ARTICLE

A REVISIT ON : GENERATION OF ELECTRICITY BY HARVESTING SUNLIGHT

D. GHOSH1,3, K. BISWAS1,3, S. BALAJI1,3, PARTHA PRATIM DAS,2,3 P. SUJATHA DEVI2,3 AND K. ANNAPURNA1,3

In the present day knowledge based society, Science transforms the Culture. This can be witnessed through the contemporary rationalized society that aroused with the advancement of science. Such revolutionised change leading to very fast modern-day walks of life can essentially be credited to the progression in light based technologies. However, in order to meet the continuously mounting demand for energy in this developed world, one has to look for the renewable energy resources and among all such resources most abundant one is the sunlight. In this context, an attempt has been made here to report a brief review on the development of solar cell, the simplest device that converts sunlight to electricity. Further, this article also presents the contributions of CSIR-CGCRI in this field.

Introduction manufacturing industries, retail market, geology, atmospheric physics and optical communications, to name part from being one of the indispensible a few. The development of optical fibres enabled faster component of nature for existence of life on earth, and easier communications leading to instantaneous data the light, has widely been explored by mankind A transfer within a fraction of second through internet to meet with his basic needs in every stages of connections across the world, thereby connecting people development. Light based technologies began to develop irrespective of their geographical locations. Thus in the with the invention of incandescent bulb to meet with the present era, light based technologies have become vital for requirement of vision during night. This then went through almost all human activities. And it is due to the immense several stages of modification over centuries to ultimately societal benefits of these technologies that Nobel prizes evolve as the energy efficient compact fluorescent lamps have been awarded several times to encourage researchers (CFLs), high intensity discharge lamps and solid state in this sector. lighting devices like Light Emitting Diodes (LEDs) and Organic LEDs (OLED) that serve to lit away the darkness However, as we all know, to every pro there is a con. in the present era. In addition to lightning applications, In order to meet with the energy needs of these developing the discovery of lasers in 1960s brought about a revolution technologies, today mankind is exposed to one of the in the world of technology due to its immense applications greatest threat of 21st century – the energy crisis. But here in various sectors which include medical science, also, the ray of hope to overcome such crisis is the sunlight, which on the other hand is the most abundant renewable 1 Glass Division, CSIR-Central Glass and Ceramic Research Institute, energy source, and the most common device for the purpose Kolkata - 700 032, India. is – solar cell. When sunlight of suitable energy is incident 2Sensor and Actuator Division, CSIR-Central Glass and Ceramic Research Institute, Kolkata - 700 032, India. on a solar cell which is basically a p-n junction, due to 3CSIR-Network of Institutes for . photovoltaic effect an electron–hole pair is produced. The e-mail : [email protected] ; [email protected] presence of electric field across the junction of this device

VOL. 81, NOS. 11–12 337 TABLE I: Historical developments in solar cell technology from 1839 to 2007 [1]

Year Achievements/Milestones

1839 Discovery of photovoltaic effect by Edmond Becquerel in electrolytic cell made up of two metal electrodes upon exposure of sunlight 1873 Demonstration of photoconductivity of selenium by Wilough by Smith 1876 William Grylls Adams and Richard Evans Day discovered that illuminating a junction between selenium and platinum generates a photovoltaic effect. 1883 Development of cells made from selenium (Se) Fig. 1. Working principle of a typical p-n junction solar cell. wafers 1905 Publication of a legendary paper on “” by causes the electron-hole pair to drift apart to produce Albert Einstein (Nobel Prize in the year 1921). electricity provided recombination process is avoided. Fig.1 1932 Demonstration of photovoltaic effect in CdS by Audobert and pictorially represents the working principle of a typical p- Stora n junction solar cell. Efforts to convert sunlight to electricity 1954 Realization of photovoltaic technology by Bell Telephone using photovoltaic cells dates back to centuries. But the Laboratories with the development of Si-PV cell capable of use of sunlight to generate fire, lighting for religious converting solar energy into electricity suitable for running purposes, burning wooden ships of enemies in the war field, electrical equipment for daily usage. Production of silicon solar cell with 4 % efficiency. etc. was seen from as early as the 7th century B.C [1]. However, the photovoltaic effect was first experimentally 1958 Ted Mandelkorn of U.S. Signal Corps Laboratories produces n-on-p silicon PV cell by diffusing P-type layer on n-type observed in 1839 by a French scientist Edmond Becquerel layer. who demonstrated an increment in electricity upon exposure 1958 First PV-powered satellite, Vanguard I, was launched by to sunlight when two metal electrodes were immersed in United States Naval Research Laboratory. an electrically conducting solution. Later in 1876, Adams 1963 Production of practically viable silicon photovoltaic module and Day discovered photocurrent generation from selenium by Sharp Corporation (Se) under exposure to sunlight. It was in 1905, Albert 1963 Installation of 242 watt PV array on a lighthouse by Japan Einstein reported photoelectric effect along with his famous which was world’s largest array at that time. theory of relativity and subsequently he was bestowed with 1966 Launching of first Orbiting Astronomical Observatory of 1 Nobel Prize in the year 1921. Photovoltaic effect in CdS kilowatt PV array by NASA was discovered in 1932 by Audobert and Stora. After two 1976 Development of first PV cell by D. Carlson decades, in the year 1954 revolutionary progress in and C Wronski photovoltaic cell took place at Bell labs, USA with the 1981 First solar-powered aircraft, the “Solar Challenger” built by development of silicon photovoltaic which is capable of Paul MacCready, flied from France to England across the converting sunlight into electric power with an efficiency English Channel. of 4% having potential to actually run electrical equipment. 1982 First solar powered car “the Quiet Achiever” developed by Successively with further improvements, 11% efficiency for Hans Tholstrup Si-photovoltaic was achieved at the same Lab. Initially it 1985 Record breaking 20% efficiency for silicon solar cells under was explored for space research applications but continuous one sun conditions by the University of South Wales efforts are being made to use it for domestic purpose also. 1986 The first commercial thin film power module released by Although Photovoltaic devices are simple in design ARCO Solar requiring very little maintenance along with their biggest 1994 Development of gallium indium phosphate and gallium advantage of easy construction as stand-alone systems to arsenide solar cells which exceeded 30 % conversion give outputs from microwatts to megawatts, major efficiency. bottlenecks for this technology to take-off to replace fossil 1999 Development of tandem solar cell photovoltaic by Spectrolab and National renewable Energy. Implementation of Solar fuel energy are higher cost per unit of electric energy and concentrator systems mounted on tracking systems. limited conversion efficiency. Table 1 lists historical 2007 Research group at university of Delaware led by Allen Barnett advancements of photovoltaic technology in chronological achieved a record breaking 42.8% efficiency under standard order1. terrestrial conditions

