Read The correlation of proton and neutron damage in photovoltaics text version

THE CORRELATION OF PROTON AND NEUTRON DAMAGE IN PHOTOVOLTAICS

Scott R. Messenger, Edward A Burke, Justin Lorentzen SFA, Inc., Largo, MO 20774, USA Robert J. Walters, Jeffrey H.Warner, Geoffrey P. Summers US. Naval Research Laboratory, Washington, DC 20375, USA Susan L. Murray Emcore Corporation, Albuquerque, NM 87123, USA Christopher S. Murray Sandia National Laboratories, Albuquerque, NM 87123, USA Christopher J. Crowley. Nabil A. Elkouh Creare Incorporated, Hanover, NH 03755, USA

ABSTRACT

Damage correlations resulting from neutron and proton irradiation are described using the nonionizing energy loss (NIEL) approach. The method is applied on data generated on single junction GaAs/Ge solar cells. INTRODUCTION As part of a NASA-funded project to incorporate InGaAs-based TPV cells with a general purpose heat source (GPHS) to power deepspace missions, proton and neutron irradiation studies were undertaken relevant to this space radiation environment. The source in the GPHS is composed of PUOZ111. While the heat produced from the GPHS can be harnessed by the TPV cells to produce power to the spacecg5, the neutron radiation spectrum emitted by isotopic Pu in the GPHS is expected to produce displacement damage in the TPV device. This paper will provide a methodology to correlate radiation damage due to a neutron radiation spectrum using protons, which are often much more convenient to use for radiation testing. The simulation of neutron damage effects using protons can be advantageous for many reasons. Aside from the availability and high cost of procuring neutron beam time, the more significant problem of neutron-induced activation can lead to high postradiation radioactivity levels, which can limit measurement capabilities. This study aims to provide informationon how to use protons to simulate the effects due to a neutron environment. Particle correlation is based on nonionizing energy loss (NIEL) which relates damage to the number of atomic displacements produced [ . ] 23. A schematic representation of the GPHS configuration of interest is shown in Fig. 1. TPV cells will be mounted adjacent to the top and bottom sides of two stacked GPHS units. Figure 2 shows the neutron

#-

Fig. 1. Schematic of the general purpose heat source

(GPHS) [l]. figure shows three GPHS units. The TFW This

cells will be located adjacent to the top and bottom surfaces of two units stacked as above. radiation spectra produced by the GPHS of interest, as well as several from other fission neutron sources. In Fig. 2, the following neutron energy distributions from four fission sources are plotted: 1) the PUOZspectrum from GPHS Q1 [I], 235Uspectrum produced from White 2) the Sands Missile Range free.field spectra at 24" from the source (WSMR FF 24"). 3) the spectrum produced by the 252Cfsource at the National Institute of Standards and Technology (NIST) and 4) the spectrum produced by the 235Usource at the Sandia SPR II facility. The data plotted are normalized to one source neutron. That is, the integral over neutron energy for each curve is unity. We wish to determine the TPV degradation produced by such a neutron spectrum over the prototypical mission lifetime (14 years). However, using such high energy neutrons for device testing in this experiment will produce several highly radioactive isotopes. As these TPV devices

0-7803-87074/05/%20.00 02005 IEEE.

559

are grown on InP substrates, the prevalent neutroninduced isotope will be the 114mln metastable state, having a half-life of 49.5 days. As most facilities cannot handle radioactive materials, and those that can have limitations, it is expected that these devices will have to cool down for many half-lives until they can be handled for measurement. Therefore, it is desirable to be able to predict the radiation degradation of these devices using a non-activating radiation particle. Protons seem to be a convenient and viable option for such studies.

1.o

the appropriate NlEL value. Characteristic radiation cuwes have been generated for several solar cell technologies. The computer code 'SAVANT" developed jointly by both NASA and NRL implements the displacement damage dose approach for several solar cell technologies [SI.

10' I

1 1 I I

1 I

Td lOeV, Oa b As

GaAs

I

4

1

NI

E

2

0.9

--

Eleciron

5

e

t

0.7

0.6

p

f

= e

n

0.5

0 0.4

E 0.3 m

1v

104

109

p

0

0.2

io' 10.1 100 io' Partlcle Energy (MeV)

io*

10'

$ 0.1 * 0.0

Fig. 3. Electron, neutron, and proton nonionizing energy loss (NIEL) calculations for GaAs.

10.1

104

100

10'

Neutron Energy (MeV)

I .o

4

Fig. 2. Neutron density energy spectrum emiced from several neutron fission sources, normalized to one neutron.

