Prospects of Microwave Heating in Silicon Solar Cell Fabrication A Review

Pages 11
Views 3
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
IOSR Journal of Electrical and Electronics Engineering (IOSR-JEEE) e-issn: ,p-ISSN: , Volume 6, Issue 3 (May. - Jun. 2013), PP Prospects of Microwave Heating in Silicon Solar Cell
IOSR Journal of Electrical and Electronics Engineering (IOSR-JEEE) e-issn: ,p-ISSN: , Volume 6, Issue 3 (May. - Jun. 2013), PP Prospects of Microwave Heating in Silicon Solar Cell Fabrication A Review S. D. Das CEGESS, Bengal Engineering and Science University, India Abstract : Solar energy is only renewable energy source that does not depend on earth's resources. Current industrial manufacturing of solar cells use manufacturing steps that have many sustainable issues. So it becomes very important to promote technologies that are more sustainable than current state of the art. This paper will look into prospects of microwave heating for silicon solar cell fabrication in this perspective. Microwave heating is very different from conventional heating techniques and application of microwave for semiconductor processing purposes has many advantages. It provides avenue to crystallisation of amorphous solids, reduction of defects, selective heating, reduction in processing time, sintering of powdered materials and so on. Sustainability analysis of microwave heating over conventional heating will be presented to judge its suitability in solar cell fabrication. Status of different technologies using microwave heating which could replace or assist in different fabrication steps of solar cells will be reviewed. Based on these reviews, possible fabrication schemes of silicon solar cells will be presented. Keywords - microwave heating, solar cells, silicon, nanomaterials, sustainability I. Introduction Microwave radiation is part of electromagnetic spectrum with frequencies ranging from 300 MHz to 300 GHz (Fig. 1). Dielectric heating is heating of material using high frequency electromagnetic radiation and microwave heating is subcategory of dielectric heating at frequencies above 300 MHz. Use of frequency in the RF range (1-300 MHz) for heating purpose is not uncommon. This process of heating with electromagnetic radiation is also known as electromagnetic induction heating [1]. Most often materials are heated with in a cavity where electromagnetic waves are launched from a small dimension source such as a magnetron. These are referred to as microwave ovens. The first commercial microwave oven was available as early as 1947 from Raytheon. Microwave oven finds its most popular use in domestic application for heating of food materials. There are many industrial applications of microwave heating as well, such as melting, smelting, sintering, drying, joining, annealing [2-13] and many more. Some of the applications rely on indirect heating through susceptors [14]. Microwave heating without the use of cavity is also possible and one such example is asphalt pavement maintenance, where a heating wall is used instead of a cavity [15]. Microwave heating could be adopted for semiconductor processing as well. Microwave annealing in semiconductor processing is a potential alternative to conventional rapid thermal annealing (RTP). Microwave annealing can deliver much higher power in less time than RTP (Fig. 2) [16, 17] and this can lead to extraordinarily high ramping rates of 2000 o C/second in some materials [18]. There are many other applications of microwave heating in semiconductor processes such as non-contact heating and selective heating, defect elimination in semiconductors, ultra fast annealing, and recrystallization of amorphous semiconductors [19]. Solar cells based on microwave heating have recently been introduced by Herman et. al. [20], where microwave heating has been used to produce nanoparticles and thin film solar cells were then produced using microwave annealing procedure. In this review an attempt is made to discuss possible applications of microwave heating for different fabrication steps of silicon solar cells. Discussion on sustainability issues of microwave heating will also be presented to justify its suitability in solar cell fabrication. Also, part of the discussion will focus on nanomaterial fabrication using microwave heating, which could be used for novel third generation solar cell designs such as plasmonic based solar cells. Based on the reviews, technically feasible fabrication schemes of silicon solar cells will be presented. 