Radioluminescent nuclear batteries with different phosphor layers

Nuclear Instruments and Methods in Physics Research B 338 (2014) 112–118

Contents lists available at ScienceDirect

Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb

Radioluminescent nuclear batteries with different phosphor layers Liang Hong, Xiao-Bin Tang ⇑, Zhi-Heng Xu, Yun-Peng Liu, Da Chen Department of Nuclear Science and Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China

a r t i c l e

i n f o

Article history: Received 16 June 2014 Received in revised form 9 August 2014 Accepted 11 August 2014 Available online 7 September 2014 Keywords: GaAs photovoltaic cell Beta radioluminescence nuclear battery Phosphor layer Monte Carlo method Shockley diode equation

a b s t r a c t A radioluminescent nuclear battery based on the beta radioluminescence of phosphors is presented, and which consists of 147Pm radioisotope, phosphor layers, and GaAs photovoltaic cell. ZnS:Cu and Y2O2S:Eu phosphor layers for various thickness were fabricated. To investigate the effect of phosphor layer parameters on the battery, the electrical properties were measured. Results indicate that the optimal thickness ranges for the ZnS:Cu and Y2O2S:Eu phosphor layers are 12 mg cm2 to 14 mg cm2 and 17 mg cm2 to 21 mg cm2, respectively. ZnS:Cu phosphor layer exhibits higher fluorescence efficiency compared with the Y2O2S:Eu phosphor layer. Its spectrum properly matches the spectral response of GaAs photovoltaic cell. As a result, the battery with ZnS:Cu phosphor layer indicates higher energy conversion efficiency than that with Y2O2S:Eu phosphor layer. Additionally, the mechanism of the phosphor layer parameters that influence the output performance of the battery is discussed through the Monte Carlo method and transmissivity test. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction Microelectromechanical systems (MEMS) have undergone great development in recent decades. However, the small size and limited application of MEMS set stringent the requirements for its power supply. Traditional batteries, such as fuel and solar cells, cannot satisfy the power requirements of MEMS devices due to their large volume, short lifetime, and poor environmental adaptability. A nuclear microbattery, which is used to transform the decay energy of radioisotopes into electrical energy, is a potential candidate as an energy supply for MEMS [1]. Among the numerous competing types of nuclear microbatteries, betavoltaic microbatteries have been extensively examined [2–7]. The development of miniature satellites resulted in higher density requirements for the power supply, motivating the exploration of high-energy radioactive isotopes, such as 147Pm and 241Am. However, irradiation damage of the photovoltaic material under high-energy ray is very serious. To address this limitation, the radioluminescent nuclear battery is an awesome choice. A radioluminescent nuclear battery utilizes radioactive decay to produce fluorescence in the phosphor material through the photoelectric effect of photovoltaic cells to generate electric current, as shown in Fig. 1. The development of photovoltaic materials and technology facilitated the use of III–V photovoltaic cells in the radioluminescent nuclear battery since III–V photovoltaic cells have ⇑ Corresponding author. Tel./fax: +86 025 52112906 80407. E-mail address: [email protected] (X.-B. Tang). http://dx.doi.org/10.1016/j.nimb.2014.08.005 0168-583X/Ó 2014 Elsevier B.V. All rights reserved.

many advantages: direct band gap structures, large light absorption coefficient, and small leakage current [8]. Sychov et al. [9] fabricated an indirect-conversion radioisotope battery based on 238 Pu/ZnS/AlGaAs with 21 lW power output. Prelas et al. [10,11] investigated optoelectronic betavoltaic cells based on 85Kr and diamond, and this work indicated that the using of wide bandgap photovoltaics in nuclear energy was a promising application. Walko et al. [12] designed radioluminescent photoelectric power sources based on 3H/ZnS aerogel composite light source and different photovoltaic cells. The study showed that III–V photovoltaic cell design was smaller than hydrogenated amorphous silicon design in the volume of battery. Sims et al. [13] reported the GaP power conversion in ZnS:Ag light source and demonstrated that GaP had 23.54% and 14.59% conversion efficiencies in 968 and 2.85 lW cm–2 blue light, respectively. Hong et al. [14] optimized the parameters of a beta radioluminescence nuclear battery and obtained a 2.5% energy conversion efficiency theoretically. Fluorescence efficiency, transparency and the thickness of phosphor layer will influence the fluorescence intensity of phosphor layer. Moreover, the match between the spectrum of layer and the spectral response of photovoltaic cells will affect the photoelectric efficiency of photovoltaic cells. Therefore, the selection of phosphor layer and optimization of thickness can effectively improve the output performance and energy conversion efficiency. This paper presents a radioluminescent nuclear battery based on 147Pm/phosphor layer/GaAs and reports the best thickness ranges of phosphor layers and output performance of the battery. The influence mechanism of the phosphor layer on energy