338 SCIENCE AND CULTURE, NOVEMBER-DECEMBER, 2015 The developmental stages of solar cell can be and also highlight a part of such work that classified into three generations. The first generation mainly is presently being carried out at CSIR-CGCRI under the includes the solar cells prepared from thick wafers of single TAPSUN programme. crystalline semiconductors, mainly silicon (Si) and germanium (Ge). Especially, silicon wafers are most commonly used and are more reliable and efficient due In this type of solar cells, low cost semiconducting their chemical structure. Hence till now most widely used polymers have been widely explored since the last three solar cells are based on silicon only which exhibit an decades. The low weight, semi-transparent polymer material efficiency of 20-24%. However growing large crystals of attracted the researchers mainly for its chemical flexibility pure silicon is difficult as well as expensive. Therefore the towards modification which facilitated large scale production cost of this type of cells is exorbitantly high production at a much cheaper cost. The 1st generation and ultimately resulting in a very high cost per unit of organic solar cells consisted of a single layer of polymer energy. Another issue with cells is material sandwiched between two metal electrodes of that their efficiency degrades with increase in cell different work function. The polymers were mainly hole o temperature by about 25 C. To tackle these shortcomings, (h) conductors (p-type) with a band gap of nearly 2 eV. monocrystalline silicon wafers have been replaced with The difference in the work function of two metals brings multi- for the active material of solar cells. about an asymmetry of e and h injection into the molecular This resulted in a reduction of manufacturing cost but at * and  orbital along with the formation of schottky barrier the expense of reduction in efficiency to about 13-14%. between the p-type organic material and the metal with With an aim to bring down the production cost of solar the lower work function. However, the power conversion cell through reducing the amount of active material, second efficiency of these devices was much poor (in the range generation photovoltaics have evolved in the form of thin 10-3 to 10-2 %) due to very low charge carrier mobility films of amorphous silicon (a-Si), CuIn(Ga)Se2 (CIS), CdTe and small diffusion lengths of the primary excitons3. This or CdS or polycrystalline Si deposited on low cost can be improved by increasing the charge carrier substrates. These cells utilise only 1-10 µm thick of active concentration which can be achieved by doping with some material and hence reducing the material cost substantially. n-type organic materials, like perylene and its derivatives However, in terms of energy conversion efficiency, it is through different wet processing techniques or by co still lower than that of c-Si. For CdTe cells, the efficiency evaporation of both the matrix and dopant. The bilayer 2 is nearly 16.5% while for CIS cell it is about 12.5% . The heterojunction structure of the device containing a p-type third generation solar cell technology mainly targets at and an n-type organic material stacked between two enhancing the conversion efficiency of solar cells by electrodes matching the donor HOMO and the acceptor utilising maximum extent of incident solar energy to LUMO, exhibited effective transfer of charge carriers produce electron-hole pairs. This can be brought about by between the two layers under illumination. Reduced rate frequency conversion of incident photons which minimises of recombination due to repulsive interaction between the two major loss processes in a solar cell, one being the interface dipoles and free charge resulted in greater charge thermalisation loss resulting from the excess energy of extraction and hence enhancement in efficiency. Power photons over the band gap energy and the other being the conversion efficiency of about 3.6% is achieved under AM sub-band gap loss resulting due to non absorption of 1.5 illumination using a bilayer device of copper photons having lower energy than the band gap energy of 4 phthalocyanine and C60 . The development of bulk the semiconducting material. This attempt witnessed the heterojunction, however has significantly improved the emergence of several new types of solar cells like organic efficiency of the device in the last ten years. Intimate solar cells, dye sensitized solar cells, hybrid solar cells as blending of conjugated polymers with fullerene (and its well as exploitation of band-gap engineering techniques by derivatives) throughout the bulk showed enhanced the utilisation of multi-junction systems like tandem cells. photoconductivity due to photo induced charge transfer Use of different types of fluorophores, quantum dots, from the excited state of the polymer to the much more quantum wells in different host materials occupies the major 5,6 electronegative C60 . However it requires percolated areas of investigation in this generation of solar cells having pathways for the electrons and holes to reach the contacts high potential to provide the much required high and thus this device structures are much sensitive to the performance, low-cost photovoltaic. nanoscale morphology in the blend. Till now an efficiency The proceeding sections address the routes that of nearly 9% has been achieved in this device for single followed to improve efficiency of third generation junction and about 10% in tandem cells7,8. Continuous

VOL. 81, NOS. 11–12 339 efforts are being made through different routes to bring about an enhancement of efficiency up to 15% for large scale applications.

