharply rising above a 28 V critical voltage after an off-state stress. DRCLS measures the electric-field-induced stress in an off state 共no heating兲 from the near band edge 共NBE兲 peak, which shifts to higher energy by 26 meV/GPa.15 Figure 1共b兲 shows NBE energy increases up to 7 meV corresponding to a 0.27 GPa compressive stress at the edge of the gate on the drain side within the 2DEG channel after an off-state stress. For a 28 V critical VDG, the corresponding stress is ⬃0.25 GPa compared with this transistor before stress. With an increasing off-state stress, KPFM reveals micron-scale patches of lower potential and Fermi level 共EF兲. Figures 2共a兲–2共c兲 display the KPFM, atomic force microscope 共AFM兲 and SEM images, respectively, of gate 共G兲, drain 共D兲, and source 共S兲 in a full gate’s “upper” and “lower” scanned parts. The KPFM images show a striking evolution of surface potential with an increasing stress voltage VDS as low surface potential patches begin to grow. Initially, the lowest potential area 共upper half兲 extends along and on both sides of the gate. Between VDG = 24 and 32 V, the lowest potential area shifts to the gate foot on the drain side. As VDG increases further, this lowest potential area expands and extends from the gate foot across the extrinsic drain region. The surface potential at this extended area decreases from ⬃0.4 to ⫺1.6 V. AFM shows no surface morphology changes during these potential changes from the rectangular

97, 223502-1

© 2010 American Institute of Physics

Author complimentary copy. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

223502-2

Lin et al.

Appl. Phys. Lett. 97, 223502 共2010兲

FIG. 3. 共Color online兲 DRCLS spectrum at one of lowest potential areas 共inset兲 within extrinsic drain region after off-state stress showing 2.2 eV YB, 2.8–3.0 BB, and 3.45 eV NBE peaks.

FIG. 1. 共Color online兲共a兲 dc output characteristics and IG-off as a function of off-state stress VDS = 10– 30 V and VGS = −6 V. Inset shows the dc-IV characteristics before and after an off-state stress. 共b兲 External stress caused by applied voltage under an off-state stress.

gate foot to the drain, unlike pits previously measured from HEMTs with a T-shaped gate etched away.7 Off-state VDS ⬎ 32 V results in a broken gate, which is a crater forming at the extrinsic drain region and device failure 关Fig. 2共c兲兴. Comparing the potential and morphology results with electrical stress, one can link a device failure with a lower surface potential area. As stress voltage increases, the device properties gradually degrade toward failure, faster for VDG above a ⬃28 V critical voltage. Although no AFM or SEM

FIG. 2. 共Color online兲共a兲 KPFM results that show the evolution of surface potential under an off-state stress at upper and lower regions of AlGaN/GaN HEMTs. AFM images show upper and lower scanning areas at VDG = 36 V. The SEM image in 共b兲 shows the corresponding AFM/KPFM scanning area. The red dashed circles show regions where potentials change faster. 共c兲 The SEM image indicates where upper scan area 共rotated 45° clockwise兲 failure occurs with an increasing off-state stress.

morphology changes are evident, the surface potential dramatically changes with stress. Furthermore, failure occurs close to the region of lowest potential. Failure is defined here as a crater in the extrinsic drain region. Likewise, the lower the surface potential, the more probable failure occurs there. Therefore, surface potential appears to be a strong indicator for device failure: 共i兲 low potential 共e.g., decreasing by ⬎1.5 eV兲 patches likely indicate failure points, 共ii兲 a patch with the lowest potential is likely the initial point of failure, and 共iii兲 the lower the potential, the closer to the device failure. Figure 3 shows the DRCLS result at one of the lowest surface potential areas 关point 9 in Fig. 4共a兲兴 in the extrinsic drain region after an off-state stress. Besides the 3.45 eV near band edge emission, a 2.2 eV yellow band 共YB兲 and 2.8–3.0 eV blue band 共BB兲 are evident. YB emission is often associated with Ga vacancies. BB emission can be associated with surface16 or bulk defects. Both defects trap charge and move the Fermi level EF.

FIG. 4. 共Color online兲共a兲 off-state KPFM maps showing numbered upper and lower low potential regions and 共b兲 corresponding surface potential and average YB/NBE intensity ratio increases with an off-state stress. From the color potential scale 共red higher, blue lower兲, YB/NBE increases most at lowest potential regions 6, 8, 9, and 10. Higher potential patches display slower changes. 1 and 2 correspond to extrinsic drain and drain-side gate foot of an unstressed reference device. 共c兲 Surface potential variation vs VDG and corresponding self-consistent electrostatic defect density 共smooth lines兲.


223502-3

Appl. Phys. Lett. 97, 223502 共2010兲

Lin et al.

