WO2012123620A1

WO2012123620A1 - Thin film photovoltaic cell structure, nanoantenna, and method for manufacturing

Info

Publication number: WO2012123620A1
Application number: PCT/FI2011/050227
Authority: WO
Inventors: Constantin Simovski
Original assignee: Aalto Korkeakoulusaatio sr
Current assignee: Aalto Korkeakoulusaatio sr
Priority date: 2011-03-16
Filing date: 2011-03-16
Publication date: 2012-09-20
Anticipated expiration: 2013-09-16
Also published as: US9252303B2; WO2012123645A9; WO2012123645A3; WO2012123645A2; US20140158198A1

Abstract

A thin film photovoltaic cell structure (1) comprises a substrate (2); a first dielectric layer(3) on the substrate (2); an active layer (4) on the first dielectric layer(3); and a plasmonic light concentrator arrangement (5) on the active layer (4) for coupling incident light at a first wavelength band into the active layer (4). According to the present invention, the thin film photovoltaic cell structure (1) further comprises a second dielectric layer(6) formed of a dielectric material which is transparent at the first wavelength band and at a second wavelength band on the plasmonic light concentrator arrangement (5); the first dielectric layer(3), the active layer (4), the plasmonic light concentrator arrangement (5), and the second dielectric layer(6) being configured to form a resonant cavity for coupling incident light at the second wavelength band into a standing wave confined in the resonant cavity.

Description

THIN FILM PHOTOVOLTAIC CELL STRUCTURE, NAN0ANTEN A, AND METHOD FOR MANUFACTURING

FIELD OF THE INVENTION

The present invention relates to thin film photovolta^¬ ic cells, particularly to plasmon-enhanced thin film photovoltaic cells. The present invention also relates to nanoantennas for plasmonic light concentrator ar^¬ rangements for thin film photovoltaic cells. The pre- sent invention further relates to manufacturing of plasmon-enhanced photovoltaic cells.

BACKGROUND OF THE INVENTION

Photovoltaic cells are used in a great variety of ap- plications to convert electromagnetic radiation, e.g. the solar radiation, to electrical energy.

By a photovoltaic cell is meant here generally a semi^¬ conductor-based component converting incident electro- magnetic radiation to electrical energy through a pho^¬ tovoltaic effect. In the case of solar radiation as the primary energy source, a photovoltaic cell is usu^¬ ally called a solar cell. In general, a photovoltaic cell comprises a p-type semiconductor region, an n- type semiconductor region, and an active region. The n-type and p-type semiconductor regions form a pn- junction. Photons of incident light having energy equal to or greater than the band gap energy of the active region material are absorbed, thereby generat- ing free electrons and holes in the active material. These free charge carriers are then collected by means of electrodes connected to the different sides of the pn-j unction . One promising technological trend within the photovol^¬ taic cell development is the field of thin film cells. In thin film photovoltaic cells the semiconductor lay^¬ ers of the device are realized as thin layers with a thickness in a range from a few nanometers to some tens of micrometers. Thin film cells can provide ad- vantages in the energy conversion efficiency and in the manufacturing costs. Decreased layer thicknesses also mean lower weight of the cells and the solar pan^¬ els formed of them. A thin overall structure enables also flexible device configurations. The active layer material of a thin film cell can be e.g. p-doped amor^¬ phous silicon or gallium arsenide.

In a photovoltaic cell, before any energy conversion can take place, light has to enter the cell and pene- trate to the active region. Thus, losses due to re^¬ flection, scattering, and absorption before the active region should be minimized. Moreover, having reached the active region, the light energy should be absorbed there as effectively as possible without passing through it or reflecting back out of the cell. This is an important issue in all photovoltaic cell configura^¬ tions. In thin film cells where the active region typ^¬ ically is a layer having a thickness in a range of on^¬ ly tens of nanometers to some micrometers, effective capturing of light energy into the active region be^¬ comes a key factor for the overall efficiency of the cell. On the other hand, light capturing solutions used with conventional thick film cells, e.g. differ^¬ ent surface textures for coupling the incident light into the cell, are usually not suitable for thin film configurations .

Thus, intensive effort is focused in the field of thin film photovoltaic cells on research and development of more and more efficient solutions to improved light capturing . Recently, one of the most active research areas has been the different forms of plasmonic light concentra^¬ tors (LC) based on plasmonic nanoparticles arranged in an array on the surface of a photovoltaic cell on the side of the incident light. The operation of plasmonic nanoparticles (PNP) is based on resonant excitation of surface plasmons by the incident light. With suitable configuration of the nanoparticles and proper struc^¬ tural connection between the nanoparticles and the ac- tive region of the cell, the plasmon resonance, i.e. resonant oscillation of the electrons in the nanopar^¬ ticles, results in efficient coupling of incident light into the active region. In this coupling, the light energy in the incident plane wave is concentrat- ed into a plurality of so called hot spots located at least partially within the active region of the cell. At the same time, reflection backwards from the active region as well as transmission through it is minimized.