L. Hong et al. / Nuclear Instruments and Methods in Physics Research B 338 (2014) 112–118

113

b particles in substances usually range from a few tens of microns to hundreds of microns so that the phosphor layers can easily absorb the kinetic energy of the particles. As a result, the shielding of beta radioluminescent nuclear battery is easy to handle, and the size of the battery can become greatly

miniaturized due to its relatively short ranges. Moreover, the battery will perform more stably since the irradiation damage of the phosphor materials is minimal. 147Pm radioisotope with the specific (per unit area) activity of 5 mCi cm2 was applied to the battery in this study. As two kinds of typical phosphors, ZnS:Cu and Y2O2S:Eu were used to fabricate phosphor layer, and these parameters of which are given in Table 1. There are many ways to prepare phosphor layer: physical settlement method [15], photographic paste method, and electrophoresis method. The photographic paste method is costly, whereas the electrophoresis method produces unstable phosphor layers. The physical settlement method is more advantageous than the two aforementioned methods: easy operation and controllable thickness. In this paper, phosphor was deposited on quartz glass substrate by physical settlement method to prepare the phosphor layer, as shown in Fig. 2(b). The fluorescence intensity of phosphor layer excited by the 147 Pm radioisotope is very weak, and GaAs photovoltaic cell with small leakage current can utilize and collect the weak fluorescence. The radioluminescent nuclear battery comprises 147Pm radioisotope, phosphor layer, and GaAs photovoltaic cell with an area of 5  5 mm2 (Fig. 2(d)).

Table 1 Phosphor parameters.

2.2. Optical and electrical test

147Pm

Load

source

Phosphor layer Fluroescence Photovoltaic cell Fig. 1. Schematic of a radioluminescent nuclear battery.

conversion efficiency of the battery is discussed through the Monte Carlo method and transmissivity test.

2. Experimental 2.1. Battery prototype

Phosphor (chemical composition)

Luminescence color

Grain diameter (lm)

Y2O2S:Eu ZnS:Cu

Red Green

6.8 7.4

Fig. 2. (a)

147

The radioluminescence spectrum and fluorescence efficiency of phosphor layer which are determined by the kind of phosphor, would influence the photoelectric response of the GaAs photovoltaic cell and fluorescence intensity of the phosphor layer, respectively.

Pm source, (b) phosphor layers, (c) GaAs photovoltaic cell, and (d) battery prototype setup.

114

L. Hong et al. / Nuclear Instruments and Methods in Physics Research B 338 (2014) 112–118

The transmissivity of phosphor layer also affect fluorescence intensity. The radioluminescence spectrum were tested with a Cary Eclipse luminescence spectrophotometer (Agilent Technologies, USA) against 147Pm source. The data mode of Cary Eclipse luminescence spectrophotometer was setted as Bio/Chemi-luminescence, and the spectrophotometer obtained the relative intensity of fluorescence with a certain wavelength by photomultiplier detectors to transform photons into electrons and collect the electrons. The test conditions were consistent: the light source (combination of a phosphor layer and 147Pm source) was next to the entrance window of the spectrophotometer, and the emission slit and photomultiplier detector voltage were 20 nm and 800 V, respectively. The transmissivity of layer was measured with an UV–VISNIR spectrophotometer (Shimadzu UV-3600). I–V curves were obtained with a dual-channel system source meter instrument (Model 2636A, Keithley, USA). 3. Results and discussion 3.1. Radioluminescence spectra and transmissivity With 147Pm excitation, the relative radioluminescence spectra of the phosphor layers were presented in Figs. 3 and 4. As seen from these figures, the spectrum of ZnS:Cu is unimodal, whereas that of Y2O2S:Eu is narrow and bimodal. The spectral peak positions of the two phosphor layers are different. The thickness of the phosphor layer affects the fluorescence intensity but not the spectrum shape. Fig. 5 shows the normalized radioluminescence spectra and transmissivities of the phosphor layers. The thicknesses of ZnS:Cu and Y2O2S:Eu phosphor layer were 5.76 and 4.38 mg cm2, respectively. Transmissivity is constant within the radioluminescence spectrum, so the spectrum shape would not be altered. 3.2. Optimization of the phosphor layer parameters Figs. 6 and 7 depict the short-circuit current density (Jsc), opencircuit voltage (Voc), and maximum power output density (Pmax) of the battery vs. phosphor thicknesses. The plot of Jsc with increase phosphor layer thickness, initially increases and subsequently decreases, and which has a maximum at ZnS:Cu phosphor layer thickness of 12–16 mg cm2 and Y2O2S:Eu phosphor layer