Dye Sensitized Solar Cells Dye Sensitized Solar Cell (DSSC) can be assumed as a mesoscopic heterojunction solar cell or a mesoscopic electron-injection solar cell having significant advantages over other PV technologies in terms of the shorter energy payback time, performance under diffuse light conditions, cost effectiveness and novel architectures9.

Advantages of DSSCs

 Low production cost and particularly interesting much lower Fig. 2. Schematic representing working principle for DSSC device. investment costs compared with conventional PV technologies. A DSSC is in essence a photochemical cell which contains two electrodes and an electrolyte and generates  Feedstock availability to reach terawatt scale. electrical current by redox reactions. The working electrode  Enhanced performance under real outdoor or photoanode of the cell which consists of nanostructured, conditions mainly at diffuse light and higher porous semiconducting oxide layer, deposited on a temperature. transparent conducting oxide (TCO) glass or plastic substrate. The glass coated with fluorine-doped tin oxide  Bifacial cells capture light from all angles. (FTO) or indium-doped tin oxide (ITO) has commonly been  Cell voltage does not decrease so much under low used for this purpose. A monolayer of the light-sensitive light intensity appropriate for use indoors and in charge transfer dye has been further attached to the poor weather conditions. semiconducting oxide film. Similarly the electrically  Performance is stable over a long life-time. conducting glass substrate coated with a thin layer of platinum catalyst acts as the counter electrode. This has  Materials used in DSSC are relatively non toxic. been placed parallel with a face to face configuration to  Short energy payback time (Â 1 year). the working electrode. The interspacing has been filled with  Design opportunities such as transparency and the liquid or solid electrolyte that plays a role of conducting 10 multi- colour options. media . The model structure of a complete DSSCs device including the principal components namely the TCO,  Flexibility and lightweight. photoanode semiconductor, light sensitizing dye, electrolyte, Pt electrode along with the basic electron transport Disadvantages of DSSCs processes among them have been illustrated in Fig. 2.  Difficulties with sealing the cell causing leakage and evaporation of the electrolyte. Components of DSSC Device  Lower efficiencies compared with conventional Photosensitizers : The most widely used solar cells. photosensitizers are Ru (II) polypyridyl dyes where the light

 The corrosive nature of the iodide/triiodide redox absorption originates due to MLCT process with molar -1 -1 11 mediator. extinction coefficient () of 10,000-20,000 M cm . Functionalizing the ancillary polypyridyl ligand by the  Difficulties in scaling up the production to large substituents including alkyl, aryl, alkoxy, phenyl, modules. heterocycle etc., or by replacing the chelating anions many

340 SCIENCE AND CULTURE, NOVEMBER-DECEMBER, 2015 Fig. 3. Structure of the photo sensitizers (a) N3, and (b) N719 and (c) Z-907. popular dyes such as N719, N3, N749 (black dye), Z907, efficiency of 20% is being expected in near future from K51, K60, N886, C103, JK56 etc. have been synthesized. the device15-18. A maximum of 11.7% efficiency have been reported to be achieved with these Ru(II) dye based DSSCs till Electrolytes 12-15 2010 . The structure of three extensively used dyes N3, The most extensively used electrolyte is the iodide- N719 and black dye are illustrated in Fig. 3. - - triiodide (I3 /I ) electrolyte exhibiting superior performance Besides the Cu(II) and Zn(II) porphyrins and Zn(II) constantly and an efficiency of 11% have been achieved - - 19,20 phthalocyanines dyes, metal free coumarin dyes, indoline from the I3 /I electrolyte based DSSCs yet . dyes, triarylamine dyes, carbazole dyes, squarines, perylene Counter Electrodes dyes are quite mentionable. The cadmium chalcogenite QDs including CdS, CdSe, CdTe and their different alloys, the Platinized conducting glass is mostly used. The carbon lead-chalcogenite QDs such as PbS, PbSe and the antimony materials such as porous carbon, single wall carbon sulphide QDs have also been frequently studied. The nanotubes, graphene, conductive polymers such as PANI, PEDOT, PPy and the sulfide inorganic compounds such as organolead halide or perovskite CH3NH3PbX3 (X= Cl, Br, I) photosensitizers have been spotlighted most in recent CoS2, CuInS2, CuZnSnS4 are also being explored recent 10,15 years owing to their outstanding light harvesting properties times . with E of ~1.5 eV, optical absorption edge of ~820 nm g Photoanodes and  of ~ 150,000 M-1cm-1 16. Based on the promising improvement of the ongoing researches the conversion The R & D on DSSC promoted further based on the

Table 2. Properties of commonly used photoanodes in DSSC.

