Space Solar Cell Radiation Damage Modelling

Size: px

Start display at page:

Download "Space Solar Cell Radiation Damage Modelling"

Monica Harmon
6 years ago
Views:

1 Space Solar Cell Radiation Damage Modelling Scott R. Messenger, Ph.D. US Naval Research Laboratory, Washington, DC

2 Prediction is very difficult, especially if it s about the future. -Niels Bohr Danish physicist ( ) *Taken from M. Xapsos, NSREC 2008

3 Space Radiation and Effects

4 Outline 1. Motivation 2. The Space Radiation Environment 3. Solar Cell Space Radiation Degradation Modeling -JPL Equivalent Fluence (EQFLUX) -NRL Displacement Damage Dose (SCREAM) 4. Radiation Damage in Multijunction Solar Cells

5 Motivation Provide space solar cell community an alternate method for cell level degradation analyses Heritage method (JPL EQFLUX) requires several ground test energies Electrons (0.6, 1, 12 MeV) Protons (0.02, 0.05, 0.1, 0.3, 1, 3, 10 MeV) Alternate Method (NRL Displacement Damage Dose, DDD) needs significantly less ground data Electrons (1, >2 MeV) Protons (>1 MeV) AIAA S Space Solar Cell Qualification (Sects. 8.1 & 8.2) have been modified to include NRL DDD model as primary which can save about $150K in future space cell qualifications

6 Problem To generate ground irradiation data necessary to predict the effect of a particle spectrum (as that found in space) on a solar cell in orbit This is accomplished by reducing all of the ground data to a characteristic data set Electron and Proton Fluence Data (GaAs/Ge, 1991) Normalized Maximum Power Protons 9.5 MeV 3 MeV 1 MeV 0.5 MeV 0.3 MeV 0.2 MeV 0.1 MeV 0.05 MeV Electrons GaAs/Ge 1 Sun, AMO 25 o C Particle Fluence (cm -2 ) 0.6 MeV 1 MeV 2.4 MeV 12 MeV

7 Solution We want all data to collapse to a common basis The JPL method uses an equivalent 1 MeV electron fluence The NRL Method uses the displacement damage dose Normalized Pmax keVp keVp 500keVp 0.7 1MeVp 0.6 3MeVp MeVp 12 MeV MeVe 0.3 1MeVe keVe Proton Fit 0.1 Electron Fit 0.0 JPL Data Collapse GaAs/Ge 1.E+11 1.E+12 1.E+13 1.E+14 1.E+15 1.E+16 1.E+17 Equivalent 1 MeV Electron Fluence (cm -2 ) Normalized Maximum Power NRL Data Collapse Protons 9.5 MeV 3 MeV 1 MeV 0.5 MeV 0.3 MeV 0.2 MeV Electrons 0.6 MeV 1 MeV 2.4 MeV 12 MeV GaAs/Ge 1 Sun, AM0 T=25 o C Neutrons 1 MeV equiv Displacement Damage Dose (MeV/g)

8 Outline 1. Motivation 2. The Space Radiation Environment 3. Solar Cell Space Radiation Degradation Modeling -JPL Equivalent Fluence (EQFLUX) -NRL Displacement Damage Dose (SCREAM) 4. Radiation Damage in Multijunction Solar Cells

9 Space Radiation Environment electrons protons p + p + e - p + e- e - COVERGLASS ACTIVE CELL e - SUBSTRATE Omnidirectional Continuous Energy Spectrum p + p + PANEL Multi-particle (light and heavy ions, neutrons?) p + e - p +

10 Space Radiation Environment electrons protons p + p + e - p + e- e - COVERGLASS ACTIVE CELL e - SUBSTRATE Omnidirectional Continuous Energy Spectrum p + p + PANEL Multi-particle (light and heavy ions, neutrons?) p + e - p +

11 AE8MIN Integral Flux > E (elec/cm 2 /s) AE8MIN Electron Spectra (Static) 1.E+09 1.E+08 1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 1.E+02 1.E+01 1.E L Shell Value MeV

12 AP8MIN Proton Spectra (Static) AP8MIN Integral Flux > E (prot/cm 2 /s) 1.E+09 1.E+08 1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 1.E+02 1.E+01 1.E L Shell Value MeV