Fig. 4. Relative radioluminescence spectra of Y2O2S:Eu phosphor layers under excitation by a 147Pm source.

thickness of 17–21 mg cm2, respectively. The plot of Voc and Pmax vs. phosphor layer thickness are approximately the same. Increasing the phosphor thickness also increased the energy deposition of the b particles in the phosphor layer. This phenomenon resulted in the increase in the generation of fluorescence. However, the absorption of fluorescence would be added at the same time. The energy attenuation plots of the b particles in the phosphor layer simulated by MCNP, and those of fluorescence obtained from the transmissivity of phosphor layers are presented in Fig. 8. The energies of both b particles and fluorescence show an exponential decline with phosphor layer thickness, but the energy attenuation speed of fluorescence is lower than that of the b particles. Thus, the generation is greater than the absorption of fluorescence during the initial increase in thickness. Further increase phosphor layer thickness results in almost no longer increases in the generation of fluorescence but the absorption of fluorescence continues to increase, so fluorescence intensity would decrease. As a result, the fluorescence intensity will initially increases and subsequently decreases with phosphor layer thickness. Figs. 9 and 10 display the relative fluorescence intensity as a function of phosphor layer thickness plots, which were calculated by integrating the relative fluorescence intensity with a certain wavelength with respect to wavelength. Also, the plots of accumulation of deposition energy depends on phosphor layer thickness by MCNP simulation are presented in the Figs. 9 and 10, which attain saturated with increasing layer thickness. Assuming that the layer is perfectly transparent, namely, without self absorption of photons, the plots of relative fluorescence intensity vs. phosphor layer thickness would be similar to that of accumulation of deposition energy with increasing thickness. To compare relative fluorescence intensity and accumulation of deposition energy for various phosphor layer thickness, it is sure that the self absorption of photons is great. According to the Shockley diode equation, the output current density can be calculated as [16]:

    qV 1 ; JðVÞ ¼ J sc  J 0 exp kB T a Fig. 3. Relative radioluminescence spectra of ZnS:Cu phosphor layers under excitation by a 147Pm source.

ð1Þ

where Jsc is the short-circuit current density, J0 the reverse saturation current density, and V the output voltage. Jsc and J0 are expressed as follows [16]:

L. Hong et al. / Nuclear Instruments and Methods in Physics Research B 338 (2014) 112–118

Fig. 5. Radioluminescence spectra and transmissivities of the phosphor layers.

Fig. 6. Electric performances of the battery vs. thicknesses of ZnS:Cu phosphor layer.

Fig. 7. Electric performances of the battery vs. thicknesses of Y2O2S:Eu phosphor layer.

115

116

L. Hong et al. / Nuclear Instruments and Methods in Physics Research B 338 (2014) 112–118

J sc ¼ q

Z

1

bs ðE; T a ÞdE;

ð2Þ

Eg

J0 ¼ q

Z

1

Eg

3 h c2

h

2F a E2   i dE; exp kBET a  1

ð3Þ

where bs(E, Ta) is the fluorescence intensity with photon energy of E, q the elementary charge, Fa the environmental geometrical factor equal to p, Ta the environment temperature, h the Planck’s constant, c the speed of light in vacuum, and kB the Boltzmann constant. Correspondingly, Voc and Pmax are obtained:

V oc ¼

  kB T a J  ln sc þ 1 ; q J0

Pmax ¼ Max½JðVÞ  V:

Fig. 8. Energy attenuation of b particles and fluorescence in the phosphor layers.

ð4Þ ð5Þ

bs(E, Ta) as a function of phosphor layer thickness is similar to the plot of relative fluorescence intensity vs. phosphor layer thickness. Therefore, Jsc, Voc, and Pmax increase first and subsequently decrease with the phosphor thickness according to Eqs. (1)–(5).

Fig. 9. Relative fluorescence intensity and accumulation of deposition energy vs. thicknesses of ZnS:Cu phosphor layer.

Fig. 10. Relative fluorescence intensity and accumulation of deposition energy vs. thicknesses of Y2O2S:Eu phosphor layer.