KPFM maps for off-state stress show that decreasing surface potential correlates with increased YB and BB intensities. Figures 4共a兲 and 4共b兲 show numbered KPFM patches of an off-state stressed transistor and corresponding changes in DRCLS defect intensity for each patch, respectively. Each data point in Fig. 4共b兲 represents an average of multiple spectra. With an increasing off-state stress and decreasing potential, regions 6, 8, 9, and 10 exhibit monotonically increasing defect intensity. Thus area 9 at the extrinsic drain region has the second lowest potential and the second highest YB increase. Area 10 at the crater edge 关failure edge area 共FEA兲兴 has the highest CL YB emission but a higher KPFM potential measured prior to crater formation. Areas with higher potentials and unstressed devices exhibit negligible defect increases, e.g., areas 1–5. The BB emission exhibits similar increases in off state in regions 9 and 10. Figure 4共c兲 shows the variation in surface potential with VDG. For low potential regions such as areas 9 and 10, there is a striking change in the rate of surface potential decrease above off-state stress VDG ⬃ 20– 25 V. The corresponding increase in gap state defect density with lower EF can be calculated using a self-consistent electrostatic model that depends on acceptor level EA and EF positions in the band gap and the EF-dependent density of charged acceptors ␴A−.17 Negatively charged acceptors induce a dipole q⌬V = q2␴A−d / ␧ that shifts EF lower in the band gap, depending on the occupancy of charged acceptor sites ␴A− = ␴A0 / 关1 + exp关共EF − EA兲 / kBT兴兴, charge separation d, dielectric permittivity ␧, and total acceptor density ␴A0. As defect density increases, EF moves closer to the gap state. Using deep level spectroscopy measured trap densities of high 1012 cm−2 for similar stressed HEMTs,18 we fit the potential variation to defect densities increasing by nearly two orders of magnitude with a charge separation of approximately 8 nm, corresponding to defects located at or below the AlGaN surface. As with a gate leakage current, there are pronounced decreases in potential and increases in defect density above VDG ⬃ 25 V.3 The correlation of stress 关Fig. 1共b兲兴 and defect generation 关Fig. 4共c兲兴 thresholds supports a model of defects forming between gate and drain by an off-state field-induced stress normal to the surface that, added to lattice-mismatch strain already present, exceeds the crystal’s critical elastic energy density.11 In summary, nanoscale scanning and depth-resolved measurements of electric-field-induced stress induced by offstate operation above a characteristic VDG threshold reveal

expanding nanoscale low potential patches inside the extrinsic drain where electrically active defects grow and where, for the lowest potential, highest defect density, the device will fail. The pronounced onset of defect formation, Fermi level movement, and transistor degradation at the threshold gate-drain voltage of Joh and del Alamo11 provides a direct optical and electrostatic evidence for gate-foot trap state generation and supports the inverse piezoelectric model of AlGaN/GaN HEMT degradation. This work was supported by the Office of Naval Research Grant No. N00014-08-1-0655 共Dr. Paul Maki and Harry Dietrich兲. 1

Y. F. Wu, D. Kapolnek, J. P. Ibbetson, P. Parikh, B. P. Keller, and U. K. Mishra, IEEE Trans. Electron Devices 48, 586 共2001兲. 2 O. Ambacher, B. Foutz, J. Smart, J. R. Shearly, N. G. Weimann, K. Chu, M. Murphy, A. J. Sierakowski, W. J. Schaff, L. F. Eastman, R. Dimitrov, A. Mitchell, and M. Stutzmann, J. Appl. Phys. 87, 334 共2000兲. 3 J. Joh and J. A. del Alamo, IEEE Electron Device Lett. 29, 287 共2008兲. 4 G. Meneghesso, G. Verzellesi, F. Danesin, F. Rampazzo, F. Zanon, A. Tazzoli, M. Meneghini, and E. Zanoni, IEEE Trans. Device Mater. Reliab. 8, 332 共2008兲. 5 M. Ťapajna, U. K. Mishra, and M. Kuball, Appl. Phys. Lett. 97, 023503 共2010兲. 6 S. Y. Park, C. Floresca, U. Chowdhury, J. L. Jimenez, C. Lee, E. Beam, P. Saunier, T. Balistreri, and M. J. Kim, Microelectron. Reliab. 49, 478 共2009兲. 7 P. Makaram, J. Joh, J. A. del Alamo, T. Palacios, and C. V. Thompson, Appl. Phys. Lett. 96, 233509 共2010兲. 8 K. V. Smith, S. Brierley, R. McAnulty, C. Tilas, D. Zarkh, M. Benedek, P. Phalon, and A. Hooven, ECS Trans. 19共3兲, 113 共2009兲. 9 G. Koley, V. Tilak, L. F. Eastman, and M. G. Spencer, IEEE Trans. Electron Devices 50, 886 共2003兲. 10 C.-H. Lin, T. A. Merz, D. R. Doutt, M. J. Hetzer, J. Joh, J. A. del Alamo, U. K. Mishra, and L. J. Brillson, Appl. Phys. Lett. 95, 033510 共2009兲. 11 J. A. del Alamo and J. Joh, Microelectron. Reliab. 49, 1200 共2009兲; J. Joh, F. Gao, T. Palacios, and J. A. del Alamo, ibid. 50, 767 共2010兲. 12 L. J. Brillson, J. Vac. Sci. Technol. B 19, 1762 共2001兲. 13 P. Hovington, R. Gauvin, A. R. Couture, D. Drouin, P. Horny, and H. Demers, CASINO version 2.42, Monte Carlo simulation of electron trajectory in solids, 2002; http://www.gel.usherbrooke.ca/casino. 14 D. R. Doutt, C. Zgrabik, H. L. Mosbacker, and L. J. Brillson, J. Vac. Sci. Technol. B 26, 1477 共2008兲. 15 K. Wan, A. Alan Porporati, G. Geng, H. Yang, and G. Pezzotti, Appl. Phys. Lett. 88, 251910 共2006兲. 16 M. A. Reshchikov and H. Morkoç, J. Appl. Phys. 97, 061301 共2005兲. 17 C. B. Duke and C. Mailhiot, J. Vac. Sci. Technol. B 3, 1170 共1985兲. 18 A. R. Arehart, A. C. Malonis, C. Poblenz, Y. Pei, J. S. Speck, U. K. Mishra, and S. A. Ringel, “Next generation defect characterization in nitride HEMTS” 共unpublished兲.