13 Space Radiation Spectra 1.E km, circular, 63 o, 1 year duration Integral Spectra Differential Spectra 1.E+13 Integral Fluence ( cm -2 ) 1.E+12 1.E+11 1.E+10 1.E+09 1.E+08 1.E+07 1.E+06 AP8MIN AP8MAX CRRESPRO Quiet CRRESPRO Active PSB97 Solar Protons Proton Energy (MeV) Differential Fluence (cm -2 MeV -1 ) 1.E+12 1.E+11 1.E+10 1.E+09 1.E+08 1.E+07 1.E+06 1.E+05 1.E+04 AP8MIN AP8MAX CRRESPRO Quiet CRRESPRO Active PSB97 Solar Protons Proton Energy (MeV) *The space radiation spectra is generally considered accurate to ~2X. This is the dominant error source for any degradation analysis leading to MARGIN!!!!

14 What is a Coronal Mass Ejection?

15 Space Weather - Solar Proton Event Data October 19, 1989 Event MeV Protons GOES-7 SEM IMP-8 GME /19/ /21/ /23/ /25/ /27/ /29/ /31/ /2/ /4/ /6/ /8/ /10/ /12/1989 Date & Time Differential Flux (cm -2 s -1 sr -1 MeV -1 )

Electron and Proton Radiation Environment (CEASE data onboard TSX5, DSP21, HEO and ICO Satellites) *The present radiation environment models are known to have an accuracy of at least a factor of 2.

17 Electron and Proton Radiation Environment (CEASE data onboard TSX5, DSP21, HEO and ICO Satellites) *The present radiation environment models are known to have an accuracy of at least a factor of 2. AP(E)9 will greatly update the environment data. A beta version has been created in May, The NASA Radiation Belt Storm Probe (RBSP) experiment (launch in 2012) is expected to greatly enhance the understanding. Any model is only good to the data that is given!

22 GPS Environment Data (LANL) 1.E+08 Averaged Integral Fluence by Year (e /cm 2 /sec) 1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 1.E+02 1.E+01 1.E+00 1.E 01 1.E 02 1.E 03 1.E 04 1.E 05 1.E All data AE8MIN/AE8MAX SV41/AE8MAX 1.E 02 1.E 01 1.E+00 1.E+01 Electron Energy (MeV) Electron Energy (MeV)

23 12/10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ E E E E+09 GPS Environment Data (LANL) Daily integrated 1 MeV electron fluence 6 mil SiO2 RDC LANL 6 mil SiO2 RDC AE8MIN 6 mil SiO2 RDC AE8MAX

24 GPS Array Data with Environment Data (LANL) 1.00 Maximum Power (P max ) Remaining Factor /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/2009 Pmax/Pmax0

25 GPS Array Data with Environment Data (LANL) 1.00 Maximum Power (P max ) Remaining Factor /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/2009 Pmax/Pmax0

26 GPS Array Data with Environment Data (LANL) 1.00 Maximum Power (P max ) Remaining Factor /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/ /10/2009 Pmax/Pmax0

27 Bottom Line The environment is a significant source of error, depending on the mission and lifetime.

28 Outline 1. Motivation 2. The Space Radiation Environment 3. Solar Cell Space Radiation Degradation Modeling -JPL Equivalent Fluence (EQFLUX) -NRL Displacement Damage Dose (SCREAM) 4. Radiation Damage in Multijunction Solar Cells

29 Displacement Damage Incident particle INTERACTIONS Primary knock-on atom (PKA) PROTONS Rutherford Nuclear Elastic Nuclear Inelastic ELECTRONS Rutherford DAMAGE SIMPLE DEFECTS Vacancies Interstitials Scattered particle COMPLEXES Vacancy/impurity Multi-vacancy/interstitial Clusters

30 Effect of Particle-Induced Damage p n + p InP H3 H4 H Compensation Centers Carrier removal E ED EA EC DLTS Signal (pf) EC EA ED H7 Trap Energy 0.67 H3 H4 H5 H7 E Recombination Centers Diffusion Length Degradation Temperature (K) 0.00

31 JPL and NRL Methods NASA Jet Propulsion Laboratory (Pasadena, CA) Calculate equivalent 1 MeV electron fluence for mission Uses empirically determined relative damage coefficients (RDCs) Read EOL power from measured 1 MeV electron curve Naval Research Laboratory (Washington, DC) Calculate displacement damage dose (D d ) for mission Uses calculated values of nonionizing energy loss (NIEL) Read EOL power from measured characteristic curve *Both methods have the same general approach.

