Space Power Technological, Economic and Societal Issues in Space Systems Development Volume 7 Number 2 1988 —carfoK— publishing company
SPACE POWER Published under the auspices of the SUNS AT Energy Council EDITOR Andrew Hall Cutler, Space Studies Institute Space Power is an international journal for the presentation, discussion and analysis of advanced concepts, initial treatments and ground-breaking basic research on the technical, economic and societal aspects of large-scale, space-based solar power, space resource utilization, space manufacturing, and other areas related to the development and use of space for the long-term benefit of humanity. Papers should be of general and lasting interest and should be written so as to make them accessible to technically educated professionals who may not have worked in the specific area discussed in the paper. Editorial and opinion pieces of approximately one journal page in length will occasionally be considered if they are well argued and pertinent to the content of the journal. Submissions should represent the original work of the authors and should not have appeared elsewhere in substantially the same form. Proposals for review papers are encouraged and will be considered by the Editor on an individual basis. Editorial Correspondence: Dr Andrew Hall Cutler can be reached by telephone at (619) 284-2779, and his address is 3030 Suncrest No. 214, San Diego, CA 92116, USA. Dr Cutler should be consulted to discuss the appropriateness of a given paper or topic for publication in the journal, or to submit papers to it. Questions and suggestions about editorial policy, scope and criteria should initially be directed to him, although they may be passed on to an Associate Editor. Details concerning the preparation and submission of manuscripts can be found on the inside back cover of each issue. Business correspondence, including orders and remittances to subscriptions, advertisements, back numbers and offprints, should be addressed to the publishers: Carfax Publishing Company, P.O. Box 25, Abingdon, Oxfordshire 0X14 3UE, United Kingdom. The journal is published four times a year, in March, June, September and December. These four issues constitute one volume. An annual index and title-page is bound in the December issue. ISSN 0951-5089 Cover: An artist's concept of the enhanced configuration of the permanently manned Space Station, produced by Rockwell International. The enhanced configuration includes an upper and lower keel for attaching external payloads, a 50 kilowatt solar dynamic system mounted on the ends of the transverse boom, a servicing bay and a co-orbiting platform (not pictured). Reproduced by Courtesy of NASA, Washington, DC, USA. © 1988, SUNSAT Energy Council
SPACE POWER Volume 7 Number 2 1988 Akio Suzuki, Taizo Hirano, Akio Ushirokawa & Makoyo Nagatomo. SPS is the Next Goal of Commercial Solar Cells 131 Takashi Abe. Direct Energy Conversion from a Laser Beam by using a Relativistic Electron Beam 145 V. Poulek. Very Low Temperature Rise Laser Annealing of Radiationdamaged Solar Cells in Orbit 153 Af. Misawa & S. Kondo. Preliminary Design Study of a 1 MWe Space Nuclear Power Plant 157 Arnaldo M. Angelini. On the Possibility of Intercontinental Power Transmission via Satellite 175 Hiroshi Homma, Tatsuo Okamoto, Masao Yamauchi, Chobei Yamabe & Kenji Horii. Direct Energy Converter from Laser Energy to Electricity with Laser-produced Plasma 187 Alain Bossavit. On the Exploitation of Geometrical Symmetry in Structural Computations of Space Power Stations 199 Kiyohiko Itoh & Yasutaka Ogawa. Considerations on an Inland Rectenna 211 Space Power Titles and Abstracts 217
digitized by the Space Studies Institute ssi.org
SPS is the Next Goal of Commercial Solar Cells AKIO SUZUKI, TAIZO HIRANO, AKIO USHIROKAWA & MAKOYO NAGATOMO Summary SPS solar cells are required to be the same low cost as that of mass-produced terrestrial ones. First, we describe the development situation of terrestrial solar cells, and second, the development trends and cost prospects of future terrestrial ones. The SEG single crystal cell is approaching the limit of cost reduction, and the SOG semicrystalline cell will take its place soon. The amorphous Si cell which is now expanding its market for consumer use rather than for terrestrial use has the advantage in cost reduction. Therefore, it will be the most useful product, if its low efficiency and poor reliability - still unsolved problems-are improved. The cost of the SPS solar cell will be rather higher than that of a terrestrial one because of its high performance. Here, we discuss the factors of high price and the cost prospects of the SPS cell. Introduction One of the main problems of a solar power satellite (SPS) which converts solar energy into electrical power with solar cells is the high cost of solar cells for present satellite systems. It is expected that mass production of them would reduce the cost, but the reduction of the cost of space-use solar cells is not given a high priority of research effort compared to increasing conversion efficiency and life. The high cost of space-use solar cells is accepted because the electrical power system is a subsystem of an expensive space system carried by expensive space transportation. Considering that the SPS is possible only when a much cheaper transportation cost is realized and solar cells are the dominant cost in a total SPS system, cost reduction should be the central issue for technical research and development for solar cells. The annual production of terrrestrial-use solar cells is growing as high as 10 MW in Japan, and the cost of solar cells is becoming as low as ¥1000 in round numbers per watt of direct output of electrical power of solar cells. The cost is doubled for electrical power system made of solar cells. The current cost target is as low as ¥100 per watt, which is a similar order of magnitude as the requirement for SPS. The reduction in cost would increase the need, and eventually mass production will reduce the cost again. Thus the terrestrial-use solar cells are expected to be competitive with other electrical sources in this century. The high quantity of solar cells will be pursued even after that. For example, current research and development of amorphous Akio Suzuki & Taizo Hirano, SHARP Corporation, Tokyo, Japan, and Akio Ushirokawa & Makoto Nagatomo, Institute of Space and Astronautical Sciences, Megoro-ku, Komaba 4-6-1, Tokyo 153, Japan.