Material Isoelectric point Band gap(eV) ECB = 0.0 V Electron Highest reported Vs NHE affinity (eV) efficiency ()

Anatase TiO2 5-6 3.3 0.5 V 3.9 14.1%

ZnO 4-9 3.3 Close to TiO2 higher 4.5 6.5% than SnO2

SnO2 2.5-4.0 3.5 Lower than TiO2 4.8 6.02%

In2O3 7.1 3.6 Lower than TiO2 4.45 2.37%

Nb2O5 2.6-4.5 3.4 0.2-0.3 V 2.34 4.10%

VOL. 81, NOS. 11–12 341 pioneer work of Michael Gratzel group in 1991 based on the conversion efficiency of 1.38% and 0.63%, respectively 21 mesoporous TiO2 nanoparticles as photoanode . Incessant [25]. The J-V characteristics of the ZnO rods are exhibited research establishes TiO2 as the ideal photoanode material in Fig. 4. exhibiting the highest efficiency of over 12% so far22. The basic properties and performances of some commonly used materials as photoanode are exhibited in Table 2.

Performance Measurement The overall solar-to-electrical energy conversion efficiency,  , for a solar cell is given by the photocurrent density measured at short-circuit (Jsc), the open-circuit photo voltage (Voc), the fill factor of the cell (FF), and the intensity of light (Pin).  = (Jsc *Voc *FF)/ Pin Another fundamental measurement of a solar cell is the external quantum efficiency, which in the DSC community is normally called the incident photon to current Fig. 4. J-V characteristics of (a) as prepared ZnO rods and that annealed conversion efficiency (IPCE).The IPCE value corresponds at (b) 300 ° and (c) 600 °C, respectively. to the photocurrent density produced in the external circuit Attempts were also made to synthesize alternative under monochromatic illumination of the cell divided by photoanodes such as spinel Zn SnO and BaSnO 26,27. Such the photon flux that strikes the cell. 2 4 3 systems are expected to increase the stability and photo IPCE = Jsc() / e() conversion efficiency of the device further. Efforts also being made to utilize the sysnthesized porous photoanodes As far as the objective of our work was concerned, as scaffold layer in the newly emerged perovskite systems. we have given emphasis exclusively in fabricating various photoanodes by economically viable and effective Multi Junction Solar Cell techniques and studying on their performance in N719 dye based DSSC. We have also focussed our attention on In general, more the extraction of incident solar energy identifying alternative photoanodes to replace the existing by active material in a solar cell higher the electric power produced. However, any active material will have definite TiO2 and SnO2 based oxides. In this respect, we have synthesized different microstructures of ZnO including 1D- band gap energy that defines its extraction capacity. Hence rod, 3D- flower and particle through solution processed in order to increase absorbance of solar energy over a wider sonochemical method. The nature of structural defects in wavelength range, different semiconductor materials with ZnO rods, nanoparticles and flowers have been investigated different p–n junctions could be utilized in multi-junction by several spectroscopic studies such as photoluminescence, solar cell which brings in a significant enhancement in the Raman and EPR. An overall conversion efficiency of 1.38% conversion efficiency. While crystalline silicon solar cell was achieved for the DSSCs fabricated with ZnO rods, efficiency is practically limited maximum up to 25.6 %, whereas the particles and flowers exhibited lower the multi junction cell with GaInP/GaInAs/Ge concentrator efficiencies of 1.004% and 0.59%, respectively. The higher cell, demonstrated a record 46 % efficiency28. Most efficiency exhibited by the rods has been attributed to the interestingly, multi-junction solar cells based on III–V minimum oxygen vacancy defects along with favourable semiconductors, II–VI compounds and polymers exhibit one dimensional structure for providing direct pathway of highest efficiencies over any other present day photovoltaic electron conduction. Correlating all the experimental devices due to the considerable reduction of thermalization evidences, it has been confirmed that the nature and and transmission losses unlike other PV materials owing concentration of structural defects are quite decisive than to the presence of multiple p-n junctions with various band shape in monitoring the overall performance of ZnO based gap energies. Though it is possible to achieve very high in DSSCs23-25. The effect of post annealing temperatures on these tandem cells, implementation of various costly the native defects of ZnO rods have been studied in detail semiconducting oxides causes a significant enhancement in to understand and its ambience on DSSC performance26,27. the manufacturing cost. Thus very high price-to- o The as prepared rods and that annealed at 600 C exhibited performance ratio of these cells has limited their market in