DISPIACEMENT DAMAGE CORRELATIONS

0.9

GaAslGe

d 0.y

8

Proton irradiations can be made to simulate the same level of displacement damage as neutrons by having knowledge of NIEL, which is the rate at which atomic displacements are produced i a material [2,3]. n Analytical NlEL calculations have been performed for a number of particles and materials [2,3]. Neutron NlEL can also be calculated using neutron displacement kerma [4]. Fig. 3 shows results of NlEL calculations for electrons, protons and neutrons incident upon GaAs. These data can be used to correlate the effects of different particles on device performance [5].For example, the NlEL values for 1 MeV electrons, neutrons and rotons in GaAs are 2.66x706, P 6.00~1 and 5.33~10 MeVcm'lg, respectively. This 04, means that, in GaAs,1 MeV protons are about 2000 times more damaging than 1 MeV electrons and about 90 times more damaging than 1 MeV neutrons. Fig. 4 shows results for maximum power (1 sun, AMO, 25OC illumination conditions) degradation in GaAslGe solar cells as a function of displacement damage dose (Dd) under electron and proton irradiation [5,6.7]. Dd is a product of the particle In fluence and NlEL f5J. Fig. 4, one can see that this resulting damage curve is independent of the irradiating particle. This means that this 'characteristic" damage curve could have been obtained from using any of these particles. Therefore, the most convenient particle, usually protons in the several MeV range, can be used. After this characteristic curve is generated, the effect from any other particle, or spectrum of particles, can be calculated using

:::I

OA1 0.0

Protons

Electrons

\10"

1 Sun. AM0 T=2!ioC

io'

1

io*

10'0

h

0.2MeV

10'2

Displacement Damage Dose (MeWg)

Fig. 4. Maximum power degradation (1 sun, AMD, 25°C illumination conditions) of GaAslGe solar Celts plotted as a function of displacement damage dose. It is seen that all of the data for electrons and protons can be correlated using the NlEL and displacement damage dose concepts.

1 MeV Neutron Damage Equivalence

The analysis of determining the f MeV equivalent neutron fluence, as described in the ASTM Standard 722-94 [4], will be summarized here. An equivalent monoenergetic neutron fluence, @equiv,Eref,mat, characterizes an incident energy-fluence spectrum, @(E), in terms of the fluence of monoenergetic neutrons at it specific energy, Eref, required to produce the same displacement damage from neutrons in a specific irradiated material, mat, as @(E). @quiv,Eraf,mat is given as fotlows:

560

los

IO*

loq1 loi2

lo1'

10"

toi5

Particle Fluence (unm2)

Fig. 5.GaAslGe solar cell maximum power degadation as a result of proton, electron and neutron irradiations. The data are plotted as a function of particle fluence It can be seen how different the above radiations act on the cell degradation.

1.o

L

a

0.9

0.8

L

E a E

0.7

0.6

0.5

Table t: 1 MeV neutron equivalent fluences of the various neutron spectra given in Fig. 2. The results are given in terms of one 1 MeV neutronlcm'. For example, the total neutron fluence from the WSMR FF 24" spectrum is equivalent to 78.9% of a given 1 MeV neutron Ruence.

g

0.4

+

A

1MeV

'

2.4 MeV

9

Neutrons

1 MeVequlv.

OP1 0.0

3 08

L

IO' lOl0 IO"

1012

09YeV

DisplacementDamage Dose (MeVlg)

GaAslGe Solar Cell Appllcation

Fig, 6. Same as Fig. 4 with the inclusion of the neutron radiation results. The neutron data using the SPR-II spectrum was reduced to an equivalent 1 MeV neutron fluence and then converted to D d using the 1 MeV neutron NlEL value. All of the data for electrons, protons, and neutrons are well correlated using the NlEL concept.

GaAslGe solar cells were irradiated with neutruns from the SPR-I1 fast burst reactor at Sandia National Laboratory. The cell data were taken from Refs. [9.10]. The neutron spectrum was reduced to an equivalent 1 MeV neutron fluence as described above and the results are plotted in Fig. 5, along with electron and proton fluence data published elsewhere [6,7j. Figure 5 clearly shows how the different irradiating particles produce very different degradation characteristics upon a given device parameter. The data are then transformed into displacement damage dose using the NlEL concept [5], with the results shown in Fig. 6. figure 6 is the same as Fig. 4 with the inclusion of the neutron data. Excellent particle correlation is achieved.

