Measur asurem emen ents of RF noi ts of RF noise i e in InG InGaAs/InAlA As/InAlAs r s recessed diodes: cessed diodes: Sign Signatures of shot-noise suppr es of shot-noise suppression ssion O. García-Pérez, J. Mateos , S. Pérez, T. González A. Westlund, J. Grahn 1
Shot Noise Shot-noise S I = qI is related to the discrete character of the electronic charge and is usually observed in electronic devices when carrier transport is ballistic or is limited by an energy barrier (Schottky diodes, tunnel diodes…) full shot noise = 1 Uncorrelated carriers Shot-noise S I = qI suppressed shot noise 1 Fano Factor = S I qI Correlated carriers enhanced shot noise 1 Origin of correlations: Pauli exclusion principle (degenerate semiconductors) Long-range Coulomb interaction (strong space-charge effects) The measurement of shot noise and the value of its corresponding Fano Factor can provide valuable insight about the transport dynamics inside semiconductor devices Javier Mateos – Universidad de Salamanca 2
Shot Noise Ballistic diode Shot noise suppression due to long range Coulomb correlations U nc nc high important space-charge effects Poissonian Poissonian N Contact 2 injection injection Contact 1 thermal thermal Active region equilibrium equilibrium x =0 x = L 15 qU/K B T 80 Potential (K B T/q) 40 10 20 L 10 5 L Dc 4 0 0 -5 0.0 0.2 0.4 0.6 0.8 1.0 x / L Full shot noise is recovered Presence of a barrier that disappears when voltage is increased when the barrier disappears Javier Mateos – Universidad de Salamanca 3
Shot Noise Recessed planar diodes=Slot diodes Ballistic diode (ungated HEMTs) U nc nc Poissonian Poissonian N Contact 2 injection injection Contact 1 thermal thermal Active region equilibrium equilibrium x =0 x = L Quasi-ballistic transport in the channel 15 qU/K B T Shot noise suppression expected 1.8 80 Potential (K B T/q) 40 10 1.6 20 Potential barrier 1.4 10 5 1.2 imposed by the surface Potential (V) 1.0 4 charges at the recess 0 0.8 0 0.6 0.4 -5 0.0 0.2 0.4 0.6 0.8 1.0 0.2 x / L 0.0 -0.2 Presence of a barrier that disappears -0.4 0.0 0.1 0.2 0.3 0.4 0.5 0.6 when voltage is increased Position(um) Javier Mateos – Universidad de Salamanca 4
Slot diodes: fabrication Slot diodes with different geometries have been fabricated and characterized L S L R L D Source Drain Recess cap InGaAs InAlAs InGaAs Channel InAlAs 2-8 nm Parameter Reference design Testedrange [nm] [nm] L s 200 200-800 L r 200 200-800 L d 550 300-1000 + Diodes without recess Javier Mateos – Universidad de Salamanca 5
Slot diodes: Monte Carlo simulations L s L r L d Source Drain L s = 200nm Contact + Resistance L d = 550nm Cap layer (0.35 Ω .mm) 160nm<L r <800nm 1.2 1.2 1.2 1.0 1.0 1.0 3 A/m) 3 A/m) 3 A/m) 0.8 0.8 0.8 Current density (x10 Current density (x10 Current density (x10 0.6 0.6 0.6 L R =160 nm (Exp.) L R =160 nm (Exp.) L R =160 nm (Exp.) L R =400 nm (Exp.) L R =400 nm (Exp.) 0.4 0.4 L R =400 nm (Exp.) 0.4 L R =800 nm (Exp.) L R =800 nm (Exp.) L R =800 nm (Exp.) L R =160 nm (MC) L R =160 nm (MC) L R =160 nm (MC) 0.2 0.2 0.2 L R =400 nm (MC) L R =400 nm (MC) L R =400 nm (MC) L R =800 nm (MC) L R =800 nm (MC) L R =800 nm (MC) 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Applied voltage (V) Applied voltage (V) Applied voltage (V) Batch C7, type 5 - Cap 8nm Batch C7, type 6 - Cap 2nm Batch C7, type 2 - Cap 4nm MC simulations reproduce the experimental DC curves and their dependence on L s , L r and L d by adjusting the values of the surface charges and including the ohmic contact resistances Javier Mateos – Universidad de Salamanca 6
Slot diodes: Monte Carlo simulations L s L r L d Source Drain L s = 200nm Contact + Resistance L r = 160nm Cap layer (0.35 Ω .mm) 300nm<L d <1000nm 1.2 1.2 1.2 1.0 1.0 Current density (x10 3 A/m) Current density (x10 3 A/m) 1.0 Current density (x10 3 A/m) 0.8 0.8 0.8 0.6 0.6 0.6 L D =300 nm (Exp.) L D =300 nm (Exp.) L D =300 nm (Exp.) L D =550 nm (Exp.) L D =550 nm (Exp.) L D =550 nm (Exp.) 0.4 0.4 L D =1000 nm (Exp.) 0.4 L D =1000 nm (Exp.) L D =1000 nm (Exp.) L D =300 nm (MC) L D =300 nm (MC) L D =300 nm (MC) L D =550 nm (MC) 0.2 0.2 0.2 L D =550 nm (MC) L D =550 nm (MC) L D =1000 nm (MC) L D =1000 nm (MC) L D =1000 nm (MC) 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Applied voltage (V) Applied voltage (V) Applied voltage (V) Batch C7, type 5 - Cap 8nm Batch C7, type 6 - Cap 2nm Batch C7, type 2 - Cap 4nm MC simulations reproduce the experimental DC curves and their dependence on L s , L r and L d by adjusting the values of the surface charges and including the ohmic contact resistances Javier Mateos – Universidad de Salamanca 7