L. Hong et al. / Nuclear Instruments and Methods in Physics Research B 338 (2014) 112–118

3.3. Energy conversion efficiency The total energy conversion efficiency of the battery (gT) can be determined as:

gT ¼

Pmax ; AEb

ð6Þ

where A is the 147Pm radioisotope activity, and Eb the average beta energy of 62 keV for 147Pm radioisotope. According to the energy transfer process of the battery, gT takes the form

gT ¼ gb  gR  gTR  gPV ;

ð7Þ

where gb is the efficiency of conversion of beta particle energy into energy deposit of phosphor layer, gR the fluorescence efficiency (efficiency of conversion of energy deposit into fluorescence intensity) of phosphor layer, gTR the transport efficiency of fluorescence intensity from the phosphor layer to the surface of the GaAs photovoltaic cell, and gPV the photoelectric efficiency of GaAs photovoltaic cell. The energy attenuation trends of the b particles in the two phosphor layers are same (Fig. 8). Thus, gb is same for the two phosphor layers. The fluorescence efficiencies gR for ZnS:Cu and Y2O2S:Eu phosphor layers are 17.5% and 13% [17], respectively, with similar ratios of 1.35. Fluorescence attenuation speed in the ZnS:Cu phosphor layer is faster than that in the Y2O2S:Eu phosphor layer (Fig. 8). As a result, gTR of ZnS:Cu phosphor layer is less than that of Y2O2S:Eu phosphor layer. The total energy conversion efficiency gT of radioluminescent nuclear batteries with ZnS:Cu and Y2O2S:Eu phosphor layers are shown in Fig. 11. When the thickness of phosphor layer is 14 mg cm2, gT of the battery with ZnS:Cu phosphor layer is twice as large as that with Y2O2S:Eu phosphor layer. Assuming that the efficiencies for every processes of the battery with Y2O2S:Eu phosphor layer are a, b, c and d, these efficiencies of the battery with ZnS:Cu phosphor layer can be obtained on the bases of Eq. (7) and the aforementioned analytical results. The values of these efficiencies are presented in Table 2. The

117

photoelectric efficiency of GaAs photovoltaic cell gPV with ZnS:Cu phosphor layer is 1.48 times greater than that with Y2O2S:Eu phosphor layer, so the fluorescence from the ZnS:Cu phosphor layer is a better match to the GaAs photovoltaic cell. 4. Conclusions Two types of radioluminescent nuclear battery with ZnS:Cu and Y2O2S:Eu phosphor layers were manufactured. The influence of the phosphor layer parameters on battery performance was explored. Experimental results showed that the optimal thickness ranges of ZnS:Cu and Y2O2S:Eu phosphor layers for 147Pm radioisotope are 12 mg cm2 to 16 mg cm2 and 17 mg cm2 to 21 mg cm2, respectively. The output battery performances are codetermined by the fluorescence efficiency, transparency of the phosphor layer, and the match between the fluorescence spectrum and photovoltaic cell. The transparency of Y2O2S:Eu is superior to that of ZnS:Cu, but the fluorescence efficiency of ZnS:Cu is greater than that of Y2O2S:Eu. As a result, the latter is beneficial for fluorescence transmission to the surface of the photovoltaic cell. Additionally, the fluorescence of ZnS:Cu matches the spectral response of the GaAs photovoltaic cell better. Therefore, the battery with ZnS:Cu phosphor layer can provide higher energy conversion efficiency than Y2O2S:Eu phosphor layer. The energy conversion efficiency of the battery is very low due to various factors that restrict the overall efficiency of the battery. The low efficiency may have resulted from the very low activity and impure of 147Pm radioisotope used during the study, which caused the very low value of gb. The 147Pm source in the study was manufactured by powder metallurgic method [18]. The self absorption of photons in the phosphor layer is also an important factor to reduce energy conversion efficiency. Therefore, enhancing the activity and purity of the source, and selecting both high efficiency and transparency phosphor can improve the energy conversion efficiency of the battery. Acknowledgments Supported by the National Natural Science Foundation of China (Grant No. 11205088), the Aeronautical Science Foundation of China (Grant No. 2012ZB52021), the Natural Science Foundation of Jiangsu Province (Grant No. BK20141406), and the Funding of Jiangsu Innovation Program for Graduate Education (Grant No. CXZZ12_0146) and the Fundamental Research Funds for the Central Universities. References

Fig. 11. gT of the battery for various thicknesses with ZnS:Cu and Y2O2S:Eu phosphor layers.