32 Orbital Degradation Calculations d e(ee) d p(ep) 1MeV electron RDC(Ee,t)dEe Cpe RDC(Ep, t)de de de e JPL Method C pe is determined empirically (75-90% degradation level) p p NRL Method D d d(e de p p ) NIEL(E p )de p R ep d(e de e e ) NIEL(E R ep is determined empirically (ratio D xe /D xp ) e NIEL(Ee ) ) NIEL(1 MeV) n1 de e *Both methods currently rely on single-valued electron-proton degradation correlation. However, these methods can be adapted to account for degradation-dependent correlation (WCPEC 2006).

33 Heritage Model JPL Equivalent Fluence Method Data needed Protons E = 0.02, 0.05, 0.1, 0.3, 1, 3, 10 MeV = 3e9 to 2e12 p + /cm 2 Electrons E = 0.6, 1, 12 MeV = 2e13 to 1e16 p + /cm 2 Electron and Proton Fluence Data (GaAs/Ge, 1991) Normalized Maximum Power Protons 9.5 MeV 3 MeV 1 MeV 0.5 MeV 0.3 MeV 0.2 MeV 0.1 MeV 0.05 MeV Electrons GaAs/Ge 1 Sun, AMO 25 o C Particle Fluence (cm -2 ) 0.6 MeV 1 MeV 2.4 MeV 12 MeV

34 RDC Calculation Electron and Proton Fluence Data (GaAs/Ge, 1991) Normalized Maximum Power (10 MeV Protons & 1 MeV electrons) Protons 9.5 MeV 3 MeV 1 MeV 0.5 MeV 0.3 MeV 0.2 MeV 0.1 MeV 0.05 MeV Data exist for Pmax, Isc, and Voc for both electrons and protons Solar cell technologies available include Si, GaAs/Ge, MJ (2001) Empirical values for C ep (based on 75 90% level) Measurement intensive ($$) 80% BOL Electrons Particle Fluence (cm -2 ) GaAs/Ge 1 Sun, AMO 25 o C 0.6 MeV 1 MeV 2.4 MeV 12 MeV Electron Damage Coefficients Relative P max Damage Coefficient *Relative to 1 MeV normal incidence data, w/o coverglass Normal incidence no coverglass Coverglass Thickness 0 mil 1 mil 3 mil 6 mil 12 mil 20 mil 30 mil 60 mil Electron Energy (MeV) Proton Damage Coefficients Relative P max Damage Coefficient Coverglass Thickness 0 mil 1 mil 3 mil 6 mil 12 mil 20 mil 30 mil 60 mil GaAs/Ge (1991) *Relative to 10 MeV proton normal incidence data, w/o coverglass Normal incidence no coverglass GaAs/Ge (1991) Proton Energy (MeV)

35 Equivalent Fluence Analyses of GaAs/Ge Solar Cells Norm Pmax Norm Voc JPL Pmax vs Phi Protons 1 MeV electrons 5th order fit 10 MeV proton fluence data 1 MeV electron fluence data 1.E+10 1.E+11 1.E+12 1.E+13 1.E+14 1.E+15 1.E+16 1.E+17 1.E+18 Phi(10 MeV p+) and Phi(1 MeV e-) (cm -2 ) JPL Voc vs Phi Protons 1 MeV electrons 5th order fit 10 MeV proton fluence data 1 MeV electron fluence data 1.E+10 1.E+11 1.E+12 1.E+13 1.E+14 1.E+15 1.E+16 1.E+17 Phi(10 MeV p+) and Phi(1 MeV e-) (cm -2 ) P( ) P 0 1 C log1 10 MeV Proton and 1 MeV electron fit results 10 MeV Protons Pmax Voc Isc value error value error value error C Phix 1.92E E E E E E+11 Cep MeV Electrons Pmax Voc Isc value error value error value error C Phix 2.38E E E E E E+14 x

