Electrically active defects in semiconductors induced by radiation Ivana Capan Rudjer Boskovic Institute, Croatia http://www.irb.hr/users/capan
Outline Radiation damage Capacitance transient techniques Experimental results: radiation induced defects in n-type Si and Ge Conclusion 2
Radiation damage Primary damage is the displacement of a lattice atom to create an interstitial and a vacancy. These can: Recombine no damage React with impurities in the silicon eg VO or displace atoms eg B i which are mobile and will react with other impurities or defects these are likely to have electrical activity The intrinsic defects may react with each other eg V 2 to generate electrically active centres or possibly inert defects. 3
Effects of particle type At low levels of irradiation with electrons or gamma rays almost all the lattice damage is in the form of simple point defects which can be studied by many techniques and are relatively well understood in Si. Damage from energetic particles or heavy ions is much more complex, difficult to study and with many unknowns in terms of electrical activity and reaction energetics. Main theme of our research in the last few years has been damage using fast neutron irradiation (at TRIGA Ljubljana) to help us to understand ion damage. This SRIM simulation of a 1MeV As ion in Ge shows high concentration clusters of primary defects which react to form stable complexes eg V 2, V 3 or larger clusters of vacancies or interstitials eg extrinsic stacking faults. 4
Factors affecting the defect species created by particle irradiation or implantation Type and concentration of impurities Particle species Particle energy Particle dose (fluence) Particles dose rate (flux) Temperature 5
Electrically active defects in nsi and nge VO VV VP VV VSb VP E c 0.17 0.23 0.42 0.45 0.29 0.38 0.24 0.31 0.35 E v 300 C 220 C 150 C 170 C 160 C 125 C n-type Si n-type Ge 6
Deep level transient spectroscopy (DLTS) At typical doping levels of 10 14-10 16 cm -3 sensitivity of the technique is 10 9-10 11 cm -3 7
DLTS analysis -3-4 [ 1 exp( c t )] C( t p ) = C0 p p ln(e/t 2 ) -5-6 e = κσ ( T ) T 2 e E kt DLTS signal 1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 0.01 0.1 1 (a) -7 85 90 95 100 105 1/kT 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3 0.01 0.1 1 Pulse width (s) (b) C0 NT = 2 N D C 2 2 f ( W ) = ( W λ ) ( W λ ) W N T 0 0 1 1 / C = 2 C 0 N D 1 f ( W ) 2 0 DLTS signal 4 3 2 1 T max =89.66K; C C 0 0 70 75 80 85 90 95 100 Temperature (K) 8
High-resolution Laplace DLTS [DLTS] [Laplace DLTS @ 170K] 9
Uniaxial stress Laplace DLTS The symmetry of a defect centre is obtained from the intensity ratio of the split DLTS lines under uniaxial stress applied along the three major crystallographic directions, <100>, <110> and <111>. VO in n-type CZ:Si The splitting pattern establishes the orthorhombic-i symmetry of the VO complex. 10
Irradiatated n-type CZ Si DLTS signal (arb. units) 0.5 0.4 0.3 0.2 0.1 Si ion implantation Neutron irradiation (/ 3.5) 4 MeV electrons (/ 4) E1 E2 E3 E4 (3) (2) (1) 0.0 50 100 150 200 250 300 Temperature (K) (1) 4 MeV electrons (F = 1 10 15 cm -2 ) (2) 1 MeV neutrons (F = 2 10 14 cm -2 ) (3) 800 kev Si ions (F = 1 10 9 cm -2 ) [J. Phys.: Condens. Matter 17 (2005) S2229] 11
Si:n 0 sample annealing study (1) as irradiated (2) 100 C (3) 150 C (4) 200 C VV(=/-) VV(-/0) (T < 200 ºC) 12
Majority (electron) and minority (hole) carrier traps in nge (low resistivity) Normalized DLTS signal (a.u.) 1.0 0.5 0.0-0.5-1.0 50 100 150 200 250? Temperature (K) V-Sb [Materials Science in Semiconductor Processing 9 (2006) 606 ] 13
Annealing study C, pf 1.25 1.00 0.75 0.50 0.25 0.00 Ge:Sb Neutrons 2.4x10 12 cm -2 U b = -5.0 V U p = 0 V t p = 1 ms e n = 80 s -1 as irrad 100 o C 160 o C 200 o C 300 o C 50 100 150 200 250 C, pf 0-1 -2-3 Ge:Sb U b = -3.0 V U p = +2.0 V t p = 1 ms e n = 80 s -1 Neutrons 2.4x10 12 cm -2 as irrad 100 o C 160 o C 200 o C 260 o C 50 100 150 200 250 Temperature, K Temperature, K 14
Annealing study + C-V Normalized DLTS peak intensity (a.u.) 2.0 1.5 1.0 0.5 0.0 V-Sb??? Minority trap Carrier concentration (cm -3 ) 7x10 14 6x10 14 5x10 14 C-V @ 140K 0 50 100 150 200 250 300 Annealing temperature ( C) 0 50 100 150 200 250 300 Temperature ( C) V-Sb + Sb VSb 2 15
Uniaxial stress LDLTS measurements CZ n-type (100) Ge:Sb 5Ωcm Dose: 5 10 11 cm -2 Normalised DLTS signal 1.0 neut. elec. 0.5 V-Sb 0.0 50 100 150 200 250 Temperatue (K) 16
Point-like defects and clusters as seen by Laplace DLTS 1.2 190K 64K LDLTS signal (a.u.) 0.8 0.4 V-Sb 0.0 10 100 1000 Emission, s -1 [Solid State Phenomena 131-133 (2008) 125] 17
Vacancy clusters 0.10 ev 0.04 0.03 70 K 65 K 75 K -40.4-40.6-40.8 C, pf 0.02 0.01 lnσ n -41.0-41.2-41.4 0.00 10-5 10-4 10-3 10-2 10-1 Pulse length, s -41.6 155 160 165 170 175 180 1/kT (ev -1 ) σ n (T)=σ n0 exp(- E nσ /kt) E nσ =0.04 ev; σ n0 =1 10-17 cm 2 Akceptor-like defekt Vacancy clusters!!! [Vacuum 84 (2010) 32] 18
V-SB <110> GPa 19
V-SB <111> V-SB <100> GPa 20
C 3v trigonal <100> NS <110> 1:1 <111> 1:3 C 1h - monoclinic 1:2 1:2:2:1 1:2:1 Sb-V T=195K vacancy clusters T=65K Laplace DLTS <111> <110> Sb - V Laplace DLTS <111> <110> <100> <100> 10 100 1000 Emission, s -1 100 1000 Emission, s -1 Sb-V pair C 3v Vacancy clusters C 1h??? 21
Conclusions Annealing study of n-type Si irradiated with fast neutrons clearly shows that the concentration of the double negative charge state of the divacancy increases with annealing temperature as vacancies and/or divacancies are released from cluster related defect, the E4 defect. Formation of the minority carrier trap in n-type Sb-doped germanim: Sb-V + Sb Sb 2 -V Low dose of fast neutrons has introduced the pointlike defects and small vacancy clusters Trigonal symmetry of the Sb-V pair 22
Branko Pivac (IRB) Milko Jakšić (IRB) Radojko Jačimović (IJŠ) Željko Pastuović (Australian Nuclear Science and Technology Organisation ) Anthony R. Peaker (Uni. of Manchester) Vladimir P. Markevich (Uni. of Manchester) 23