solar cells is aiming at larger cell area, higher efficiency, higher production speed and substrate technology in the field of amorphous solar cells. The results of the current development will not necessarily meet the SPS requirement, but fundamental technology will be applied for the improvement of the performance of solar cells for SPS. Thus the next goal of research and development of terrestrial-use solar cells should be the high-performance requirement of SPS solar cells. In this paper, the main properties of terrestrial-use solar cells will be examined from the standpoint of their applicability to the space environment, and the approach to higher-performance solar cells towards an extension of terrestrial-use solar cell development will be discussed. 1. Forecast of Development of the Terrestrial-Use Solar Cell In the 1970s, the main uses of solar cells were for satellites in the USA, and for independent power sources in remote areas in Japan. And the solar cell was thought of as a special electrical part which was used by particularly limited specialists. For example, terrestrial-use cells were installed in about 2,000 places in Japan in those days as telemeter power sources; less than 10 W per system was a typical example, and even the biggest solar system for a lighthouse had only 1 KW output. The cost was about ¥30,000 per watt and it was still very expensive. Nevertheless the solar system had enough merits by comparison to other power sources, such as the commercial power source which had to erect utility poles and run wires from a commercial power station, or the heavy batteries which had to be changed at intervals, or the diesel dynamo which required maintenance. Single Crystal Solar Cells With the SEG (semiconductor grade Si) single crystal wafer process used for large- scale integration (LSI), integrated circuit (IC) technology has been developed as a mass production process as the highest refined technology. The wafers used for solar cells do not require such high quality as those for LSI use, in the respect of physical characteristics and configuration accuracy, and so low-cost wafers can be supplied in accordance with the demand for LSI use if we use wafers out of specification or standard, or we miss out some processes. In Japan, 16.5% (AMI.5) efficiency solar cells fabricated from single crystal wafers have recently been sold in the market at a price of over ¥1,000 per watt, and the market size is about 4 MW per year. The demand for wafers for solar cells will increase rapidly, but cost reduction cannot be expected only from the additional mass production effect if we persist in the present high quality for single crystal wafers. The final cost of single crystal solar cells will be about ¥800 per watt, although the figure depends on the production quality. Solar Grade Si Materials and Substrate Manufacturing Technology The final limit of the cost reduction of a single crystal solar cell depends on the Si raw material and the substrate manufacturing technology. Generally there are two methods of material purification. One is to get polycrystalline Si by reduction through halogenation of Si, and the other is direct synthesis of SOG polycrystal by means of some purification process from metallurgical Si or SiO2 not through halogenation. In Japan,
a unique method has been developed. That is the granule polycrystalline Si synthesis by continuous recycling of chlorosilicate compounds using a Fluid Floor Reactor [1]. There are two methods of manufacturing single or semicrystalline Si substrate. One is the slice method in which wafers are cut from a lump of Si made by CZ or other methods, and the other is the sheet method in which Si wafers are directly formed to their cell thickness from molten material. The latter can simplify the substrate process but it brings much material loss. The former, slice method, holds a dominant position in many respects because of good mass productivity and good material performance. Especially, for the casting method, the representative slice method is being investigated as an effective way. The solar cells that are made from casting a semicrystalline wafer of SOG material, are developed to attain almost 13% conversion efficiency in production, and the cost of unit power is becoming the same as that for a single crystal wafer. These kinds of solar cell are expected to cost in the order of ¥500 per watt. Junction Formation Process The junction of solar cells is formed by the wet process method which uses chemicals analogous to conventional semiconductor devices, or by the dry process method which uses high vacuum technology. At the existing technical level, it is still hard to choose one method over the other in respect of costs, mass production and evaluated performance of solar cells. Packing In the terrestrial-use solar cell modules, the super-straight structures using glass boards are developed and commonly used instead of the boxed type structure used in the early days. In the near future will appear the ultrathin laminating structure, or lightweight mould type structure. Amorphous Silicon Solar Cells Amorphous Silicon (a-Si) has the physical characteristics of ‘very high solar absorption' differing from single crystal Si, and theoretically needs only 1 pm thickness to operate as a solar cell. In addition it has convenient properties of no grain growth, compared to other crystalline materials such as thin film. Nowadays, the efficiency of an a-Si solar cell comes up to 10% by improving deposition technique, and this is further increasing by 1% per year. On the other hand, a-Si solar cells tend to degrade by high-intensity irradiation such as direct sunlight, so that there is some restriction on outdoor use. But if these problems are settled, it is expected to be the leading technology of economic terrestrial-use solar cells. The a-Si solar cells have another characteristic of high spectral response at short wavelength. As indoor illumination is almost entirely by fluorescent lamp in Japan, a lot of a-Si solar cells are manufactured on a large scale for indoor personal microelectronics use, such as calculators, and the total production of a-Si solar cells is rather larger than that of crystalline-type ones. Table I shows the present state and future target of the terrestrial-use solar cell, and Tables II and III show the development of crystalline and amorphous solar cells.
Table I. Present State and Future Target of Terrestrial-use Solar Cell.