28 Page MICROWAVE FREQUENCY (Hz) Figure 1. Electromagnetic spectrum showing microwave range. II. Microwave Heating Microwave heating is very different from conventional heating. In conventional heating, thermal energy is transferred into bulk of material from a heat source though conduction or convection via surface of the material to be heated. However, in microwave heating microwave radiation generates heat in the material bulk causing temperature to rise inside the material and subsequently thermal energy travels outside via the material surface. Thus conventional heating relies on heat transfer where as microwave heating relies on heat generation in the bulk, which could cause rapid rise of temperature in the material. With the rapid increase in temperature materials could soften and subsequently could melt. Once radiation is stopped cooling begins. During microwave heating some parts of sample may be heated to higher temperatures as compared to other parts of the material, this is selective heating. Selective heating happens primarily because of spatial distribution or anisotropy of material property such as dielectric function. There are various physical mechanisms of microwave heating. Mechanisms of microwave heating could be classified broadly into two different categories, dielectric polarization [16, 21] and ionic conduction [6]. Polarization mechanisms depends on the creation of dipoles or permanent dipoles already present in a material. On application of electromagnetic radiation the fields exert force on the dipoles on atomic scale. Since the fields oscillates the dipoles have to oscillate with them. But atomic and molecular bonds of the material oppose this oscillating force, which creates intermolecular friction and the dipole is unable to keep up with the phase changes of the fields. This frictional loss generates heat. Polarization mechanisms include electronic polarization, atomic polarization, dipolar polarization and space charge polarization. Electronic polarizations is polarization due to displacement of nucleus with an atom. Atomic core is too heavy to respond to the oscillations. Only the surrounding negative charge of electrons tend to oscillate. Electronic polarization occurs at very high frequencies, near ultra violet range. Atomic polarization is polarization due to displacement of an atom with in a molecule. Only infrared frequencies can excite this mode. Dipolar polarization is polarization due to permanent dipoles present in a molecule. Millimeter and centimeter frequencies can excite this mode and this is the fundamental mechanism that governs the microwave dielectric heating. Space charge polarization is polarization due to presence of free charges. The free charges (free electrons) on encountering some boundary (such as interface between two materials) creates local electrically polarized region. Dielectric heating is negligible for electronic and atomic polarizations as time scale is too long at microwave frequencies. The other mechanism of microwave heating is ionic conduction which generates heat due to drift of charge carriers in presence of electric field against electrical resistance. Dominating mechanisms of heating inside a material is thus dependent on material and frequency of operation. The most common frequencies in use lie in Industrial, Scientific and Medical frequency band out of which use of 2.45 GHz is most common. However, other frequencies such as GHz and frequencies in the range of 0.9 to 18 GHz are also in use for microwave heating purposes [22]. For semiconductor materials with low complex dielectric constant, the absorption of microwave energy is low while with high complex dielectric constant absorption of microwave energy is high. In case of highly conductive materials such as metals, microwave energy is only absorbed within a skin depth from surface and rest of the energy is reflected from the surface. For this later case, eddy current is the significant cause of heating [19]. In highly doped semiconductors selective heating could be achieved with microwave heating where microwave will only heat the highly doped regions. In 2009, Tian et. al. have shown specific layers could be targeted for heating in semiconductor hetero-structures with high and low doping layers using microwave heating [23]. 29 Page Incident Power (W/cm 2 ) Prospects of Microwave Heating in Silicon Solar Cell Fabrication A Review J/cm Time (sec) Figure 2. Comparison of microwave annealing and conventional rapid thermal annealing [16]. III. Microwave Heating Sustainability One of the main advantage of microwave heating is that it is more sustainable than conventional heating. Sustainable development can be defined as a social development which fulfills the needs of present generation without endangering the possibilities of fulfillment of the needs of future generation [24]. From this definition a technology should be sustainable when it is economically viable, socially acceptable and ecologically viable for long run. Microwave oven which can raise temperatures up to thousand or couple of thousand degree Celsius is cheaper than conventional furnace systems as system requirements such as insulation, ceramic/ quartz tubes and other expensive parts are not required. Also, microwave heating is much more efficient than conventional furnace as heating is directly produced inside the material which prevents loss of energy due to heating of ambient. In conventional furnace there is significant loss of energy due to heating of ambient and extra insulation material may be required to achieve high temperatures, which increases