Transcript S - 中興大學
What Old Microwave Tricks Can Do in the New Nano Era? for semiconductor people of course Yuen-Wuu Suen Department of Physics, National ChungHsing University 孫允武 中興大學物理系 1 140.120.11.120 OUTLINES: 1. What is so “NANO” in microwaves? SCALES: time, length, energy 2. What do people think of using microwaves? Some novel thinking! 3. What I am doing about microwaves and makes them kind of “nano”!! Remember I am in the middle of Taiwan and between no where! 2 140.120.11.120 About Microwaves: Frequency: about 1GHz~40GHz Sources up to about 1000 GHz are easily available! ---so called millimeter (submillimeter) wave! FROM ELVA-1, RUSSIA http://www.elva-1.spb.ru/ Wavelength: NOTHING NANO???? (in vacuum) 3m 30cm 100MHz 3 1GHz 3cm 3mm 10GHz 100GHz 300mm 1000GHz 140.120.11.120 E h Energy: (in vacuum) 3m 30cm E E/kB 100MHz 1GHz 4×10-7eV 4×10-6eV 5mK (4.8) 50mK 3cm 3mm 10GHz 100GHz 1000GHz 0.04meV 0.4meV 4meV 5K 50K 0.5K 300mm Something I can never remember: h=6.63×10-34J·s =4.136×10-15eV·s kB =1.38×10-23J·K-1 =8.617×10-5eV·K-1 4 140.120.11.120 Comparison Between Energy Scales Bandgap of semiconductors about 0.2~3eV Conduction-band or about 30~900meV valence-band discontinuity Energy of an optical phonon about 30~50meV Energy spacing for a electron in a typical quantum well (QW) Fermi energy for electrons in a QW few meV Ionization energy of shallow donors or aceptors Exciton binding energy 4×10-7eV E E/kB 5mK 5 4×10-6eV 50mK 1GHz about 30~100meV about 10~50meV about 5~30meV 0.04meV 0.5K 0.4meV 5K 100GHz 4meV 50K 1THz 140.120.11.120 Let’s talk about something related to microwaves E Free carrier absorption EF phonon photon k At low f (compared to 1/tscatter), it is just joule heating. p2 c ( ) nr c( 2 c2 ) try some numbers: nf (cm-3) 1022 fp (Hz) 6 0.9×1015 2 p 4n f e 2 m 1018 1014 0.9×1013 0.9×1011 m 2 c ~ Np m 1 t Some possibility here for quantum dots or lowdimensional systems 10 10 0.9×109 140.120.11.120 Besides microwave radars, satellite communications, ovens…, what else microwaves can do? 7 140.120.11.120 Let’s try electron spins and magnets!! The famous 21 cm line??? (If you know quantum mechanics very well, it comes from the hyperfine interaction. H=AS1·S2) H Radio Astronomy f0=1.42GHz 0.53Å Electron spin splittings Slower than your P4! Very “NANO”!! (electron spin resonance ESR) For free electrons f (GHz)=28.0B(tesla) 0 In semiconductors DE=g*mBB g=2 depend on host semiconductors and fields GaAs 0.44, GaN 1.98, InSb –51, Ge –2.5, Si 1.98 8 140.120.11.120 Cyclotron resonance eB c m B z fc(GHz)=28.0(m0/m*)B(tesla) For electrons confined in a two-dimensional interface 1 1 E (n ) c g m B B 2 2 Landau levels 9 140.120.11.120 What can you do about the length scale? Why is it so long compared to “NANO”? The light travels so fast!! even divided by a reflective index. l=c/nf Then let’s translate it to something (somewave) slower. Maybe it will become more “NANO”. It comes out to be an acoustic wave. Or you want it on a surface, and then it should be a surface acoustic wave (SAW). cs~ 3000 m/s~ 10-5c 10 140.120.11.120 Let’s see the scales again! E 40meV 400meV f for EM of the same l as SAW 1013Hz 1014Hz 30mm 3mm 3m lSAW E E/kB 4eV 40eV 1015Hz 1016Hz 1017Hz 300nm 30nm 3nm 30cm 3cm 3mm 300mm 100MHz 1GHz 10GHz 100GHz 1000GHz 4×10-7eV 4×10-6eV 0.04meV 0.4meV 4meV 5K 50K 5mK (4.8) 11 50mK 0.5K 400V 140.120.11.120 How to generate SAWs? First you need a piezoelectric substrate. Electric field Mechanical strain You must place an inter-digit (comb-like) electrode on the surface as a transducer. OR One can use optical or thermal methods---but less defined properties of SAWs. laser SAW heat SAW lSAW 12 You need an e-beam writer to get “NANO”. 140.120.11.120 MORE about SAWs 1. A SAW on the surface of a piezoelectric substrate travels with a spatially modulated electric field, which gives a wave-like electric potential variation near the surface. 2. A SAW without an associated piezoelectric field is useless for microelectronics or nanoelectronics. For example, water wave. 3. The propagation properties of the SAW are very sensitive to the electrical or mechanical properties near the surface. Therefore, it is very useful for sensor applications. 