Relax s500 January 08

Relax s500 January 08Edit

Work in progress!!!

This page details the fabrication process of January 2008's Relax device. The substrate is the 2DEG s500. For the realistaion of this device we follow  4 insulating steps (Ohmic Contacts, Alignement Marks, Mesa, Gates):

1. Cleaning
   

   1. aceton bath with 40% ultrasounds fot 2'
   2. triclorethylene bath @100°C for 1', then  2' with ultrasounds
      

2. Ohmic Contacts

   1. Spin-coating with 50g/L PMMA @5krpm, 2krpm/s, for 60s at full speed.
   2. Cleaning the backside of the substrate with IPO in cotton sticks and sticking to a Si substrate with 30g/L PMMA
      
   3. PMMA "annealing" fot 45' @160°C
      
   4. Ebeam lithography
   5. Development: 1 MIBK- 3 IPO for 45s then IPO 10s
   6. Fred's Ohmic Contact [Ni40Ge600Au1200Ni250Au1000] (in Angstroms)
      
   7.  Annealing:
      1. Fred's Program #600 for oven cleaning
      2. Setting PID param at (100; 12; 1,6)
         
      3. Program 32
         1. ->5'' to 120°C for 40''
         2. ->10'' to 370°C for 2' (370°C is Ge-Al eutectic point)
         3. ->15'' to 440 for 50'' (just above 420°C, point at which the Au diffuses inside GaAs
      4. Do not forget to set back the PID parameters to the standard ones: PIDstd= (200; 10; 2)
         
   8. Contact test: linear I/V, with (0, 0) and (65microA, 50mV) => 1kOhm
3. Alignement Marks
   1. Spin-coating with 40g/L PMMA @5krpm, 2krpm/s, for 60s at full speed.
   2. Cleaning the backside of the substrate with IPO in cotton sticks and sticking to a Si substrate with 30g/L PMMA
   3. PMMA "annealing" fot 1h10min @160°C
   4. Ebeam lithography
      
   5. Development: 1 MIBK- 3 IPO for 45' then IPO 10'
   6. 20 nm TI- 200 nm Au evaporation
   7. Lift-off:
      1. 80°C triclorethylene for ~2min, with ambient room triclorethylene flashes
         
      2. acetone
      3. isopropanol
4. Mesa
   
   1. Spin-coating with 40g/L PMMA @5krpm, 2krpm/s, for 60s at full speed.
   2. Cleaning the backside of the substrate with IPO in cotton sticks and sticking to a Si substrate with 30g/L PMMA
   3. PMMA "annealing" fot 1h @160°C
   4. Ebeam lithography
      
   5. Development: 1 MIBK- 3 IPO for 45'then IPO 10'
   6. Desoxydation: HCl (37%) 5''
   7. 100nm Al evaporation (VAS, asking 66nm, so input thicness=76nm)
   8. Lift-off:
      1. 80°C triclorethylene for ~2min, with ambient room triclorethylene flashes from time to time (every 15s , 3-4 times)
         
      2. acetone
      3. isopropanol
   9. Etching with 1 H2O2 - 3 H3PO4 - 80 H2O sauce in 3 steps of 20'' give 45+-1 nm
   10. Warning if you ever see this before etching
   11. Check that etched regions are electrically disconected from mesa even in presence of light exciation (we check here R>50MOhm, limit of the aparatus)
       
5. Gates
   1. Dose Test  1
      
   2. Dose Test 2
   3. Desoxydation: HCl (10%) 20''
   4. Spin-coating with PMMA A2 @5krpm, 2krpm/s, for 60s at full speed.
   5. PMMA "annealing" fot 1h @160°C
   6. Ebeam lithography
   7. Development: 1 MIBK- 3 IPO for 30'then IPO 10
   8. 50nm Al evaporation (VAS, asking 32nm, so input thicness=42nm)
   9. Lift-off:
      1. 80°C triclorethylene for ~2min, with ambient room triclorethylene flashes from time to time (every 15s , 3-4 times)
         
      2. acetone
      3. isopropanol
   10. Electric contact tests fail. Too many desoxydation steps (and Al decaping with NaOH) killed the 2DEG (steps 4 and 5 repeated because of issues, see 4.10, 5.1)