C0NCLUSIONS

Using the concept of nonionizing energy loss (NIEL), neutron and proton radiation damage in GaAs sofar cells have been correlated. The application of a neutron radiation energy spectrum, such as that found in most nuclear reactors including that from a GPHS, has been described in detail. In this application, the overall radiation effect of the neutton energy spectrum is reduced to an I equivalent ' MeV neutron fluence. NlEL can then be used in damage correlation. This correfation is expected to simplify the radiation testing procedure as one can use protons in the MeV energy range to simufate the effects of

neutrons.

561

J.,',.

where @(E)is the incident neutron energyfluence spectral distribution. FD,,&(E) is the neutron displacement damage function for the irradiated material as a function of energy (i.e. the displacement kerma discussed above), and FD,Eref,mpt is the disptacement damage reference value designated for the irradiated material and for the specified equivafent energy, Eref, which is given in the appendices of [4]. Eref is typically chosen to be 1 MeV. The energy limits on the integral in (1) above are determined in practice by the incident energy spectrum and by the material being irradiated. The 1 MeV equivaient displacement damage reference values for Si and GaAs, (i.e. FD,IM ~ V . Sand F0,, ~ are 95 and 70 MeVmb. respectively. (The conversion of displacement kerma to NlEL is accomplished by a changeof units from MeVmb to MeVcm'lg. For Si and G a s . the 1 MeV neutron NlEL values are 2 . 0 4 ~ 1 0 ~ and 5.83~10~MeVcm'lg, respectively.) Applying (1) to the radiation spectra contained in Fig. 2, we arrive at the equivalent 1 MeV equivalent fluences in Si and GaAs which these spectra are equivalent to. The results of these calculations are given in Table 1. As an example, a total fluence of I d 2 nlcm' using the WSMR FF 24" spectra would be equivalent to a fluence of 7.89~10'' n/cmZ of monoenergetic 1 MeV neutrons.

0.0

'

'

' ..''..I ' ' "'''i

' ' ,.,..I

''

j....''

' ' ,'..'-I

''

."ld

' '

' ''>'

IO'*

I D '

REFERENCES

[l] M.E. Anderson, "Neutron Flux, Spectrum, and Dose Equivalent Measurements for a 4500-W(th) 238Pu02 General Purpose Heat Source", Mound Laboratory Monsanfo Report MLM-3248, May, 9, 1985.

121 G.P. Summers, E.A. Burke, P. Shapiro, S.R. Messenger and R.J. Walters, "Damage Correlations in Semiconductors Exposed to Gamma. Electron, and Proton Radiations", IEEE Trans. Nucl. Sci., 40(6). 1993, pp. 13729. [3] S.R. Messenger, E.A. Burke, M.A. Xapsos, G.P. Summers, R.J. Walters, I. Jun and T. Jordan, 'NIEL for Heavy tons: An Analytical Approach", I Trans. Nucl. Sci., 50(6), 2003, pp. 1919-23. [4] 'Standard Practice for Characterizing Neutron Energy Fluence Spectra in Terms of an Equivalent Monoenergetic Neutron Fluence for Radiation-Hardness Testing of Electronics", ASTM International Standard E722-94 (Reapproved- ZOOZ), American Society for Testing Materials.

[5] S.R. Messenger, G.P. Summers, E.A. Burke, R.J. Walters and M.A. Xapsos, "Modelling Solar Cdl Degradation in Space: A Comparison of the NRL Displacement Damage Dose and JPL Equivalent F!uence Approaches", Pmgress in Photovoltaics: Research and Applications, 9, 2001, pp. 103-121.

[ I 3.E. Anspaugh, "Proton and Electron Damage S Coefficients for GaAslGe Solar ceils", Proc. 22nd IEEE Photovoltaic Specialist Conference, 1991, pp. 1593-1598.

[i'l

Anspaugh,

B.E. GaAs Solar Cell Radiation Handbook,

JPL Publication 96-9, 1996.

[8] To obtain a beta version of the SAVANT program, please contact the Space Environments Effects Office at NASA Marshall Space Flight Center (http://see.msfc.nasa.govl)

[9] R.J. Walters, G.P. Summers, S.R. Messenger and E.A. Burke. 'Displacement Damage Dose Analysis of Neutron Irradiated Single and Dual Junction GaAs-based Solar Cells", Proc. 16th Space Photovoltaic Research and Technology Conf., NASA Glenn Research Center, Cleve land, OH, USA, 1999, pp. 3943.

[TO] R. A. Flock, Final Report for AF Wright Patterson Contract# F33615-88-C-2815, January 1991.

562

Information

The correlation of proton and neutron damage in photovoltaics

4 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

673354


You might also be interested in

BETA
Radiation Health Protection Manual
Microsoft Word - Ion-Neutron_Equivalence--IBL_and_LANSCE_by_E._Bielejec,_et_al..doc
ASTM Complete Index