342 SCIENCE AND CULTURE, NOVEMBER-DECEMBER, 2015 PV industry. However, the high power to weight ratio has 10,000 MW by 2017, and a total of 100,000 MW by 2022 opened up unique application fields towards aerospace and 31. At present, Rajasthan with its installed power of 1170 terrestrial applications. The futuristic approaches in tandem MW tops the state-wise list followed by Gujarat (1000 cells include integrating III-V compound semiconductor MW), Madhya Pradesh (603 MW) and other states are materials on Si substrate addressing the critical issues with ramping up their solar capacity. the material growth, quality of grown crystals as well as However in this attempt another major bottleneck is reproducibility. the efficiency of a Si-photovoltoic cell, which is limited Silicon Solar Cell due to three loss mechanisms: Despite the advances in different types of solar cells, i) Recombination loss of the photo generated till date the solar cell market is dominated by silicon minority carriers in the silicon bulk and at the photovoltaics. Although silicon solar cells provide the surface. highest energy conversion efficiency amongst all, still it Attempts to minimise the thickness of cell material falls much short to be compared to that of commercial tends to increase the chances of recombination of grid electricity. Two main reasons for this are, (1) greater the photo generated carriers. But the cost per unit of electricity and (2) its low conversion implementation of a passivation layer at the front efficiency due to intrinsic or extrinsic losses in addition to and rear surfaces can enhance the effective lifetime restricted utilisation of incident solar energy. Three factors of these minority carriers. Normally, this that contribute to the fabrication cost for solar modules passivation is achieved either by reducing the are, cost of the silicon substrate (50%), cell processing surface defects (surface passivation) or by using a (20%) and module processing (30%)29. Thus, reducing the cost of silicon substrate remains one of the most important technique of band bending at Si surface creating issues in the PV industry which led to the emergence of electric field (field effect passivation). Deposition low cost multi-crystalline silicon wafers or thin films. As or growth of dielectric films like amorphous silicon it was discussed earlier, monocrystalline Si PV are the most nitride (SiNx) containing high positive charge popular with its maximum efficiency of 24.7% followed density or alumina (Al2O3) with high negative fixed by 20% for multi-crystalline Si PV and 14% for thin film charge density can provide not only chemical Si PV. With continued efforts in lowering the cost per passivation but also field effect passivation. Field unit all over the globe, usage of solar energy effect passivation can also be obtained by is increasing exponentially for both industrial and domestic introducing high-low junction with the same type purposes over two decades. The global cumulative of impurities (p+-p or n+-n) or p-n junction with photovoltaic power is reported to be from just around 100 opposite doped types. The p+-p combination is MW in 1992 to around 178,391 MW in 2014 with commonly employed at the rear side of p-type projected target of 233,000 MW by 201530. Worldwide in silicon solar cells thus creating ‘Back surface Field’ many countries, governments even have taken initiatives (BSF). However, the conventional Al (BSF) yields in providing economic incentives for investments in solar a maximum VOC of about 630 mV, corresponding PV. European countries were the pioneers along with to rear surface recombination velocity (Seff) ranging Japan in this solar PV production. However, United States from 200-600 cm/sec, which is insufficient for high was topping the list for installed photovoltaics for many efficiency Si solar cells. Recent investigations years up to 1996 after which Japan ousted US and remained revealed thermal and plasma ALD deposited Al2O3 leader until 2005. Later, Germany took the world’s lead films can provide a very high passivation effect in producing solar power and recently approaching 40,000 with very low surface recombination velocities MW mark. Interestingly, cost of solar also declined S Â5 cm/s on p-type and n-type Si(typically 1- significantly due to improvements in technology and last eff 4cm) after annealing at moderate temperatures32. year, 2014 has seen a steep increase in production of solar The ALD synthesized Al O films can limit the cells and modules together with more deployment of 2 3 emitter saturation current density to -10 and -30 photovoltaics in Asia particularly in China and India. China fA/cm2 thereby producing a V up to 700 mV is expected to continue its rapid growth and enhance its OC PV capacity to 70,000 megawatts by 201730. In national which indicates that the level of passivation scenario, the installed grid connected solar power capacity achieved by Al2O3 is higher than that obtained by 33 is 4,229.36 MW and India expects to install an additional thermal SiO2, a-Si:H and SiNx .

VOL. 81, NOS. 11–12 343 ii) Heating loss arising due to series resistance in the improvement in efficiency36,37. The reflection loss gridlines and bus bars and also at the interface can also be minimised by texturing the front between the contact and silicon. surface with pyramid or inverted pyramid like structures through different chemical etching and A metallic top contact is necessary to collect the vapour deposition techniques. photo-generated carriers. But the front metal grid of a conventional solar cell consisting of horizontal Apart from above three conventional losses in a solar fingers and vertical bus bars are associated with cell, another factor which limits the conversion efficiency three major loss mechanisms: one being the of solar cell is the incomplete utilisation of the incident resistive loss due to lateral conduction of majority solar energy that mainly arises due to the fixed bandgap carriers in the emitter to the metal lingers, second of Si. Photons having energy less than the band gap of Si being the resistive loss in the metal fingers and passes through the cell unabsorbed while photons with busbars, and third is the loss of light due to shading energy greater than the bandgap causes heating of the cell effect of metal fingers. To minimize these losses, resulting in thermalisation losses. This can be overcome grid spacing and designing should be optimized. by modulating the incident spectrum in such a way that Since the power loss from the emitter depends on photons having energy on either side of the band gap of the cube of the lone spacing, short distance silicon is converted to fall within the frequency range which between the fingers and multiple bus bars is Si-PV can absorb. Such spectral conversion can be attained desirable34. Fingers of low resistivity material by three possible phenomena, namely Upconversion, having higher height to width aspect ratio, is Downconversion and Downshifting of photons. The observed to be highly beneficial to reduce the following sections present a detailed study of the role of power losses in a solar cell. According to HB UC, DC and DS mechanisms on the efficiency of solar Serreze, a tapered busbar has lower losses than a cells. busbar of constant width. Smaller the unit cell, finger width and finger spacing, lower the losses. Spectral Conversion