36 Initial Omnidirectional Spectrum Fluence (cm 2 MeV) JPL Equivalent Fluence Method 5000 km, circular, 60 0 orbit (1 year duration) Proton Energy (MeV) Equivalent 1 MeV Electron Fluence w/c ep Relative P max Damage Coefficient Proton Damage Coefficients Coverglass Thickness 0 mil 1 mil 3 mil 6 mil 12 mil 20 mil 30 mil 60 mil *Relative to 10 MeV normal incidence GaAs/Ge data, (1991) w/o coverglass, based on P max Proton Energy (MeV) 1 MeV Electron P max Degradation 1 MeV Electron Fluence (e - /cm 2 ) km, circular, 60 o orbit (1 year duration) GaAs SiO 2 Coverglass Thickness (mil) Normalized Pmax GaAs/Ge 1.E+11 1.E+12 1.E+13 1.E+14 1.E+15 1.E+16 1.E+17 Equivalent 1 MeV Electron Fluence (cm -2 ) 200keVp 300keVp 500keVp 1MeVp 3MeVp 9.5MeVp 12 MeV 2.4MeVe 1MeVe 600keVe Proton Fit Electron Fit

37 JPL Model Pros/Cons Pros: Heritage (developed in the 1980s) Widely available and already incorporated into many space radiation suites (SPENVIS, Space Radiation TM, OMERE, etc.) Cons: Much ground test data needed ($$) Requires 1 MeV electron (to 12 MeV) AND 10 MeV proton data Currently available for Si (1982), GaAs/Ge (1996), MJ (1999) Program not particularly user friendly in FORTRAN version Entire calculation is technology specific (every design change needs requalification, $$) Calculation of omnidirectional RDCs for covered cells not trivial and coverglass and technology specific Assumes that the total damage can be characterized by 1 MeV electrons which may not be appropriate for proton-dominated orbits

38 Proposed Model NRL Displacement Damage Dose Method Data needed Protons E = 3 MeV = 1e10 to 1e13 p + /cm 2 Electrons E = 1 and 5 MeV = 1e13 to 1e16 p + /cm 2 Normalized Maximum Power Electron and Proton Fluence Data (GaAs/Ge, 1991) Protons 3 MeV Electrons GaAs/Ge 1 Sun, AMO 25 o C 1 MeV 5 MeV Particle Fluence (cm -2 )

39 NonIonizing Energy Loss NIEL= Rate at which energy is lost to nonionizing events; analogous to LET or stopping power for ionizing events (UNIT=MeV/cm or MeVcm 2 /g) NIEL(E) min ( T d ) d(,e) T(,E)L[T(, E)]d d Differential scattering cross section for displacements Recoil energy Lindhard partition factor NOTE: Energy dependence of NIEL similar to experimental RDCs

40 NIEL in Si and GaAs Si *T d = 21 ev GaAs *T d = 10 ev, Ga & As Si NIEL (MeVcm 2 /g) Proton Electron Neutron GaAs NIEL (MeVcm 2 /g) Proton Electron Neutron Particle Energy (MeV) Particle Energy (MeV) *NIEL calculation available for any charged particle in any material *Neutron NIEL determined from Displacement Kerma calculation NIEL (MeVcm 2 /g) = KERMA (MeVmb) x (10-27 N A /A)

41 Displacement Damage Dose Analysis of GaAs/Ge Solar Cells Normalized Maximum Power GaAs NIEL (MeVcm 2 /g) Protons Measured Data 9.5 MeV 3 MeV 1 MeV 0.5 MeV 0.3 MeV 0.2 MeV 0.1 MeV 0.05 MeV Neutrons 1 MeV equiv Particle Fluence (cm -2 ) w/ NIEL Proton Electron Neutron GaAs/Ge 1 Sun, AMO 25 o C Electrons 0.6 MeV 1 MeV 2.4 MeV 12 MeV Particle Energy (MeV) GaAs *T d = 10 ev, Ga & As Normalized Maximum Power Characteristic Curves D d Protons 9.5 MeV 3 MeV 1 MeV 0.5 MeV 0.3 MeV 0.2 MeV D d (1 MeV electron) Electrons 0.6 MeV 1 MeV 2.4 MeV 12 MeV Neutrons 1 MeV equiv. GaAs/Ge 1 Sun, AM0 T=25 o C Displacement Damage Dose (MeV/g) Calculated NIEL gives energy dependence of damage coefficients (well matched to RDCs) Characteristic curves can be fit to simple expressions (similar to JPL handbook method) Characteristic curve can be generated using minimal ground test data (only 1 proton and two electron energies)

