1886 was founded Alexander III Imperial Electrotechnical Institute

Transcription

1 SIC TECHNOLOGY RESERCH ND MODELING St Petersburg Electrotechnical University "LETI" Department of Micro- and Nanoelectronics Professor Victor Luchinin IEEE MOS-K GS Workshop, Sept. 16, 2011, Helsinki

2

3 1886 was founded Prof. lexander Popov, the 1st elected Director, Radio pioneer 1899 lexander III Imperial Electrotechnical Institute The first specialised electrical engineering higher education institution in Europe cademician G. I. lferov, Nobel Price Winner in Physics 2000, LETI graduate

4 Experimental research and modeling of bulk SiC crystal growing Growth model of SiC politypes in the matrix replication process under the non-equilibrium conditions Experimental research and modeling of the structure formation of SiC and ln layers on heterogeneous substrates Modeling of graphene layers on SiC surface formation Experimental research and modeling SiC power devices

5

6 1955 Method Lely 1976 Method LETI (ETU)

7

8 T g T s T max r T 1 r K sm T>2000 C (а) z T z 3 (б) K 5 sm

9 Simulation tool Softimpact

10 3 4

11

12

13

14 B B B B B C C C 2H 3C 4H 6H 8H 10H 15R B C B C B B C C B C B B C B C B C B C B C B C B B C C B C B C Si 3C 2H G M S W Г, % G ij

15 (0001) X [ 0001] F V divdgrad 2 X 2 D ; D X 2 2 F gradф x, F Ф 2 D X C 2 D X KT F i i 0 i F [ 1120 ] i X KT АФ D i X i s 0 V

16 Time selection i s - reproduction i s - transformation 2 ktv i Ф D 0 1

17 V кр, мкм/с H 15R 6H 8H 27R 21R T=2320 K (0001) C 33R - переход в 3C - переход в 6H 39R 19R j , м

18 ед S S S M мат р. j j изб. j ds dt ds dt dis dt des dt S' изб уп d i S dt CRYSTLLISTION MEDIUM Matrix Imperfection increase уп.j Delay of the odering process уп.структ.j р ел. l l S изб.j Crystal matrix уп. j n j 1 1 exp E / RT акт exp E / матр. j RT

19 Method of control 19 19R Control of not-equilibrium degree of crystallisation process Insertion of new Sructurial information Kinetic growing structural phase transition (self structuring in the open system) Matrix coping (replication) Evolutional selection R 33R Parametres and criteria of control j, j τ τ j S R 8H 21R 6H 15R 4H -C Tb 3 C ds dt j ΔS j изб τ S 2 1 2H 3C ln

20

21

22 SiН 4 С 3 Н 8 Si t ~ 1390 C SiC Si Equipment and process parameters Carrier gas: Н 2, process gases: SiH 4, С 3 Н 8 Substrate temperature: ± 5 С; C 3 H 8 flow: ml/min; SiH 4 flow : ml/min; Н 2 flow : 2 16 l/min. Pressure in chamber during growth: 1 Torr RF-heating of substrates at 30 khz

23 Тemperature Porous silicon formation Porous silicon 1000 С etching 1300 С carbonization 1390 С growth Time

24 Lattice mismatch Si and SiC approximatly 20%, Differences of temperature coefficients of linear extension ~8% Epi-layer 3C-SiC, 0,5 3 μm Buffer layer 3C-SiC, nm Nanoporous silicon layer, nm Silicon substrates p-, n- type (100) Characteristics of 3C-SiC-layer Hall concentrations: сm -3 ; Electron mobility: μ = сm 2 /Vс; Specific resistance: ρ v = 10-3 Оhm сm

25

26 SiC/l 2 O 3 Т С, V 1 /hour ln/si

27

28 ev Energy diagram for atom placed on binary compound surface tom placed on binary compound surface growth

29

30 Experimental and theoretical dependence of layer composition on methane content in gaseous mixture when deposited on cold substrate (P = Па) 1 - U=520 В, 2 - U=620 В.

31 Temperature dependence of SiC layer growth rate for different ratio of deposition flux main components: o Layers with Si excess, Layers similar to stoichiometric composition of SiC, Layers with C excess. 1 Si/C=1, 2 Si/C=0.7, 3 Si/C=1.2

32

33 - carbon bridges drift - evaporation of Si atoms

34 R3 Reconstruction cell Blue vectors are the basic cell vectors. The carbon atoms of the graphene coincidence cell are not shown

35 X- 100 V/div; Y 1 m/div U rev max = V X- 1 V/div; Y 200 m/div R on = 0.5 Оhm

36 Simulation tool TCD Silvaco

37 Cree substrate Forward current 1 А Reverse voltage 300 V Q rr = 0.9 nq 3.3 кv

38 Since the temperature control inside the closed cavity is rather complicated, a modeling of the thermal fields and fluxes in the crucible is required The different crystal structure free energy proximity requires to use the models of non-equilibrium nonlinear thermodynamics with ordering time factor separation Under highly non-equilibrium conditions of binary compound precipitation the analysis of chemical and structure ordering can be based on molecular kinetics models For Graphene on the SiC surface system the crystallochemical model of the structure formation (can be used) considering for the binding energy

39

40