the cost of the conventional furnace system. From the perspective of ecological systems sustainable technology should use renewable resources, processes should be efficient and emissions or disposed materials from the technology should not harm the ecology in anyway. Interpretation of these boundary condition for different technologies will be different for different places. Thus sustainability of a technology is evaluated with respect to its local environment [25]. Fig. 3 demonstrates the interaction of ecosphere and technosphere for microwave heating process, in general. Conventional heating system and microwave heating system differs in their process of conversion of electricity to heat and corresponding instrumentation. Definitions of renewable and non-renewable resources can be adopted from Dewulf et. al. work [26]. Microwave oven consist of four parts, namely, the source (magnetron), transmission waveguide, applicator and the electronic control unit for source and applicator. The magnetron source, applicator and transmission waveguide all are primarily made of metal while controlling unit consists of electronic grade semiconductor materials. These materials can readily be recycled. However, some microwave ovens may use ceramic insulators like beryllium oxide which may be harmful to health. Other sources of non-renewable materials are paints, wiring insulation, and non-degradable plastics. The single cycle gaseous waste depends upon the material being heated and long term wastes harmful to ecology are the nonrenewable materials. However, some of these same non-renewable materials may also be used in conventional furnace along with more harmful materials such as glass wool (as insulator) in some systems, which are also hazardous to health. Thus microwave oven is more sustainable than conventional furnace system by virtue of cost benefits, being more efficient and less non-renewable material use. Further, microwave processing itself is expected to achieve good combination with distributed renewable energy such as solar cell, wind energy and small hydro systems as microwave processing could be typically characterized by short time, small scale and distributed type processes [27]. 30 Page Ecosphere Resources Non-renewable Resource Renewable Resource Microwave radiation Conversion to heat Electricity Materials Instrumentation Sample Purging gas Single cycle waste Long term wastes Heat Gases Technoshpere Ecosphere Figure 3. Interaction of ecosphere and technosphere of microwave heating process. IV. Microwave Heating in Silicon Solar Cell Fabrication Crystalline silicon solar cells are most widely used commercially available solar cell whose market share is largest among all forms of solar cells. A crystalline silicon solar cell is basically a p-n junction diode operated in the 4 th quadrant so that power could be extracted from it. Commercially available cells typically have n on p configuration where n-region is known as emitter and p-region is known as base. A simple schematic of silicon solar cell is shown in the Fig. 4. Commercial crystalline silicon solar cells are manufactured using five basic processing steps: 1. Saw damage removal and texturing 2. p-n Junction formation 3. Anti-reflection coating and surface passivisation 4. Electrical contact formation and annealing of contacts 5. Edge Isolation After production of solar cells, solar panels are made according to wattage requirement of the applications. 4.1 p-n Junction formation The emitter region is formed by doping a p-type silicon wafer with n-type impurity in n on p configuration. A saw damaged removed and textured p-type silicon wafer is doped with n-type dopant using diffusion process in diffusion furnace/ LPCVD system. This process is followed by phosphorous glass removal (when phosphorous is used as n-type dopant and POCl 3 is used as source of phosphorous) which completes the p-n junction formation process. Two different types of emitter doping profile, deep emitter and moderately doped or a shallow emitter with high surface concentration could be used. Both these profiles with good quality surface passivization can lead to reduced surface recombination and hence increased emitter collection efficiency [28]. However, to get good quality ohmic contacts heavily doped emitter regions and base regions are a requirement. Thus the standard solar cell configuration becomes n+ p p+. The heavily doped emitter and base (n+ & p+) regions can cause recombination losses and hence needs to be minimized. The front contact recombination losses due to heavily doped emitter region is minimized by keeping the emitter thickness thinner than the diffusion length of minority carrier, while a built-in electric field is created at the back surface of the solar cell or the base. The back surface field helps in minority carrier transport helping the largest number of minority carriers cross the junction quickly so that recombination losses are kept at minimum. In this case only recombination losses at front outer surface accounts for loss of photogenerated