42 Dd Analyses of GaAs/Ge Solar Cells Dd(E) (E) NIEL(E) NIEL(E) NIEL(E ref ) n1 P(D P 0 d ) 1 C log1 D D d x 1.0 Norm Pmax E+07 1.E+08 1.E+09 1.E+10 1.E+11 1.E+12 Dd(prot) and Dd(1 MeV elec) (MeV/g) 1.00 GaAs/Ge Proton Pmax 1 MeV Electron Pmax Proton Pmax data 1 MeV Electron Pmax data Protons Pmax Voc Isc value error value error value error C Dx 1.10E E E E E E+08 Electrons Pmax Voc Isc value error value error value error C Dx 6.90E E E E E E+09 n Norm Voc GaAs/Ge Proton Voc 1 MeV Electron Voc Proton Voc data 1 MeV Electron Voc data Norm Isc GaAs/Ge Proton Isc 1 MeV Electron Isc Proton Isc data 1 MeV Electron Isc data E+07 1.E+08 1.E+09 1.E+10 1.E+11 1.E+12 Dd(prot) and Dd(1 MeV elec) (MeV/g) E+07 1.E+08 1.E+09 1.E+10 1.E+11 1.E+12 Dd(prot) and Dd(1 MeV elec) (MeV/g)

43 MJ Solar Cell Radiation Response in terms of D d *GaAs NIEL used in the correlation 1.1 Spectrolab EOL 3J Cells n/p cells 1.1 Emcore 3J Cells Norm P mp D d proton electron Energy (MeV) C = 0.3 D x = 3x10 9 MeV/g n = 1.6 Rep = 0.3 Data from Marvin Displacement Damage Dose (MeV/g) (E) (E) NIEL(E) NIEL(E) NIEL(E ref ) n1 Norm P mp P(D P Displacement Damage Dose (MeV/g) *Experimentally determined variables (C, D xp, D xe, n) proton electron Energy (MeV) ) D 1 C log1 C = D x = 1.2x10 9 MeV/g n = 1.8 Rep = 0.17 d d xe Rep 0 D D x xp D

44 Emcore ATJ Solar Cell Radiation Response 12 MeV 2 MeV 1 MeV 0.6 MeV EffFitElec 0.05 MeV 0.1 MeV 0.3 MeV 1 MeV 2.5 MeV 10 MeV EffFitProt *This shows that the Emcore ATJ multijunction (3J) solar cell is wellbehaved in that indeed a single characteristic curve is generated from the ground data. This data set is included in SCREAM development.

45 NRL Displacement Damage Dose Method Differential Fluence (cm -2 MeV -1 ) Incident and SDS (Isotropic) NonIonizing Energy Loss (2003) SiO 2 Coverglass Thickness 3 mil 12 mil 30 mil 5000 km, Circular Orbit 60 Inclination 5 year mission Uncovered Proton Energy (MeV) GaAs NIEL (MeVcm 2 /g) Proton GaAs *T d = 10 ev, Ga & As Proton Energy (MeV) D d (MeV/g) Total Mission Dose 5000 km, circular, 60 o (1 Year Mission) GaAs SiO 2 Thickness (mil) Normalized Maximum Power D d (1 MeV electron) GaAs/Ge 1 Sun, AM0 T=25 o C P max Degradation Displacement Damage Dose (MeV/g)

46 NRL Model Pros/Cons Pros: Few ground test measurements needed (3) Ground test particle energies can be conveniently chosen Shielding algorithm is independent Allows for rapid analysis of emerging cell technologies Allows for easy trade studies Can combine data from different experiments Allows for alternate radiation particles (neutrons, alphas, etc.) Cons: Lack of heritage (developed in the mid-1990s) More suited for sufficiently thin devices (~few mm) Uniform damage deposition required over active region Program currently not available to general public No interlaboratory cross calibration of method