Also, optimum width of busbar (Wb) occurs when Up / Down conversion and down shifting : Up- 35 resistive loss in busbar equals its shadowing loss . conversion is a non-linear anti-stokes luminescence process iii) Photon losses due to surface reflection where two or more low energy photons are required to generate one high energy photon. This phenomenon was The reflection losses occurring at the top surface first discovered by Auzel in the 1960s38. The phenomenon of a solar cell can be arrested by application of was usually observed in rare earth based materials. Up- single or double layer anti reflection coatings converting materials have applications in diverging areas deposited through chemical vapour deposition, such as display, solid-state lasers, biomarkers, temperature spray, spin-on or screen printing techniques. Plasma sensors etc. The application of upconverting materials for enhanced chemical vapour deposition (PECVD) Si based solar cells to enhance its efficiency has been enables deposition of these layers of uniform and proposed by Trupke et al.39, and theoretically demonstrated controlled thickness. TiO , Si N , SiO , Al O , 2 3 4 2 2 3 that the upper EQE limit of UC- silicon solar cell systems Ta O are most commonly used as ARC material. 2 5 can be enhanced to 47.6 % for single junction cells. In Recently a-SiNx:H and a-Si:C:H have proved to this concept, with an up-converting layer integrated with be a promising material for ARC and has been solar cell, sub band gap light could be converted to high observed to enhance the efficiency of mc-Si solar energy photons that can be absorbed by active material. cells from 9.84% to 14.25% while in c-Si, a double Fig. 5 presents implementation of an upconverting later on layer SiO2-TiO2 ARC has resulted in 37%

Fig.5. Schematic presentation of upconversion (UC), Downconversion (DC) and Downshifting (DS) mechanism.

344 SCIENCE AND CULTURE, NOVEMBER-DECEMBER, 2015 the rear side of a bifacial solar cell along with possible layers. So far there is no practical demonstration of energy level transitions. The same figure also represents efficiency enhancement using DC material on Si-PVs. the concepts of frequency down conversion and down CSIR-CGCRI has developed some of the materials which shifting with respective layers placed on front side of the show DC phenomenon. Rare earths such as Pr3+/Yb3+ co- solar cell. The first up-converter on a silicon solar cell has doped low phonon fluoro-tellurite glass systems were been demonstrated by Shalav et al., in the year 200540 developed which shows efficient downconversion emission which shows enhancement in the photo-current much less from Pr3+ ions to Yb3+ ions (theoretical DC efficiency than 1%. ~180%) near the bandgap of Silicon under UV-Vis excitation where the response of Si-PVs are very poor. So far, upconverting materials used for Si solar cells Also, Eu2+/Eu3+-Yb3+ co-doped silicate based oxyfluoride were based on phosphors which are not transparent and glass and glass-ceramics samples have been developed and only applicable on the rear side of the bifacial Si-PV. In frequency down-conversion has been demonstrated from this regard CSIR-CGCRI has developed some of rare earth Yb3+ ions near 1 µm under broad UV excitation which is doped transparent glass systems which show efficient attributed to the cooperative energy transfer from Eu2+/ upconversion properties under IR (1550 nm) light defect centres to Yb3+ ions. Several other rare earth ion excitation. The advantage of these materials is that they pairs in glass or glass ceramic hosts were under examination have minimum scattering losses compare to phosphors and for potential application of down converting materials to are transparent in UV-Vis-NIR region which can be improve the efficiency of Si-PVs. potentially utilised on either side of the Si-PVs for its efficiency enhancement. Frequency downshifting involves the conversion of one high energy photon, which is inefficiently absorbed Down-conversion is a process where one high energy by the active material of the solar cell to one low energy photon will cut in to two or more low energy photons. photon that it can absorb. This phenomenon obeys Stokes Initially, the Down converting mechanism has been Law and the change in wavelength is known as the Stokes explored decades back while working with phosphors which shift. Ideal luminescent materials used for downshifting convert Vacuum Ultra Violet light to visible light for should exhibit the following properties: unity quantum lighting applications. Trupke et al.41, showed that, using a efficiency; a wide absorption band in the region where the silicon solar cell with an ideal DC, a conversion efficiency radiation is required to be absorbed; high absorption of 38.6% could be achieved under unconcentrated sunlight coefficient; a narrow emission band, corresponding to a which will potentially reduces the thermalization losses region where the cell quantum efficiency is high; and good encou-ntered in Si-PVs. Fig. 3 describes the proposed separation between its absorption and emission bands to configuration of single junction, bifacial solar cell integrated minimize losses due to recombination. Not only in solar with down converting, down shifting and upconverting glass cells, downshifting of luminescence has been widely explored in different systems for several applications like lasers, production of white LEDs etc. Hoevel et.al., first applied the concept of luminescent downshifting on top of PV cells and studied its effect on spectral response of solar cell which showed that this system has the potential to enhance the efficiency of the device42. Thereafter it has been studied widely on different types of solar cells. For LDS encapsulated c-Si solar cells, an enhancement of about 40% has been reported by Richards et al43. Fig. 6. Solar spectrum and its modulation by application of UC, DC or DS layer on single junction, Since transition metals exhibits bifacial solar cell. broad absorption band, they are