4. The SAW is very useful on detecting the material properties of the length scale lSAW. 5. The SAW can provide a controllable electrostatic-like field in the scale of lSAW. Of course the distribution is flying. 13 140.120.11.120 A SAW Delay Line as a Detector Almost anything changed here can be detected. You can build an oscillator including this SAW sensor, or you can hook up to an expensive vector analyzer to measure S’s. 14 140.120.11.120 Anything GOOD to use SAW detectors?! 1. No contact! 2. Short wavelength compared to EM signals at the same frequency. 3. Low energy compared to EM signals at the same wavelength. 4. We can use SAWs to detect the special length scale in the sample via the size-resonance of SAWs and the sample. 15 140.120.11.120 Besides those old goodies and odds, what’s new and “NANO”? Some thinking…….. 1. Make energy levels in nanostructures microwave active, so that one can use microwave doing something. 2. One can use SAW sensor to detect some nano features in nanostructures. 3. One can use SAW to drive electrons or holes to anywhere your want, where anywhere=nanostructures or quantum dots……. 16 140.120.11.120 Microwave spectroscopy of a quantum-dot molecule T. H. OOSTERKAMP et al Nature 395, 873 - 876 (1998) Photon resonances in a double-dot sample. The metallic gates (1, 2, 3 and F) are fabricated on top of a GaAs/AlGaAs heterostructure with a twodimensional electron gas (2DEG) 100 nm below the surface. 17 Measured pumped current through the strongly coupled double-dot. 140.120.11.120 Determination of the complex microwave photoconductance of a single quantum dot H. Qin, F. Simmel, R. H. Blick, J. P. Kotthaus, W. Wegscheider, and M. Bichler Phys. Rev. B 63, 035320 (2001) Bias dependence of the quantum dot conductance in the vicinity of a single resonance. Amplitude |A| of the photoconductance measurement obtained with the two-source setup. Energy level alignment. 18 140.120.11.120 Experimental setup for the two-source measurement:Two millimeter waves with a slight frequency offset generated by two phase-locked microwave synthesizers ( f =18.08 GHz and df =2.1 kHz) are added, doubled and filtered. You can see some old microwave goodies in their setup. 19 140.120.11.120 Single-electron acoustic charge transport on shallow-etched channels in a perpendicular magnetic field J. Cunningham et al Physical Review B 62, pp. 1564-1567 (2000) fSAW=2.716 25 GHz, I=nefSAW They are trying to make a new electric current standard. 20 140.120.11.120 Flying potential and flying electrons 21 140.120.11.120 Quantum computation and spintronics Quantum computation using electrons trapped by surface acoustic waves C. H. W. Barnes, J. M. Shilton, and A. M. Robinson Physical Review B 62, pp. 8410-8419 (2000) (a) Schematic diagram showing the effective potential due to a SAW passing across a Q1DC; (b) potential through the center of (a), parallel to the Q1DC. 22 140.120.11.120 A Flying Qubit quantum bit Spin separation!! 23 Spin operation!! 140.120.11.120 Toward a quantum computer a SAW quantum-gate network Controlled-NOT gate Probably you need to know some quantum mechanics before going any further!? 24 140.120.11.120 Let’s talk about something opto…. If I can drive electrons around, I think I can drive electrons and holes (or excitons) around. Then I want to select a place for them to recombine at the time I suggest…..and I want…. Acoustically Driven Storage of Light in a Quantum Well C. Rocke, S. Zimmermann, A. Wixforth, J. P. Kotthaus, G. Böhm, and G. Weimann, Phys. Rev. Lett. 78, 4099 (1997); 25 140.120.11.120 What will happen if I drive electrons and holes to a quantum dot ( laser if you want)? Or to an array of quantum dots (lasers)… or … (anything you can imagine) Photon trains and lasing: The periodically pumped quantum dot C. Wiele, F. Haake, C. Rocke, and A. Wixforth PHYSICAL REVIEW A 58, 2680 (1998) 26 Quantum-dot laser with periodic pumping C.Wiele et al Physical Review A 60, pp. 4986-4995 (1999) 140.120.11.120 What are we doing in NCHU? 1. We have build a type-II phase lock loop (PLL) for pulsed microwave signal to detect very small phase variations due to absorption of microwave or SAW signals by electron systems. The resolution of the phase is better than 0.01 degree with average of –100dBm input power. The sensor is SAW delay line or coplanar waveguide. 