Table II. Development of Crystalline Solar Cell. 2. Cost of Solar Cell The cost of terrestrial-use solar cells has been cut to less than l/2Oth of the initial cost in the early days in Japan, following the subsequent rise in demand and the development of manufacturing techniques. The cost has reached about ¥1,000 per watt. Production capacity Fig. 1 is increasing to about 10 MW per year, that is 1,000 times that of the early days of production. Fig. 1 shows the actual and projected cost learning curve for terrestrial solar cells. The curve is drawn on the expectation for future development according to the guidelines of the ‘Sunshine Project'. The symbol © of Fig. 1 shows the results for solar cells made from single crystal in the Japanese market, and the symbol @ shows the predicted cost learning curve of the
Table III. Development of Amorphous Solar Cell.
TABLE IV. Characteristics of Mass Production Terrestrial Cell and SPS Cell.
crystalline cell developed several years ago. The market price has risen slightly higher than expected. In the near future, it is expected that the semicrystalline cell will be substituted for the single crystal cell and after that, the amorphous Si cell will take the place of the semicrystalline cell. In that case, it is the premise that the reliability problem - one of the weak points of the a-Si cell - should be settled and an improvement in conversion efficiency should be achieved almost equal to that of crystalline-type solar cells. Around the year 2000, the production capacity of solar cells will be more than 10 GW world wide and commercial solar cells will be coexistent with the crystalline-type cells with high efficiency and a-Si cells with lower cost. In all cases, it is supposed that a-Si cells will be the market leaders and then the cost of crystalline and a-Si cells will be several hundreds of yen per watt and under ¥100 per watt, respectively. Cost Analysis of Terrestrial Use Solar Cells The relationship between cost and production quantity of semicrystalline terrestrialuse solar cells is shown in Fig. 2 as relating to two factors, substrate materials and others. The cost of the substrate Si materials occupies the larger part. In the case of small production scales, it is especially notable. This is the reason why the repayment rate on plant investment for producing SOG Si raw materials is very effective. On the other hand, taking the case of amorphous solar cells shown in Fig. 3, the rate of material costs is small and the other costs are mostly the depreciation of equipment and the operational costs in the deposition process. As mass production grows in scale, the material cost rate of amorphous Si cells increases, and will ultimately become about half of all a-Si cell costs.
3. Solar Cells for SPS The cost of SPS solar cells is required to be of the same order as that of the terrestrialuse cell target in the Sunshine Project. However, the target performance of terrestrialuse solar cells is not always satisfactory for an SPS programme and it is necessary to improve their performance at many points. For example, SPS solar cells have to have as high a conversion efficiency as 17.5% at AMO and with lower degradation in the space environment, with phenomena such as particle irradiation. Besides lighter weight, superior strength in mechanical properties and compactness in the cargo are required. Table IV shows the expected performance of terrestrial-use solar cells at a production scale of 10 GW per year, that of SPS solar cells, and the additional improvements in the terrestrial use solar cell for SPS. Investigation of the Production Process We now follow the production process of a typical solar cell according to the flow chart in Fig. 4. Source Material Preparation Process (Si Halide, Polyciystalline Si). The process for synthesizing for metallurgical Si by reduction from silica is used widely in the steel and non-ferrous industry. Si Halide and Polycrystalline Si. Although the process for direct production of SOG Si is investigated, without the intermediate step of Si halide, it is not easy to abandon the process through Si halide, in view of the purity of silicon which is essential for obtaining a higher conversion efficiency. In the Japanese Sunshine Project, the NEDO
method (The Fluid Floor Reactor Method) has been developed in place of the conventional Siemens method, and it has succeeded in long-term operation in the demonstration plant that attempts recycling of the by-product. Si Substrate. For current space-use solar cells, the highly purified single crystal Si wafer is applied, but for SPS solar cells SOG semicrystalline Si wafer will be able to be applied. However, the solar cell made from semicrystalline Si has generally poor radiation resistance because of its numerous crystal defects, so that additional development is necessary to improve its electrical performance. A larger cell (for example, about 10 cm square) will be used, and it is necessary to reduce the cell thickness (for example, less than 100 /zm) in order to reduce the weight and the radiation degradation of space-use solar cells. Thus it is necessary to reduce the manufacturing yield in the case of large and thin semicrystalline Si wafer with larger internal stress. Junction Forming Process, Contact Forming Process and Wiring Technique. It is necessary that the depth of P-N Junction is as shallow as 0.2 /zm uniformly in order to improve the shorter wavelength response of AMO sunlight. The shallow junction needs to stretch fine grid lines around the active area of the cell, and the screen printing method of contact formation (used for present terrestrial-use cells) should be improved to get a finer pattern. The history of advancing contact formation and the interconnection of solar cells is the reliability improvements in the solar cell and module itself. The contact of the space-use solar cell which requires excellent reliability has the composite structure of Ti/Pd/Ag by the vacuum evaporation method, in order to provide resistance against humidity during storage and testing on
the ground, and against great stress during launch and mission life in space. Terrestrialuse solar cells have the contacts formed by mass production low-cost screen patterning, because it is packaged in a strong humidity-proof structure, although it is exposed to severe outdoor weather. Although the SPS cell which aims for economic advantage should basically apply the process used for terrestrial cells, new developments in materials and process technology are likely to be necessary, because the usual process for terrestrial-use cells brings disadvantages in electrical output, reliability and mechanical characteristics. Radiation Shield The present space-use solar cell uses the expensive cover glass for a radiation shield. For the large area of an SPS array, it is necessary to develop new materials for the radiation shield which are cheaper and processed more quickly. As one idea, the solar array may be covered with a transparent resin film which has superior radiation resistance. Application of Amorphous Si Cells for SPS Amorphous Si cells deposited on glass substrates are effective, because the substrates act as a shield from irradiation. However, these cannot be used if the following problems are not solved. The problems of amorphous Si cells are the low efficiency mentioned above and degradation by sunlight. The effective thickness of amorphous Si cells is very thin, less than 1 /zm, so it is thought that amorphous Si cells have good radiation resistance because of their small absorption. But we have not yet achieved good results in our preliminary investigation by comparison to normal Si cells. It is thought that a-Si has smaller activation energy than crystalline Si, sufficient to confuse the atomic structure, like photon degradation. It is also thought that a lightweight array for space will be realized only when a-Si cells with lower degradation against irradiation are developed. There is a report that 1,620 watt per kg in the power-weight ratio of a-Si cells deposited on a flexible blanket (for example, Kapton) has been realized. This is five times more than the 300 watts per kg of one made of crystalline Si [2], When a-Si is suitable for space use and a large-area sheet array is required, it will become unnecessary to carry solar cells fabricated on the ground into space, because they will be able to be made in space where a high vacuum is readily available [3]. 