VOL. 81, NOS. 11–12 345 Fig.7. Solar Simulator set-up (Inset: (a) Picture of glass on solar cell under solar simulator ; (b) I-V curve of standard cell with and without glasses prepared at CSIR-CGCRI) used as sensitizers to rare earth ions and thus helps to bring in their conversion efficiencies, solar power is becoming about effective downshifting of photons as observed in affordable and Solar PV is progressively contending with Cr3+- Yb3+ and Cr3+-Nd3+ pairs44,45. Such downshifting of conventional energy sources in many countries. India also photons from UV to visible has been effectively realised going towards this by undertaking many strategies. So, we in Ce3+ -Eu3+ and Ce3+- Tb3+ doped metaphosphate glass have entered in an era where Sun (solar power) would be prepared in CSIR-CGCRI. A part of the UV energy illuminating dark nights.  absorbed by Ce3+ is efficiently transferred to Tb3+ and Eu3+ which exhibits strong luminescence in the green and red References region where EQE of c-Si solar cell is high. Also the 1. The History of Solar, https://www1.eere.energy.gov/solar/pdfs/ quantum yield of the dopant ions is much higher than that solar_timeline.pdf reported in other host matrices. Further doping the Ce3+- 2. P.C. Choubey, A. Oudhia and R. Dewangan, A review: Solar 3+ 2+ cell current scenario and future trends, Recent Research in Tb metaphosphate glasses with Mn resulted in a broad Science and Technology, 4(8): 99 (2012) luminescence covering almost the entire visible portion of 3. G.A. Chamberlain: Organic solar cells: A review, Solar Cells, the spectrum ranging from 320-750 nm. This proves to be 8, 47(1983). highly promising since the EQE of c-Si solar cell is high 4. P.Peumanns and S.R. Forrest, Very-high-efficiency double in the visible region. These glasses have been tested for heterostructure copper phthalocyanine/C60 photovoltaic cells, their performance on solar cell under solar simulator Appl. Phys. Lett., 79, 126 (2001). equipped with Xenon arc lamp as illuminator fitted with 5. N.S. Sariciftci, L. Smilowitz, A.J. Heeger, and F. Wudl, Photoinduced electron transfer from a conducting polymer to AM 1.5 filter to get 1 sun test condition. This is the buckminsterfullerene, Science 258, 1474 (1992). conventional procedure to test solar cell performance in 6. G. Yu, J. Gao, J.C. Hummelen, F. Wudl, and A.J. Heeger, Polymer the laboratory condition as illustrated in Fig. 7 along with photovoltaic cells: Enhanced efficiencies via a network of test result and the picture of the glass sample exhibiting internal donor-acceptor heterojunctions, Science 270, 1789 (1995). luminescence under light exposure in the inset. Currently 7. He, Z. et al. Enhanced power-conversion efficiency in polymer efforts are being carried out to integrate luminescent glass solar cells using an inverted device structure, Nat. Photon., 6, layers onto Si-solar cells for improved conversion 591 (2012). efficiency. As a consequence of exertions towards lowering 8. You, J. et al., A polymer tandem solar cell with 10.6% power production costs of photovoltaics along with improvements conversion efficiency. Nat. Commun., 4, 1446 (2013)

346 SCIENCE AND CULTURE, NOVEMBER-DECEMBER, 2015 9. K. Kalyansundaram , Dye Sensitized Solar Cells, CRC Press, 26. P. P. Das and P. S. Devi, Formation of Self-Assembled Defect-