47 Analytical Model Comparison Proton Dd to 1 MeV electron equivalent fluence 1MeV electron D C p 1 D * NIEL(1) Ce dp xe 1 D xp 1 MeV electron fluence to equivalent proton Dd D dp Ce * NIEL(1) Cp 1MeV electron Dxp 1 1 Dxe *General rule of thumb: MeV e - /cm 2 ~ MeV/g

48 Orbit Examples (SPENVIS) 1 year, 700 x km, 63.4 o, w/ solar event protons Fluence Spectra GaAs/Ge Cell degradation vs. Coverglass Thickness Integral Fluence (cm -2 ) 1.E+16 1.E+15 1.E+14 1.E+13 1.E+12 1.E+11 1.E+10 1.E+09 1.E+08 1.E x km, 63.4 o, ESP Total Fluence, 1 year Trapped Protons Solar Protons Trapped Electrons Energy (MeV) Normalized Remaining Parameter x km, 63.4 o, 1 year (Proton-dominated) Pmax JPL Voc JPL NRL Proton Dd Pmax NRL Proton Dd Voc GaAs/Ge Coverglass Thickness (mils SiO 2 )

49 Orbit Examples (SPENVIS) 15 year GEO w/ solar event protons Integral Fluence (cm -2 ) 1.E+17 1.E+16 1.E+15 1.E+14 1.E+13 1.E+12 1.E+11 1.E+10 1.E+09 1.E+08 1.E+07 Fluence Spectra 15 year, GEO, ESP Total Fluence Trapped Protons Solar Protons Trapped Electrons Energy (MeV) Normalized Remaining Parameter GaAs/Ge Cell degradation vs. Coverglass Thickness GEO, 15 year (Electrons & Protons) Pmax JPL Voc JPL NRL Proton Dd Pmax NRL Proton Dd Voc GaAs/Ge Coverglass Thickness (mils SiO 2 )

50 Why Do The Models Agree? The energy dependence of the experimentally determined RDCs closely match that of the calculated NIEL. Relative P max Damage Coefficient Protons SJ GaAs/Ge 2J InGaP/GaAs/Ge 3J InGaP/GaAs/Ge CIGS NIEL GaAs JPL MJ RDCs SRIM MJ RDCs *Parameters normalized to value at 10 MeV Energy (MeV) Relative P max Damage Coefficient Electrons SJ GaAs/Ge 2J InGaP/GaAs/Ge 3J InGaP/GaAs/Ge CIGS 1 MeV Equiv. NIEL GaAs (n=1.7) 1 MeV Equiv. NIEL CIGS (n=2) *Parameters normalized to value at 1 MeV Energy (MeV)

51 DDD Implementation History Quickbasic code (user proton spectrum input) Developed by NRL with Ed Burke (mid 1990 s) Transformed into Mathcad (mid 1990 s to present) Enabled both proton and electron conditions in 2005 NASA GRC support to Maxwell Labs (AP8/AE8) 1 st version of SAVANT (late 1990 s) Developed for inclusion into NASA Environmental Work Bench 1 st NASA Living With a Star (LWS SET) Data Mining NRA NRL and NASA GRC created a FORTRAN based, stand alone, Windows TM based, version of SAVANT (2003) distributed by NASA MSFC (SEE Program, now defunct) Only beta version produced (large laundry list created) SCREAM (Solar Cell Radiation Environment Analysis Models) Transformed MATHCAD version into MATLAB based executable Available by request Spenvis/Mulassis (MC SCREAM) All of the components are there except solar cell damage info The enabling component is MULASSIS to calculate the SDS User interface created by Daniel Heynderickx (DHC, Consultancy)

52 SAVANT DDD Analysis Code SAVANT: Solar Array Verification and ANalysis Tool (NASA, NRL, OAI)

53 SCREAM Integral Radiation Spectrum Input (user input) Converts the log log spectral data into a 10 th order polynomial fit Differentiates the polynomial fit to get the differential spectrum Shielding Range data Protons and ions: SRIM 2006 Electrons: ESTAR Slowed Down Spectrum (SDS) Calculation Slab model adjoint calculation based on Haffner model (1967) Benchmarked with Space Radiation TM, Mulassis Calculation of DDD (integral of NIEL*SDS over energy) NIEL data from WINNIEL2 (ions) and MATHCAD (electrons) Calculation of Parametric Degradation Uses 5 empirically determined parameters for each metric