4. Cost Forecast of Solar Cells for SPS From the above, the calculation of the expected costs of SPS solar cells at production capacity of 10 GW is shown in Fig. 5, compared with that for terrestrial-use solar cells. In this Figure the cost is shown according to each process for material, substrate and cell fabrication. As for a-Si, the cost of the cell substrate is not contained, because array substrate is used as the cell substrate at the same time. Crystalline solar cells for SPS are considered to increase costs because it is necessary to improve the conversion efficiency in comparison to terrestrial-use cells. A crystalline solar requires, for example, high purification of source material, thinning of Si substrate, a shallower junction and fine pattern contact formation, and development of the contact material to improve its reliability and resistance against the space environment, and of radiation shielding materials. A-Si cannot be free from the cost-raising factors of a new venture,
such as improvement in material purification, the cell structure, the cell fabrication processes and the introduction of protective materials against space radiation in order to obtain higher output, reliability and a light weight. Cost Reduction of SPS Cells The scale of SPS is thought to be such that its output covers from 10 MW to 10 GW per system. One example shows that the fabrication capability of SPS is two systems per year, and at maximum 60 systems in total, according to feasibility studies in the USA. As the production numbers of SPS directly influence SPS costs, they must be assumed at first. But in this paper it is assumed that terrestrial-use solar cells are manufactured through a mass production process against the background of large demand, and that SPS cells with special requirements are produced according to a terrestrial process. From the above, the cost efficiency of mass production for SPS cells is ignored in some measure. In the case of a single crystal solar cell in Fig. 5, the SPS cell cost is expected to be ¥400 per watt, twice as much as ¥200 per watt for a terrestrial-use solar cell, and in the case of amorphous Si solar cell, the SPS cell cost is expected to be about ¥250 per watt by comparison to ¥80 per watt for terrestrial use. Conclusion It is necessary that the SPS solar cell costs as little as a terrestrial cell in large-scale manufacturing, has high power and high reliability nearly the same as that of a current space-use solar cell. Although we cannot expect a low cost for SPS cells, in so far as
the mass production process for space-use cells is used, it is best to apply the fabrication technology for low-cost terrestrial-use cells and to improve on their performance to space cell level. It is expected that the technology of terrestrial-use solar cells will be promoted to the cost level of several hundreds of yen per watt for amorphous Si cell in a production scale of 10 GW per year by 2000. The cost of the SPS cell will be from ¥250 to ¥400 per watt as the terrestrial-use cell adds the necessary performance suitable for space use. Solar cell development for SPS, coupled with development for low-cost terrestrialuse cells, is one target of technical development of mass production for commercial cells. REFERENCES [1] Hongu, T. (1985) Research promises future breakthroughs in solar energy, Business Japan, 30, p. 45. [2] Ovshinsky, Stanford R. & Yang, Jeffrey (1985) A figure of merit evaluation of amorphous silicon alloy solar cells, Proceedings of the 1985 International Conference on Solar and Wind Energy Applications, p. 75. [3] Ushirokawa, A. & Suzuki, A. (1982) Manufacturing experiment of the large area thin film silicon solar cell for SPS in space, Proceedings of The Space Station Symposium, 1, p. 313.
digitized by the Space Studies Institute ssi.org
Direct Energy Conversion from a Laser Beam by using a Relativistic Electron Beam TAKASHI ABE Summary In space activity, the energy remotely transmitted by means of a laser beam is valuable for many purposes. This energy can be used as electric power through an energy converter. In this paper, a new energy converter is proposed, in which a relativistic electron beam is used. In this method, the operation of high power density is possible while maintaining a high efficiency, since the unconverted laser beam is transmitted rather than absorbed in the converter. 1. Introduction A remotely powered vehicle is most promising for future space activity [1], The laser beam is a favourable candidate for the remote transmission of power generated at a central power station, because it diverges less during transmission. In a remotely powered vehicle, a laser beam can be mainly used in two ways: (1) it can be used as power for a propulsion system; and (2) it can be used as electric power through an energy converter. In this paper, we consider the latter. To be practical, a laser converter must (1) be properly matched to the laser wavelength, (2) exhibit high energy conversion efficiency, (3) operate at high power density, (4) have a high ratio of peak power to system weight, (5) operate reliably, and (6) not be excessively expensive to manufacture. Of these necessary converter characteristics, the most important are high efficiency and high power density operation. A high conversion efficiency and the capability to operate at a high power density are closely linked. The need for good conversion efficiency is obvious. However, what happens to the fraction of energy that is not converted, especially at high power density, is crucial. If this power is converted to heat, high temperatures result which may reduce efficiency and require major cooling systems. Converters that transmit unconverted power rather than absorb it could have significant advantages in thermal management, efficiency, stability, and reliability. From this perspective we propose a method in which the relativistic electron beam (REB) is used in order to convert the laser energy to electric power. 2. Interaction Between Laser and REB We consider a laser beam which propagates in parallel to REB (see Fig. 1). We have analysed earlier a phenomenon which occurs in this situation, and here we first review Takashi Abe, Institute of Space and Astronautical Sciences, Meguro-ku, Komaba 4-6-1, Tokyo 153, Japan.