Table 2 Battery efficiency. Phosphor layers

gT

gb

gR

gTR

gPV

Y2O2S:Eu ZnS:Cu

abcd 2abcd

a a

b 1.35b

c
d >1.48d

[1] D.-Y. Qiao, X.-J. Chen, Y. Ren, W.-Z. Yuan, A micro nuclear battery based on SiC Schottky barrier diode, J. Microelectromech. Syst. 20 (3) (2011) 685–690. [2] Y.-P. Liu, X.-B. Tang, Z.-H. Xu, L. Hong, P. Wang, D. Chen, Optimization and temperature effects on sandwich betavoltaic microbattery, Sci. China Technol. Sci. 57 (1) (2014) 14–18, http://dx.doi.org/10.1007/s11431-013-5413-0. [3] X.-B. Tang, Y.-P. Liu, D. Ding, D. Chen, Optimization design of GaN betavoltaic microbattery, Sci. China Technol. Sci. 55 (3) (2012) 659–664, http://dx.doi.org/ 10.1007/s11431-011-4739-8. [4] X.-B. Tang, D. Ding, Y.-P. Liu, D. Chen, Optimization design and analysis of Si–63Ni betavoltaic battery, Sci. China Technol. Sci. 55 (4) (2012) 990–996, http://dx.doi.org/10.1007/s11431-012-4752-6. [5] M. Lu, G.-G. Zhang, K. Fu, G.-H. Yu, D. Su, J.-F. Hu, Gallium Nitride Schottky betavoltaic nuclear batteries, Energy Convers. Manage. 52 (5) (2011) 1955– 1958. [6] D.-Y. Qiao, X.-J. Chen, Y. Ren, B. Zang, W.-Z. Yuan, A nuclear micro-battery based on silicon PIN diode, Acta. Phys. Sin. 60 (2) (2011). 020701(1)– 020701(9). [7] M.A. Prelas, C.L. Weaver, M.L. Watermann, E.D. Lukosi, R.J. Schott, D.A. Wisniewski, A review of nuclear batteries, Prog. Nucl. Energy 75 (2014) 117– 148, http://dx.doi.org/10.1016/j.pnucene.2014.04.007. [8] K.E. Bower, Y.A. Barbanel, Y.G. Shreter, G.W. Bohnert, Nuclear Batteries Based on III–V Semiconductors, Polymers, Phosphors, and Voltaics for Radioisotope

118

[9]

[10]

[11]

[12]

L. Hong et al. / Nuclear Instruments and Methods in Physics Research B 338 (2014) 112–118 Microbatteries, Boca Raton, London, CRC Press, New York, Washington D.C., 2002, p. 306. M. Sychov, A. Kavetsky, G. Yakubova, G. Walter, S. Yousaf, Q. Lin, D. Chan, H. Socarras, K. Bower, Alpha indirect conversion radioisotope power source, Appl. Radiat. Isot. 66 (2) (2008) 173–177, http://dx.doi.org/10.1016/ j.apradiso.2007.09.004. M.A. Prelas, E.J. Charlson, E.M. Charlson, J. Meese, G. Popovici, T. Stacy, Diamond photovoltaic energy conversion, Second International Conference on The Application of Diamond Films and Related Materials, 1993. M.A. Prelas, G. Popovici, S. Khasawinah, J. Sung, in: M.A. Prelas, P. Gielisse, G. Popovici, B.V. Spitsyn, T. Stacy (Eds.), Wide Band-gap Photovoltaics, Springer, Netherlands, 1995. R.J. Walko, C.S. Ashley, C.J. Brinker, S.T. Reed, C.L. Renschler, T.J. Shepodd, R.E. Ellefson, J.T. Gill, L.E. Leonard, Electronic and photonic power applications. The Radioluminescent Lighting Technology Transfer Conference, Albuquerque, NM (USA), 1990, pp. 25–26.

[13] P.E. Sims, L.C. Dinetta, A.M. Barnett, Proceedings of the 13th Space Photovoltaic Research and Technology Conference, Washington, D.C., 1994, p. 373. [14] L. Hong, X.-B. Tang, Z.-H. Xu, Y.-P. Liu, D. Chen, Parameter optimization and experiment verification for a beta radioluminescence nuclear battery, J. Radioanal. Nucl. Chem. (2014) 1–7, http://dx.doi.org/10.1007/s10967-0143271-2. [15] H.-F. Qin, T.-L. Guo, Preparation of tetrapod-shaped ZnO nanomaterial field emission cathodes by deposition method, Acta Phys. Sin. 57 (2) (2008) 1224– 1228, http://dx.doi.org/10.7498/aps.57.1224. [16] J. Nelson, Photons in, electrons out: basic principles of PV, The Physics of Solar Cells, Imperial College Press, London, 2003, pp. 17–39. [17] G.W. Ludwig, J.D. Kingsley, The efficiency of cathode-ray phosphors, J. Electrochem. Soc. 117 (4) (1970) 348–353. [18] D. Raybould, D.G. Morris, G.A. Cooper, A new powder metallurgy method, J. Mater. Sci. 14 (10) (1979) 2523–2526, http://dx.doi.org/10.1007/BF00737047.