and injected minority carriers. These recombination losses are severe for wavelengths less than 500 nm as heavily doped silicon's absorption coefficient is very high at these wavelengths. If the doping emitter profile is deep and surface recombination is high, the open-circuit voltage (V oc ) and short circuit current density (J sc ) are both affected severely [29], which intern affects the solar cell efficiency. Very shallow junctions (~100 nm) with high doping densities ~10 20 cm -3 are very difficult to make using diffusion methods because conventional annealing steps that follows causes emitter profile to change due to stimulation of defect formation which intern causes lateral diffusion of dopants [30]. For ion implantation method high temperature annealing step is absolutely necessary for the removal of damages and activation of dopants. But with high temperature annealing, diffusion of dopants is still a concern [31]. Quality of junction determines the efficiency of a solar cell and quality of junction depends on the dopant, dopant source, doping and annealing techniques. Typical industry standard POCl 3 junctions are of nm 31 Page deep and have cm -3 surface dopant levels. Rohtagi et. al. [32] have reported 19% cell efficiency with phosphorous implanted and diffused emitters. front contact emitter base back contact Figure 4. Simple diagram of a common monocrystalline silicon solar cell. Junction formation can also be carried out using microwave heating by recrystallization of ion implanted amorphous silicon. Microwave recrystallization of amorphous silicon can be done with and without the use of susceptors. Vemuri et. al. [33] have recently shown that 40 seconds of susceptor assisted microwave annealing can completely recrystallize amorphous silicon doped with ion implantation technique with 1300 W magnetron and 2.45 GHz frequency. The complete recrystallization was supported by Raman Spectra results which showed peak at ~520 cm -1 for all annealing times greater than 40 seconds. Complete activation of dopants was also shown with increase in annealing time. Lowest sheet resistivity of 81 /sq. was obtained for 180 kev, As + cm -2 and 100s annealing time. Further, diffusion for arsenic dopants have been shown to be less than conventional RTA. However, a band of defect is observed where the amorphous to crystalline interface lies with in the as-implanted structure. This was attributed to migration and precipitation of vacancies at interface. A similar study was done by Thomson et. al. [1], with single mode microwave oven which was operated at 3000W at 2.45 GHz. The dopant species in this case was boron (p-type for silicon). The higher power requirement is due to non susceptor assisted annealing. The results showed sheet resistance 300 /sq is achievable. A low ~6.5 nm shift in emitter profile was observed due to diffusion of boron. Lower value of implanted boron sheet resistance ( 100 /sq) with annealing time of 600s is possible and this was shown by Alford et. al. [34]. Alford et. al. achieved these results for susceptor assisted case and attributed the low activation of boron to lattice damage and reverse activation. Fong et. al. [35] achieved 81% crystallization of as-deposited amorphous silicon of 40 nm thick (without implantation) on glass substrate at a temperature of 550 o C with 1000W elliptical applicator (2.45 GHz). This result was obtained using SiC susceptor and for annealing time of 600s. The grain size of the resulted poly-crystalline structure was ~200 nm. It was observed better crystallization could be obtained with higher microwave power and higher temperature of annealing. A comprehensive study on junction formation was carried out by J Borland et. al. [36] in 2011 with implanted phosphorous and boron in silicon and under various annealing condition and techniques. Simulated results of this study showed greater than 20% efficiency is achievable using peak doping concentration of ~ cm -3 and with junction depth of nm (Fig. 5). In deeper junctions surface activation level is limited by dopant concentration due to increase in surface lifetimes which could result in 100% dopant activation. Thus deeper junctions would be preferable for higher solar cell efficiency [37]. Experimental results showed 5 minutes microwave annealing can achieve lower sheet resistance values ( 100 /sq.) as compared to higher temperature 750 o C (90min) and 850 o C (60min) with conventional furnace anneals at the greater than cm -2 phosphorous implant doses and at as-implanted junction depths of nm. For this study microwave annealing at 500 o C was carried out. Boron activation concerns using microwave annealing have also been highlighted in this study. M-H Tsai et. al. [38] have proposed two step microwave annealing method for activation of implanted boron. However, this method only yielded 436 / sq. 32 Page Figure 5. Solar cell efficiency as a function of emitter peak doping level for different junction depths (simulated results adopted from J Borland
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!