the earlier results [2]. As a result of non-linear interaction between the laser beam and the REB, the energy of the laser beam is converted in two ways-(l) the acceleration of REB, and (2) a scattered wave with lesser frequency. The scattered wave is determined through the following dispersion relation; where w (w0) and k (&0) are a frequency and wave number of the scattered (incident) electromagnetic wave, wpe ( = (4^we2/»z)1/2) the plasma frequency of the REB, (y^CTtZoeVtnV)1/2) the cyclotron frequency defined by the magnetic field of the incident laser beam, ve the thermal velocity of the REB and y is defined by 1/(1- ©Vc2)1/2. Here n is the density of the REB, e the unit charge, m the mass of electron, Io the energy flux of the incident laser beam, v the velocity of the REB and c the speed of light. The dispersion relation (1) is derived under the following condition; where kD is defined by Wp/ve. This condition is satisfied for the REB having a satisfactorily small temperature (i.e. small velocity spread). When the scattered wave derived from Eq. (1) is unstable (i.e. an imaginary part of the complex frequency (a, (=Im(a>y) is positive), the growth of the unstable scattered wave and the acceleration of REB can take place. Fig. 2 shows the frequency of the scattered wave for typical parameters. The frequency of the scattered wave is lesser than that of the incident laser beam, and becomes smaller for greater y. The acceleration of the REB is caused by a plasma wave which is induced in the REB as a result of a nonlinear interaction between the laser and REB. The upper limit of the fraction of energy which is converted to the REB acceleration is
corresponding to the phase velocity of the plasma wave in the REB and W, is the energy density of the incident laser beam. In large 7 limit, it becomes where P=v/c. On the other hand, the fraction of energy which is converted to the scattered wave is Here IF, is the energy density of the scattered wave. In large 7 limit, T], becomes The conversion efficiencies r)b and t]s are depicted in Figs. 3 for typical parameters. Since gives an upper limit of the efficiency, it cannot be greater than 1. Hence the following quantity is depicted instead of t]b, where Max (,) represents the greatest value among the arguments. Both efficiencies increase with 7 value and the density of the REB. In general, ys is much smaller than lbThe acceleration of the REB and the generation of the scattered wave are caused as an unstable phenomenon. This phenomenon appears in a timescale of Hence the phenomenon appears spatially in a scale of This interaction length L is depicted in Fig. 4. The length L increases with the larger value of 7, the weaker intensity of the incident laser beam and the lesser REB density.
Absolute values of the parameters which appears in Figs. 2-4 can be determined, when the incident laser beam is chosen. Now we consider the CO2 laser which is a most promising power laser, and has a wavelength of 10.6 /zm. In Table I, the density of the REB, and the intensity of the laser beam are tabulated. These are found to be attainable.
3. Conversion System from Laser to Electric Power In the previous section, it was shown that the energy of the incident laser beam is converted to REB kinetic energy and the scattered wave with lesser frequency (process I). To complete the conversion system from laser to electric power, this energy must
be converted to electric power (process II). Hence, a conversion system from laser to electric power is composed of these processees. Here we consider the method in process II. First, we consider the scattered wave. The scattered wave has a sufficiently lesser frequency if the density of the REB is small and the y value is large enough. It is well known that, in the microwave region, the electromagnetic wave can be rectified with good efficiency [3]. Hence, through the rectification, the power of the electromagnetic wave can be converted to DC electric power. If an appropriate value of the REB density and the y value is chosen so that the scattered wave is in a microwave region, this energy can be converted to DC electric power through the rectification. The fraction of energy converted to the acceleration of the REB is much greater than that to the scattered wave. Hence the conversion of the kinetic energy of the REB increased through the interaction with the laser beam is more important. We propose a method in which a microwave is generated intermediately, and the resultant microwave is converted to DC power. To generate a micro wave, an idea of the free electron laser can be applied [5]. That is, the accelerated REB is introduced to the wiggler field (the rippled magnetic field). Through the interaction with the rippled magnetic field, the REB generates the electromagnetic wave expending its kinetic energy. The efficiency of generating the electromagnetic wave increases with the intensity of the rippled magnetic field. The generated electromagnetic wave has a frequency; where kf is a wave number 2^/A defined by the spacing A of the rippled magnetic field. For an appropriate spacing of the rippled magnetic field and an appropriate y value, the generated electromagnetic field falls in a microwave region. The efficiency of the free electron laser is not so high. Hence the remaining REB must be reused to save its energy. In this closed system, a steady state operation is inevitable. To maintain the steady state of the REB, the intensity of the rippled magnetic field must be controlled so that only the excess energy of the REB is converted to the microwave. Fig. 5 shows a schematic drawing of the system. The REB which is accelerated through the interaction with the incident laser beam (region I) is turned by 180°, and passes through the rippled magnetic field in which the microwave is generated (region II). After that, it is turned by 180° and, again, passes through the region interacting with the incident laser. At regions I and II, the microwave is generated which can be