(2010). Free Zn2SnO4 Nanostructures from Binary Oxides without the 10. A. Hagfeldt, A, G. Boschloo, L. Sun, L. Kloo and H. Pettersson, Kirkendall Effect, Inorg. Chem., 53, 10797 (2014). Dye Sensitized Solar Cells, Chem. Rev., 110, 6595 (2010). 27. P. P. Das, A. Roy, S. Das, P. S. Devi, Enhanced stability of Zn SnO with N719, N3 and eosin Y dye molecules for DSSC 11. S. Ardo and G. J. Meyer, Photodriven heterogeneous charge 2 4 transfer with transition-metal compounds anchored to TiO application, Phys. Chem. Chem. Phys., DOI: 10.1039/ 2 c5cp04716a (2015). semiconductor surfaces, Chem. Soc. Rev., 115, 38 (2009). 28. M. A. Green, K. Emery, Y. Hishikawa, W. Warta, and E. D. 12. D. B. Kuang, C. Klein, S. Ito, J. E. Moser, R. Humphry-Baker, Dunlop, tables (Version 45), Progress in N. Evans, F. Duriaux, C. Gratzel, S. M. Zakeeruddin and M. Photovoltaics: Research and Applications, 23, 1, (2015) Gratzel, High Efficiency and Stable Mesoscopic Dye Sensitized Solar Cells Based on a High Molar Extinction Coefficient 29. T. Saga, Advances in crystalline silicon solar cell technology Ruthenium Sensitizer and Nonvolatile, Adv. Mater., 19, 1133 for industrial mass production, NPG Asia Mater., 2(3), 96 (2010) (2007). 30. https://en.wikipedia.org/wiki/Solar_energy 13. S. A. Haque, E. Palomares, B. M. Cho, A. N. M. Green, N. 31. https://en.wikipedia.org/wiki/Solar_power_in_India Hirata, D. R. Klug and J. R. Durrant, Charge Separation versus 32. A. G. Aberle, Surface Passivation of Crystalline Silicon Solar Recombination in Dye-Sensitized Nanocrystalline Solar Cells: Cells-A Review, Prog. Photovolt: Res. Appl., 8, 473 (2000) the Minimization of Kinetic Redundancy, J. Am. Chem. Soc., 33. B. Hoex, J. Schmidt, R. Bock, P. P. Altermatt, and M. C. M.Van 127, 3456 (2005). de Sanden, Excellent passivation of highly doped p-type Si 14. X. Yang, Y. Numata and L. Han, Highly efficient dye-sensitized surfaces by the negative-charge-dielectric Al2O3, Appl. Phys. solar cells: progress and future challenges, Energy. Environ. Sci., Lett., 91, 112107 (2007). 6, 1443 (2013). 34. Stefan Braun, Giso Hahn, Robin Nissler, Christoph Pönisch,Dirk 15. M. Ye, X. Wen, M. Wang, J. Iocozzia, N. Zhang, C. Lin and Z. Habermann, The multi-busbar design:an overview, Energy Lin, Recent advances in dye-sensitized solar cells: from Procedia 43, 86 ( 2013 ) photoanodes, sensitizers and electrolytes to counter electrodes, 35. HB Serreze ,Optimizing Solar Cell Performance by Simultaneous materialstoday, 18, 155 (2015). Consideration of Grid Pattern Design and Interconnect 16. N. G. J. Park, Organometal perovskite light absorbers toward a Configurations, 13th IEEE Photovoltaic Specialists Conference. 20% efficiency low-cost solid-state mesoscopic solar cell, Phys. :1-8 (1978) Chem. Lett., 4, 2423 (2013). 36. B.Swatowska, T. Stapinski, K.Drabczyk, P. Panek, The role of 17. M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. antireflective coatings in silicon solar cells the influence on their Snaith, Efficient hybrid solar cells based on meso-superstructured electrical parameters, Optica Applicata, XLI, (2), (2011) organometal halide perovskites, Science, 338, 643 (2012). 37. K.Ali, S. A. Khan, M. Z. Mat Jafri, Effect of Double Layer 18. N. J. Jeon, H. G. Lee, Y. C. Kim, J. Seo, J. H. Noh, J. Lee and (SiO2/TiO2) Anti-reflective Coating on Silicon Solar Cells, Int. S. I. Seok, o-Methoxy substituents in spiro-OMeTAD for efficient J. Electrochem. Sci., 9, (2014) inorganic–organic hybrid perovskite solar cells, J. Am. Chem. 38. Francüois Auzel, Upconversion and Anti-Stokes Processes with Soc., 136, 7837 (2014). f and d ions in Solids, Chem. Rev., 104, 139 (2004) 19. B. O’Reagan and M. Gratzel, A low-cost, high-efficiency solar 39. T.Trupke, M. A. Green, and P.W¨ urfel., Improving solar cell cell based on dye-sensitized colloidal TiO2 films, Nature, 353, efficiencies by upconversion of sub band- gap light, J. Appl. 737 (1991). Phys., 92(7), 4117 (2002). 20. Q. Yu, Y. Wang, Z. Yi, N. Zu, J. Zhang, M. Zhang and P. Wang, 40. A. Shalav, B. S. Richards, T. Trupke, K.W. Kr¨amer, and H. U. 3+ High-efficiency dye-sensitized solar cells: the influence of lithium G¨ udel, Application of NaYF4: Er up-converting phosphors ions on exciton dissociation, charge recombination, and surface for enhanced near-infrared silicon solar cell response Appl. Phys. states, ACS Nano, 4, 6032 (2010). Lett., 86, 013505 (2005). 21. M. Gratzel, Solar energy conversion by dye-sensitized 41. T. Trupke, M.A. Green, P. Wurfel, Improving solar cell photovoltaic cells, Inorg. Chem., 44(20), 6841 (2005). efficiencies by down-conversion of high energy photons, J. Appl. 22. A. Yella, H. W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Phys., 92 (3), 1668, (2002). Nazeeruddin, E. W. G. Diau, C. Y. Yeh, S. M. Zakeeruddin and 42. H.J. Hovel, R.T. Hodgson and J.M. Woodall, The Effect Of M. Gratzel, Porphyrin-sensitized solar cells with cobalt (II/III)– Fluorescent Wavelength Shifting on Solar Cell Spectral based redox electrolyte exceed 12 percent efficiency, Science, Response, Solar Energy Materials, 2, 19, (1979) 334, 629 (2011). 43. K.R. McIntosh,G.Lau, J. N. Cotsell, K. Hanton, D. L. Ba¨tzner, 23. P. P. Das, S. Mukhopadhyay, S. A. Agarkar, A. Jana and P. S. F. Bettiol and B. S. Richards, Increase in External Quantum Devi, Photochemical performance of ZnO nanostructures in dye Efficiency of Encapsulated Silicon Solar Cells from a sensitized solar cells, Solid State Sciences, 48, 237 (2015). Luminescent Down-Shifting layer, Prog. Photovolt: Res. Appl., 17, 191(2009) 24. A. Jana, P. P. Das, S. A. Agarkar and P. S. Devi, A comparative study on the dye sensitized solar cell performance of solution 44. S. Ye, J. Zhou, S. Wang, R. Hu, D. Wang, and J. Qiu, Broadband 3+ 3+ processed ZnO, Solar Energy, 102, 143 (2014). downshifting luminescence in Cr -Yb codoped garnet for efficient photovoltaic generation, Opt. Express, 21(4) 4168 25. P. P. Das, S. A. Agarkar, S. Mukhopadhyay, U. Manju, S. B. (2013) Ogale and P. S. Devi, Defects in Chemically Synthesized and Thermally Processed ZnO Nanorods: Implications for Active 45. E. Bovero, E. Cavalli, D. Jaque, J. García Solé, A. Speghini, 3+ 3+ Layer Properties in Dye-Sensitized Solar Cells, Inorg. Chem., Cr ’!Nd energy transfer in the YAl3(BO3)4 nonlinear laser 53, 3961 (2014). crystal, J. Appl. Phys., 98, 023103 (2005)

VOL. 81, NOS. 11–12 347