54 Start-Up Page

55 Materials Menu Options

56 Cell Technology Options

57 Degradation Parameter Options

58 After selecting HEO_TrP.txt and entering, the data are plotted

59 After choosing CMX and 6 mils, and pressing Slowed-Down Spectrum ( Primary Spectrum button selected) At this time, you can save the SDS using the FILE - Save SDS menu option

60 After pressing Displacement Damage Dose

61 MC-SCREAM Incident differential radiation spectra (SPENVIS) Can choose several proton, electron, and solar proton models Calculation of the slowed down spectra after having passed through shielding (analytical, MULASSIS) Monte Carlo model Several material layers can be used to describe cell Calculation of total DDD for the mission (SPENVIS) NIEL data from WINNIEL2 (ions) and MCAD (electrons) Already implemented in Mulassis Determination of the expected cell degradation Uses 5 empirically determined parameters for each metric Will have several cell technologies included User entry option for proprietary/developmental cells

62 Output: 1.Slowed down spectra for each incident behind shielding 2. DDD for each particle

63 Outline 1. Motivation 2. The Space Radiation Environment 3. Solar Cell Space Radiation Degradation Modeling -JPL Equivalent Fluence (EQFLUX) -NRL Displacement Damage Dose (SCREAM) 4. Radiation Damage in Multijunction Solar Cells

64 Multijunction Solar Cell Radiation Response Monolithic 3J InGaP 2 /GaAs/Ge (AM0, 1 sun ~28%) I 1 I 2 I 3 InGaP 2 GaAs Ge 0.8 m 3 m 300 m Monolithic: Current-limiting, I cell =minimum(i 1,I 2,I 3 ) Whichever cell is the softest will control the overall cell performance. MJ cell design can alleviate this to some extent by forcing the cell to degrade by the most radiation hard subcell. The (beginning-of-life) BOL properties are slightly sacrificed in such a design for better (end-of-life) EOL behavior.

65 Ground Test MJ Solar Cell Data P max Proton Degradation vs. Displacement Damage Dose (D d ) 1.0 Remaining Factor of P max Proton Energy 30 kev 50 kev 70 kev 100 kev 150 kev 250 kev 380 kev 1 MeV 2 MeV 3 MeV 5 MeV Displacement Damage Dose (MeV/g) D d = NIEL x Another good example of NIEL correlation! *T. Sumita, M. Imaizumi, S. Matsuda, T. Ohshima, A. Ohi, and T. Kamiya, Proc. 19 th EPVSEC, Paris, 2004.

66 Ground Test MJ Solar Cell Data P max Proton Degradation vs. Displacement Damage Dose (D d ) Remaining Factor of P max Proton Energy 30 kev 50 kev 70 kev 100 kev 150 kev 250 kev 380 kev 1 MeV 2 MeV 3 MeV 5 MeV GaAs InGaP Displacement Damage Dose (MeV/g) *T. Sumita, M. Imaizumi, S. Matsuda, T. Ohshima, A. Ohi, and T. Kamiya, Proc. 19 th EPVSEC, Paris, Two separate damage curves are apparent! Which damage curve will apply in space?

67 Proton-Induced QE Degradation in MJ Cells 1.0 InGaP/GaAs/Ge 1.0 InGaP/GaAs/Ge Quantum Efficiency kev 50 protons kev Protons Solid lines: Unirradiated Dashed lines: 1x10 12 p + /cm 2 Quantum Efficiency kev protons 100 kev Protons Solid lines: Unirradiated Dashed lines: 1x10 12 p + /cm Wavelength (nm) Wavelength (nm) 1.0 InGaP/GaAs/Ge 1.0 InGaP/GaAs/Ge Quantum Efficiency kev protons 400 kev Protons Solid lines: Unirradiated Dashed lines: 1x10 12 p + /cm 2 Quantum Efficiency MeV protons 1 MeV Protons Solid lines: Unirradiated Dashed lines: 1x10 12 p + /cm Wavelength (nm) Wavelength (nm)