rectified and converted to DC electric power. The incident laser beam goes out after interacting with the REB. Hence any fraction of the laser beam is not absorbed in the system and is not reduced to a heat to be rejected. 4. Discussion We consider an optimal parameters to design the system depicted in Fig. 5. When the electrons are turned at the region between regions I and II, they lose their energy by producing the radiation of the electromagnetic field. Considering an electron turning in a uniform magnetic field, the energy loss due to a synchrotron radiation is [4], Here B is the intensity of the uniform magnetic field. For larger value of y, the energy loss due to the synchrotron radiation becomes large while the conversion efficiencies r/b and i]s become large. When we consider the turning by a radius of 1 m, the rate for the REB to lose its kinetic energy is The frequency of the synchrotron radiation is where is a frequency in a microwave region. It means that y must be as small as possible in order to attain a smaller 2. Since the smaller 2 is necessary for a compactness of the system, it suggests that y must be chosen to be as small as possible. As for a frequency of the scattered wave, it becomes smaller and reaches the microwave region for a larger value of y. However the request for a larger value of y is contrary to other factors; the reduction of the synchrotron radiatin loss and the interaction length L (i.e., the compactness of the system). Hence we must give up the idea of using the scattered wave and pay attention to using only the kinetic energy of the REB, since the fraction of energy converted from the laser beam to the scattered wave is much smaller in comparison to the REB kinetic energy. Within this perspective, we must consider (1) the reduction of synchrotron radiation loss, (2) the compactness of the system and (3) an appropriate efficiency. Requirements (1) and (2) suggest that the y value must be selected to be as small as possible, within the constraint of attaining an appropriate conversion efficiency. For example, if the parameters <yAf=27tX 1010 (sec-1) and y=2 are chosen, we obtain 2 = 24 cm, cor=2^x 1.67 x 108 (sec-1) and Ir/mc1y=\5 x 10-6 (sec-1). A high power density operation becomes possible for a greater density of the REB. When the density of the REB becomes large, the stability of the REB ring may become a critical problem. In general, the cooperation phenomenon is expected to occur in
parallel to the increase in REB density, and it may break the stability of the REB ring. The study relating to this phenomenon remains to be conducted. In parallel to the REB acceleration, the thermalization of the REB (i.e., the spread of the energy spectrum of the REB) is induced [2]. This thermalization of the REB may break condition (2) which guarantees the acceleration of the REB through the nonlinear interaction. Furthermore it may reduce the efficiency of the free electron laser [5]. It is an important factor when the efficiency is considered for the energy conversion system in a closed form. 5. Conclusion A new energy converter from a laser energy to DC power is proposed. In this system, the acceleration of the REB through the interaction with the incident laser is utilized. The increment of the kinetic energy of the REB is converted to DC power, generating a microwave intermediately. Compactness and an appropriate efficiency can be accomplished by choosing appropriate parameters. Furthermore, the operation of high power density keeping a high efficiency is possible, since unconverted laser energy is transmitted rather than absorbed in the converter. REFERENCES [1] DeYoung, R.J. et al. (1983). NASA SP-464. [2] Abe, Takashi (1986) Physics of Fluids, 29, p. 3394. [3] Gutmann, R.J. & Borrego, J.M. (1979) IEEE Transactions, M.T.T. 27, p. 958. [4] Landau, L.D. & Lifshitz, E.M. (1975) The Classical Theory of Fields, Pergamon Press. [5] Kroll, N.M. & McMillin, W.A. (1978) Physical Review A, 17, p. 300.
Very Low Temperature Rise Laser Annealing of Radiation-damaged Solar Cells in Orbit V. POULEK Summary Solar cells of all space objects are damaged by radiation in orbit. This damage, however, can be removed by laser annealing. A new in-orbit laser regeneration system for both body- and spin-stabilized space objects is proposed. An in-orbit laser regeneration system was originally proposed only for spin-stabilized geostationary communication satellites. The results of the research on very low temperature rise laser annealing show that a laser regeneration system could be used also in body-stabilized space objects such as an international space station or geostationary communication satellites. For successful annealing of solar cells damaged by 10 years' radiation dose in orbit it is necessary for the temperature rise in the incidence point of the laser beam to reach about 400°C. By continuous regeneration, however, between two annealing cycles the solar cells are hit by about two orders of magnitude lower radiation dose. This makes it possible to carry out the regeneration at a temperature rise well under 1°C! If an optimal laser regeneration system is used, such low temperature rise laser annealing of radiation-damaged solar cells is possible. A semiconductor GaAlAs diode laser with output power up to 10 mW CW was used for annealing. Some results of the very low temperature rise annealing experiment are given in this paper. Introduction In the second half of the 1970s research work was begun with the aim of regenerating radiation-damaged solar cells by means of a laser beam. In the course of research work it has been found that solar arrays can be regenerated by a laser beam which heats up the cell to a temperature of about 400°C. There are some problems, however, caused by these high temperatures. For example, a very important problem of the in-orbit laser regeneration of radiation-damaged solar arrays is to ensure that the repeated laser annealing will neither damage the solar cells nor the complete solar panel [1]. It is also difficult to find a suitable laser for in-orbit application. Experiments with the very low temperature rise laser annealing of radiationdamaged solar cells were begun with the aim of overcoming these problems. Low Temperature Rise Laser Annealing Experiment Continuous annealing has been selected for the in-orbit laser regeneration system because between two annealing cycles the solar cells are hit by a low radiation dose. V. Poulek, Czechoslovak Academy of Sciences, Institute of Physics, 180 40 Prague, Czechoslovakia.