Noise measurements Measurements performed on wafer with a VNA Agilent PNA-X N5244 (with dedicated receivers for high sensitivity noise power measurements) in the range between 20 and 30 GHz . PNA-X N5244 Device Reference plane V DC S 11 + Power density Javier Mateos – Universidad de Salamanca 8
Noise measurements x Non-recessed diode with L = 1.3 m L s = 200 nm, L r = 160 nm, L d = 300, 500, 1000 nm 20 3.0 10 L D =300nm L D =300nm L D =300nm L D =550nm 2.5 L D =550nm L D =550nm L D =1000nm 15 L D =1000nm L D =1000nm No recess (1300nm) No recess (1300nm) 2.0 2 /Hz) No recess (1300nm) Current (mA) Fano Factor 2qI Full Shot Noise -21 A 10 1.5 1 S I (x10 1.0 5 0.5 0.0 0 0.1 0 2 4 6 8 10 12 14 16 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Voltage (V) Current (mA) Voltage (V) L s = 200 nm, L d = 550 nm, L r = 160, 400, 800 nm 20 3.0 10 L R =160nm L R =160nm L R =400nm 2.5 L R =160nm L R =400nm L R =800nm 15 L R =400nm No recess (1300nm) L R =800nm L R =800nm 2 /Hz) 2.0 No recess (1300nm) Current (mA) No recess (1300nm) Fano Factor Full Shot Noise 2qI -21 A 10 1.5 1 S I (x10 1.0 5 0.5 0 0.0 0.1 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 2 4 6 8 10 12 14 16 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Current (mA) Voltage (V) Voltage (V) Javier Mateos – Universidad de Salamanca 9
Noise measurements x Non-recessed diode with L = 1.3 m L s = 200 nm, L r = 160 nm, L d = 300, 500, 1000 nm 20 3.0 10 L D =300nm L D =300nm L D =300nm L D =550nm 2.5 L D =550nm L D =550nm L D =1000nm 15 L D =1000nm L D =1000nm No recess (1300nm) No recess (1300nm) 2.0 2 /Hz) No recess (1300nm) Current (mA) Fano Factor 2qI Full Shot Noise -21 A 10 1.5 1 S I (x10 1.0 5 0.5 0.0 0 0.1 0 2 4 6 8 10 12 14 16 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Noise level much lower 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Voltage (V) Current (mA) Voltage (V) than full shot noise level L s = 200 nm, L d = 550 nm, L r = 160, 400, 800 nm = shot noise suppression 20 3.0 10 Increase of noise at high bias, L R =160nm L R =160nm L R =400nm 2.5 L R =160nm L R =400nm L R =800nm when current saturation starts 15 L R =400nm No recess (1300nm) L R =800nm L R =800nm 2 /Hz) 2.0 No recess (1300nm) (not visible without recess) Current (mA) No recess (1300nm) Fano Factor Full Shot Noise 2qI -21 A 10 1.5 1 S I (x10 1.0 5 0.5 0 0.0 0.1 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 2 4 6 8 10 12 14 16 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Current (mA) Voltage (V) Voltage (V) Javier Mateos – Universidad de Salamanca 10
Noise measurements x Non-recessed diode with L = 1.3 m L s = 200 nm, L r = 160 nm, L d = 300, 500, 1000 nm 20 3.0 10 L D =300nm L D =300nm L D =300nm L D =550nm 2.5 L D =550nm L D =550nm L D =1000nm 15 L D =1000nm L D =1000nm No recess (1300nm) No recess (1300nm) 2.0 2 /Hz) No recess (1300nm) Current (mA) Fano Factor 2qI Full Shot Noise -21 A 10 1.5 1 S I (x10 1.0 5 0.5 0.0 0 0.1 Stronger shot noise 0 2 4 6 8 10 12 14 16 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Voltage (V) Current (mA) Voltage (V) suppression for longer L d L s = 200 nm, L d = 550 nm, L r = 160, 400, 800 nm (and not depending on L r ) 20 3.0 10 L R =160nm L R =160nm L R =400nm 2.5 L R =160nm L R =400nm L R =800nm 15 L R =400nm No recess (1300nm) L R =800nm L R =800nm 2 /Hz) 2.0 No recess (1300nm) Current (mA) No recess (1300nm) Fano Factor Full Shot Noise 2qI -21 A 10 1.5 1 S I (x10 1.0 5 0.5 0 0.0 0.1 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 2 4 6 8 10 12 14 16 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Current (mA) Voltage (V) Voltage (V) Javier Mateos – Universidad de Salamanca 11
Noise models: discussion Barrier limited transport Shot noise S r =F·2 qI Rc Ls Lr Ld Ohmic region Thermal noise Ohmic region Thermal noise Always at equilibrium S s = 4 K B T Bias dependent resistance S d = 4 K B T R c +Rs R d (V) S s S r S d R c +R s R r R d (V) Javier Mateos – Universidad de Salamanca 12
Recommend
More recommend