2. We are setting up a simple e-beam writer. 3. We are fabricating high-frequency SAW transducers. 4. We are trying to digging small “nano” holes. 5. We are making lots of microwave connection cables. 27 140.120.11.120 Pulsed RF/Microwave PLL and Gated Averaging System DC-Coupled Frequency Modulation (FM) Directional Couplers (B) (b1) 黑色:低頻訊號 橙色:高頻訊號 Integrator 積分器 (I) (b2) Double-Balance Mixer PLL Precision Counter (C) Why pulsed? 1. Use low average power to prevent from heating 2. Use gated averaging technique to avoid direct EM interruption 3. Avoid the reflection and multiple reflection signals (M) RF or (A) Microwave Generator (b3) Pulse Generator (P) Gated Average Time Delay (J) Step Attenuator Power Splitter(S) (s2) (s1) High Speed Diode Switch (D) (H)Power Detector Amplifier (E) (G) Intensity Output StainlessSteal StainlessSteal Semirigid Coax Z (F) SAW Emitter IDT Semirigid Coax Sample Impedance Match Network Active LDES Region 28 What’s different from others: SAW Receiver IDT (F) Z Impedance Match Network Cryogenic Environment We use type II PLL, homebrew sample-&-hold circuits, and cheap lock-in amplifiers. 140.120.11.120 An improved homodyne amplitude detection scheme (if you still want to know some details, and still awake) Ref. Signal 0º 90º mixer To PLL 90º hybrid ~0 Power splitter A home-made vector meter?? 29 To amplitude detection Signal from the sample 140.120.11.120 Signal Gating & Averaging: ~200 ms set by lock-in amp s1(t) Peak power about –30dBm RF/Microwave pulse train 3~4 ms set by lock-in amp 0.2~2 ms set by pulse shaping circuit s1(t) time delay Direct coupled EM s2(t) signal of mixer or power detector s4(t) signal after SH Reflected signals sampling delay set by pulse generator s3(t) fed into lock-in sampling gate set by a pulse generator fed into the controlling node of a sample-and-hold circuit 30 140.120.11.120 Our system is working------ 31 Guess which one is the PLL system? 140.120.11.120 A semiconductor chip attached on the SAW delay line Chip tied on the SAW delay line BeCu SR coax 5mm He3 sample holder 32 IDT SAW transducer 140.120.11.120 Detection by Phase Lock Loop (PLL) PLL system phase=f1=b1l1 f0 =f1+ fs fs=bsls =b1l1+bs(B)ls sample Sample under detection Type-II PLL 33 Df0 =0 =Df1+D fs(B) =Db1l1+Dbs(B)ls known B:the parameter (magnetic field) changed in the experiment u:velocity of the wave D can be measured very accurately. 140.120.11.120 Reference From sample Keep at a constant phase difference Reference From sample Reference Df Due to the change of sample conditions Tuning the frequency to match the phase From sample 34 140.120.11.120 SAW Delay-Line Sensor L l Du Df l 1 u0 f0 L 2 1 ( xx ／σ 2 K eff kSAW 35 2 eff σ xx /σ M 2 1 (σ xx /σ M ) 2 K 2 ) M xx 0 , q xx 2f 0 , 2 l M u0 (1 2 ) GaAs:3.6×10-7 W-1 GaAs/LiNO3(Y-Z):1.8×10-6 W-1 140.120.11.120 Coplanar Waveguide (CPW) Sensor xx 0 , q 0 xx G 2 2D d eff W1 m 50W meandering CPW total length ls Electric field 36 140.120.11.120 Coplanar Waveguide (CPW) Sensor Some formulae: jm ( j ) jL(G jC) 0 eff L 1 Z0 C 2D xx d eff 37 m 1 m0 mr 0 eff m 2 Np m ln D W Z0 l F m 1 C , W m L 1 m G 2 D 1 W1 2 d eff m 1 2 rad b m 1 8 m or 2 D 8 2 Df m L f d xx l Z 2 u s 0 L 2 eff 1/ 2 W 1 140.120.11.120 Data read from SAW delay line ns =2.5×1011 f0=120MHz T=0.3K GaAs/AlGaAs 2DES cm-2 SAW Intensity Df/f0 0.001% =1 =2 0 2 4 6 8 10 12 B (Tesla) 38 140.120.11.120 Data read from CPW: Amplitude (arb. unit) 10 amplitude 8 6 Df 4 200kHz =2 2 0 =1 f0=1.39GHz 0 2 4 6 8 10 B (T) 39 140.120.11.120 More data: 30 T=0.3K f0= 2.92G f0= 25 2.92G 2.16G 20 1.74G 1.74G 1.39G Df Amplitude (arb. unit) 2.16G 1.39G 15 0.96G 10 0.96G 0.59G 0.59G =2 5 0 2 4 6 B (T) 40 500kHz =1 8 =1 =2 10 0 2 4 B (T) 6 8 10 140.120.11.120 High frequency IDT pattern made by e-beam lithography Still, not “nano” enough! 41 140.120.11.120 Time delay response of a pulse microwave input 42 140.120.11.120 So-Called Flows of MW modules, Graduate students, ……….. 43 140.120.11.120 UnderConstruction ^0^ 1. l<1mm e-beam writer Nano….. Nano….. 2. Acoustoelectric effect V or A spins 3. Quantum dots, spins, spintronics Nano….. Nano….. Nano….. Nano….. Nano….. Nano….. Nano….. Nano….. 4. Buying source of higher frequency, maybe to THz. 44 140.120.11.120