Only point defects are created in the silicon lattice due to a low radiation dose. It is possible to regenerate these defects at very low temperatures. By higher radiation doses the point defects tend to create defect clusters, dislocation loops or even dislocation loop networks. These defects are much more stable than the point defects [2]. A temperature rise of about 400°C is necessary to regenerate such more complex defects. In the first experiments a low radiation dose of 1.1010 cm-2 2MeV protons was applied to boron doped n+/p silicon photovoltaic cells. The equivalent 1 MeV electron dose is about 5.1013 cm-2, i.e. a half year dose in Clarke orbit. Radiation fluence was under 1.106 cm-2 s-1 to avoid the thermal effect. A semiconductor GaAlAs diode laser with output power up to 10 mW CW was used for annealing. This laser emits on the 830 nm wavelength. A laser beam with a diameter of 100 fim was scanned over the whole area of the 4 X 4 mm2 samples with a velocity up to 2 cm/s. From the above data it can easily be calculated that the temperature rise in the centre of the laser beam was well under 1°C. In these experiments mostly only short circuit current was investigated under AM 1.5 conditions. Short circuit current of the radiation-damaged samples was compared before and after laser annealing. It was found out that in optimum annealing conditions photovoltaic cells regain about 50% of the short circuit current which was lost by radiation damage (Fig. 1). From the results it is evident that in this experiment laser annealing is a nonthermal phenomenon. In recent reports injection-enhanced annealing has been described, which enables nonthermal annealing of the radiation damage in InP and even in silicon. Yamaguchi [3] reports that forward bias and photon-induced injection annealing is possible in InP. According to Barnes [4] forward bias injection annealing was possible also in radiation-damaged silicon solar cells. Results mentioned in this paper seem to be due to photoinjection annealing. Laser Regeneration System Important progress in the field of GaAlAs semiconductor diode laser arrays is reported by Spectra Diode Laboratories Inc. Laser arrays with output power of 5.4 W CW and 11 W quasi-CW were realized at SDL [5]. In the near future GaAlAs diode laser
arrays with output power of 10-15 W CW in a diffraction limited output beam can be expected on the market. Also, the efficiency of these devices is very enhanced. At present electrical to optical efficiency of 39% is reported by SDL [6]. Semiconductor diode laser arrays were first proposed for an in-orbit regeneration system five years ago [7]. Due to the results of the research of the very low temperature rise laser annealing and progress in the field of very high power diode laser arrays, an in-orbit laser regeneration system can be substantially improved. It is now proposed not only for spin-stabilized satellites but also for body-stabilized space objects. The scheme of the in-orbit laser regeneration system for body-stabilized spacecraft with flexible or semi-rigid retractable solar arrays is presented in Fig. 2. Solar arrays are irradiated during deployment from an array of diode lasers. The final beam is as wide as the solar array. In this method, scanning of the laser beam is very simple. A laser regeneration system for spin-stabilized satellites was proposed three years ago. The scheme of the modified system is presented in Fig. 3. The regeneration system can work quite independently of the power and control system of the satellite. No change is necessary in satellite construction. Conclusion Very low temperature rise laser annealing for radiation-damaged photovoltaic cells has been demonstrated for the first time by use of the GaAlAs diode laser. It has been found that low dose radiation damage can be annealed without thermal effects. Photon-induced recombination enhanced injection annealing could be the reason. Due to the very low temperature rise no laser-induced damage of the solar cells nor damage to the complete solar panel is possible. Further progress in this field is possible, e.g. still lower radiation doses could be used and/or higher laser power could be applied. A scheme for an in-orbit laser regeneration system for both body- and spin-stabilized spacecraft is proposed in this paper. The main advantages of the very low temperature rise laser annealing of radiation-damaged solar cells in orbit are the following:
• no laser-induced damage of solar cells, nor damage to the complete solar panel; • 50% regeneration of the solar cells; • no change in satellite construction; • a quite independent regeneration system. Acknowledgement I am grateful to Dr. I. Wilhelm of the Nuclear Centre of Charles University, Prague, for proton implantation of all photovoltaic cells in this experiment. REFERENCES [1] Poulek, V. (1984) Proceedings of the 4th European Symposium ‘Photovoltaic Generators in Space', ESA SP-210, pp. 143-148. [2] Narayan, J. & Holland, O.W. (1984) Journal of Applied Physics, 56, pp. 2913-2921. [3] Yamaguchi, M., Ando, K.., Yamamoto, A. & Uemura, C. (1985) Journal of Applied Physics, 58, pp. 568-574. [4] Barnes, C.E. & Samara, G.A. (1986) Applied Physics Letters, 48, pp. 934-936. [5] Harnagel, G., Welch, D., Cross, P. & Scifres, D. (1986) Lasers and Applications, June 1986, pp. 135-138. [6] Thornton, R.L., Burnham, R.D., Paoli, T.L., Holonyak, N. & Deppe, D.G. (1985) Applied Physics Letters, 48, pp. 7-9. [7] Poulek, V. (1983) Paper IAF-83-69, Acta Astronautica, 11, pp. 697-700.