68 SRIM Simulation 3J InGaP 2 /GaAs/Ge structure 3 Proton Energies 63.1 kev (0.6 m range) 251 kev (2 m range) 1 MeV (11 m range) InGaP 2 GaAs Ge 0.8 m 3 m 300 m Monoenergetic, normal incidence( normal SRIM run) Monoenergetic, omnidirectional TRIM.DAT Spectrum, omnidirectional SRIM input file Spectrum, non-omnidirectional

69 Monoenergetic, Unidirectional Irradiations Energy Absorbed by Recoils (kev/um) 1.E+01 1.E+00 1.E-01 InGaP 63 kev (norm.) 251 kev (norm.) 1 MeV (norm.) GaAs 1.E-02 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 Depth (um) *Results from SRIM 2003 v.26 ( Ge Remaining Factor of P max InGaP degradation 0.6 Proton Energy 30 kev kev 70 kev GaAs degradation kev 150 kev kev 380 kev MeV 2 MeV MeV 5 MeV Displacement Damage Dose (MeV/g) *T. Sumita, M. Imaizumi, S. Matsuda, T. Ohshima, A. Ohi, and T. Kamiya, Proc. 19th EPVSEC, Paris, Typical ground test conditions (not space conditions) Nonuniform vacancy distribution Bragg Peak at end of track Different energies can preferentially degrade one sub junction This effect is not seen in 1 MeV electron irradiation (longer ranges)

70 Monoenergetic, Omnidirectional Irradiation 1.E+01 InGaP GaAs Ge Energy Absorbed by Recoils (kev/um) 1.E+00 1.E kev (omni.) 251 kev (omni.) 1 MeV (omni.) 1.E-02 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 Depth (um) More uniform vacancy distribution than a unidirectional beam Bragg peak not seen *Results from SRIM 2003 v.26 using special input file (TRIM.DAT) which specifies random incident angles (via direction cosines) over 2 geometry

71 Energy Spectrum, Omnidirectional Irradiation 1.E+00 InGaP GaAs Ge Energy Absorbed by Recoils (kev/um) 1.E-01 1.E-02 HEO orbit, 3 mils SiO2 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 Depth (um) Representative of exposure in the space radiation environment The vacancy distribution profile is nearly uniform over active region No special effects due to low energy protons apparent! *Results from SRIM 2003 v.26 using special input file (TRIM.DAT) which specifies random incident angle and energy to simulate HEO spectrum (3 mil SiO 2 )

72 Blanket Structure for Monte Carlo Simulation Assumed Structure for Monte Carlo N-Particle extended (MCNPX) Simulation 4 mil CMG Coverglass 2 mil DC Adhesive 11 m IMM (assumed Ge) 5 m Ag Back Metallization 2 mil GE566 Adhesive 2 mil Kapton

Omnidirectional Proton Irradiation (Entire Structure) Bottom-Side

73 Ionizing Energy Deposition In Blanket Monte Carlo (MCNPX) simulation of Tacsat4 Radiation Environment on IMM blanket structure TACSAT4 Omnidirectional Proton Irradiation (Entire Structure) Bottom-Side Radiation Top-Side Radiation 4 mil CMG 2 mil DC m IMM 5 m Ag 2 mil GE566 2 mil Kapton

74 Energy Deposition vs. Depth in Blanket Bottom-Sided Top-Sided Sum Irradiation from bottom Irradiation from top cell

75 Effect of Thin Shielding on Spectrum Irradiation Omnidirectional Irradiation Energy Going to Recoils (kev/um/ion) Circular, 5093 km, 57 o,1 year Orbit No shielding 1 um ZnO 10 um SiO2 3 mils SiO2 (3 mils ~ 75 m) E-02 1.E-01 1.E+00 1.E+01 Depth (um)

Solar Cell Radiation Environment Analysis Models (SCREAM)

Solar Cell Radiation Environment Analysis Models (SCREAM) 2017 Space Environment Engineering & Science Applications Workshop (SEESAW) Scott R. Messenger, Ph.D. Principal Space Survivability Physicist Scott.messenger@ngc.com