Preliminary Design Study of a 1 MWe Space Nuclear Power Plant M. MISAWA & S. KONDO Summary This paper describes the results of a preliminary design study performed on a 1 MWe heat pipe based space nuclear power system capable of being launched in the Space Shuttle. Two basic power conversion technologies were considered - thermionic conversion and the Rankine cycle with potassium working fluid. Design variables were optimized to minimize systems mass under given thermal and dimensional constraints. Turbine inlet temperature was the most important constraint for the Rankine cycle based system, while maximum allowable fuel temperature was the most significant constraint for thermionic conversion. The specific masses for the Rankine and thermionic systems were 9.7 and 5.6 kg/kW, respectively. This preliminary study indicates the feasibility of a 1 MWe nuclear power system for shuttle launch and suggests that thermionic conversion is preferable to the Rankine cycle. Introduction Electrical power demands for various space facilities will increase as they become more advanced [1-3]. For power generation in excess of 100 kWe, nuclear systems are believed to be more economical than solar power systems [16]. Furthermore, nuclear energy is the only realistic power source for deep space missions. Space nuclear power programs are presently underway in the United States, [4-7, 16], The United States is pursuing the SP-100 program as well as others in order to develop, demonstrate and make available space reactor systems capable of providing up to a megawatt of electrical power. Thermoelectric conversion powered by a lithium cooled, uranium nitride fuelled fast reactor was selected in August 1985 for the SP-100. The program is currently in an engineering development phase involving ground testing in a simulated space environment. There are no Japanese plans to develop space nuclear power due to the lower power demands in the current Japanese space program. If Japanese space development involving international cooperation proceeds, there will be a need for high power technologies in the late 1990s and beyond. Thus it is not too early for Japan to investigate the feasibility of megawatt scale space nuclear power sources for future missions. M. Misawa, Department of Nuclear Engineering, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan; and S. Kondo, Nuclear Engineering Research Laboratory, Faculty of Engineering, University of Tokyo, 2-22 Shirane Sirakata, Tokai-mura, Naka-gun, Ibaraki-ken 319-11, Japan. Note: This article was rewritten for style by the Editor of Space Power (Andrew Cutler). Any defects of style, as well as any errors or mis-statements which have crept into this detailed and insightful paper are the fault of the Editor.
Two energy conversion technologies were considered in the present study, and size and mass constraints were chosen to make the reactor system suitable for space shuttle launch. During the optimization process, certain essential subsystems and design constraints were found to effect system mass strongly and these are pointed out to clarify technology efforts required in the next design phase. There are three major sections to this paper-one presenting overall design requirements and top level requirements for the design study; the second describing the basic concepts and subsystem design procedures for a space nuclear powerplant; and the third presents the result of the optimization process and discusses the values chosen for the independent variables and design constraints. Overall Design Considerations Top level requirements for this design study are summarized in Table I. In addition, the following requirements were addressed in the design: • guaranteed safety during transport and launch; • guaranteed safety at end of life; • core disintegration in case of inadvertent re-entry; • containment of fission product gases in the system; • assembly completed on the ground; • requires only a single shuttle launch; • not rendered useless by any single point failure. Fission product gases are retained in the core in this design to ensure environmental acceptability regardless of where in space the reactor is used. The system is fully assembled on the ground and launched in a single shuttle flight. A boron plug was centred in the core to ensure the reactor does not accidentally achieve criticality before normal operation begins. At the end of the reactor's life it will be separated from the associated payload and boosted to an orbit with a 300-year life-long enough for radiation to decay to an acceptable level before re-entry. One megawatt design output was selected because Japanese space development in the 21st Century will require hundreds of kilowatts of baseload power for applications such as experimental facilities, manned space stations, and electric propulsion systems. In particular, nuclear-electric propulsion is one of the most promising propulsion technologies for deep space missions where nuclear power is the only realistic energy
System Description Basic Concepts. The principles on which a space nuclear power reactor operates are the same as those which govern terrestrial installations. The system consists of a reactor, heat transport system, electrical generating system, and a heat rejection system. Electrical generation and heat rejection are carried out somewhat differently than they are on Earth. In space, heat can only be rejected by radiation - and the fourth power dependence of rejected heat on rejection temperature encourages the use of a substantially higher temperature for the low temperature end of the generating cycle than would be considered on Earth. The generally high systems temperature due to this opens the possibility of using any of a variety of conversion cycles-such as the Brayton, Rankine or Stirling cycle or of using thermoelectric or thermionic converters. The Rankine cycle and thermionic conversion were selected as reference schemes after a comparison of the options above. The Rankine cycle system is composed of 9 subsystems: a reactor; core heat pipes (Li/TZM); neutron and gamma shields; 4 potassium Rankine loops (each containing a boiler, condenser, turbine and electromagnetic pump); radiator heat pipes (Na/SUS); radiator armor and fins; reactor controls; power conditioner; and support structure. A schematic of the Rankine cycle system is shown in Fig. 1(a). The reactor core is composed of hundreds of fuel elements and a beryllium reflector. The core is cooled by lithium heat pipes. The condenser sections of the core heat pipes boil the potassium and superheat the vapors to 1500 K. Shell and tube heat exchangers were used in this design and a two stage impulse turbine was selected due to its simplicity. Four power conversion loops each generating 250 kWe were used to prevent single point failure. The boiler and condenser are common to the four energy conversion loops. Sodium heat pipes are used in the radiator, which is armored with enough beryllium to be micrometeoroid resistant. The high temperature loop must resist liquid lithium and potassium at the operating temperature. Tantalum, molybdenum, columbium and tungsten have excellent corrosion resistance under these conditions, so their alloys are the materials of choice for the Rankine loop.
RkJQdWJsaXNoZXIy MTU5NjU0Mg==