Many of the recently reported research activities in the field of plasmonic light concentrators are focused on metallic nanoparticles located on top of a thin film solar cell and covered. The metallic nanoparti- cles can be randomly distributed separate particles of silver or gold. These kinds of nanoparticles for light coupling are disclosed e.g. in US 2009/0250110 Al . Al^¬ ternatively, a plasmonic light concentrator can be im^¬ plemented as an array of regular two-dimensional nano- particles like nanostrips as disclosed e.g. in EP 2109147 Al . The theory and principles of operation of both types of those nanoparticles and variations thereof are widely discussed in the scientific litera^¬ ture .

Common for both randomly organized nanoparticles and regularly arranged two-dimensional nanoparticles is that the wavelength/frequency band of the plasmonic enhancement is relatively narrow. With randomly dis^¬ tributed nanoislands, typically no more than about 10 % of the wavelength/frequency band where photovoltaic conversion can take place in the active region is covered. With an array of nanostrips, the double- frequency operation due to two quadrupole plasmon res^¬ onances can typically provide a maximal coverage of about 20% of the available photovoltaic wave- length/frequency band. Forming a light concentrator as a double-periodic grid of nanostrips can increase the efficiency via the multi-frequency resonance thereby achieved up to half of the useful photovoltaic band. So called nanoantennas (NA) arranged as an array on a photovoltaic cell provide an alternative way to imple^¬ ment plasmonic light concentrator arrangements. In a nanoantenna, the hot spot is created not due to exci^¬ tation of a collective mode in the regular grid (called surface plasmon polariton) as is the case in a regularly arranged array of two-dimensional nanoparti- cles, but due to excitation of the eigenmodes of the antenna unit (called localized surface plasmons) . Nat^¬ urally, the plasmon resonance is influenced to some extent by the electromagnetic interaction between the neighboring antennas of the array, but the major fac^¬ tor is anyway the configuration of a single nanoantenna. One known nanoantenna configuration is the bow-tie nanoantenna consisting of two oppositely placed sub- stantially triangular nanopatches. The local field at the plasmon resonance is concentrated in the gap be^¬ tween the apexes of the triangles. The center of this hot spot is located inside the substrate on which the nanoantennas are formed. The displacement of the hot spot into the substrate results from its higher per^¬ mittivity compared to that of the free space. Another known nanoantenna type is the dimer type configuration comprising two adjacent circular nanopatches. The de^¬ sign and operation principles of both bow-tie and di^¬ mer type nanoantennas are widely discussed in the sci^¬ entific literature.

An important feature of nanoantennas is that any strict antenna array regularity is not required. More^¬ over, single antenna units are not very sensitive to the manufacturing tolerances thereof. The more toler- ant geometrical dimensions of a nanoantenna array al^¬ low use of manufacturing equipment with lower cost than e.g. the electron or ion beam lithography usually required for manufacturing regularly distributed arrays of two-dimensional plasmonic nanoparticles . Thus, from industrial-scale manufacturing point of view, nanoantennas provide a very promising approach to implement plasmon enhanced photovoltaic cells. On the other hand, also the known nanoantenna configurations suffer from a narrow wavelength/frequency band of the plasmon resonance. For bow-tie antennas, the dipole- type resonance band typically covers only about 5 % of the available band of the photovoltaic effect.

As an approach entirely different from the plasmonic light concentrators, it is known to enhance the light capturing efficiency of a photovoltaic cell by placing the active region within a resonant cavity, usually called a standing wave Fabry-Perot resonant cavity. In general, such a resonant cavity is formed by two re- fracting, substantially lossless layers located at op^¬ posite sides of the active region to serve as reflect^¬ ing end elements of the cavity. The cavity is designed to form a standing wave confined in the thus formed cavity. Resonance enhances the optical field within the active region and thus increases the light captur^¬ ing efficiency. The energy which is absorbed in the active region is reimbursed by the incident light flux. A standing wave Fabry-Perot cavity can be manu^¬ factured with lower costs than e.g. the more compli^¬ cated regular plasmonic light concentrators. However, the light capturing enhancement covers again a very narrow portion of the available wavelength/frequency range of the photovoltaic effect, typically 5 - 7 %. More broadband standing wave Fabry-Perot type cavities are also known. However, they comprise complex multi^¬ layer structures located on both sides of the active photovoltaic region, the multilayer structures neces^¬ sitating nanometer-scale precision of manufacture. The costs of such cavities are comparable with the costs of the regular plasmonic light concentrators, thus making them unsuitable for use in typically very cost- critical thin-film solar cells.

As is clearly seen in the prior art description above, there is a strong demand in the field of thin film photovoltaic cell structures for more efficient light capturing solutions which, preferably, could be manu^¬ factured with reasonable manufacturing costs.

OBJECT OF THE INVENTION

The object of the present invention is to provide a thin film photovoltaic cell structure with improved light capturing efficiency and enabling cost-efficient industrial-scale manufacturing. Another object of the present invention is to provide a nanoantenna which is suitable, for example, for use in such thin film pho- tovoltaic cell structures. It is also an object of the present invention to provide a cost-efficient method for manufacturing such thin film photovoltaic cells.

SUMMARY OF THE INVENTION The above objects are achieved by the present inven^¬ tion characterized by what is presented in claims 1, 5, and 16. According to a first aspect, the present invention is focused on a thin film photovoltaic cell structure comprising a substrate; a first dielectric layer on the substrate; an active layer on the first dielectric layer for generating free charge carriers via a photo- voltaic effect; and a plasmonic light concentrator ar^¬ rangement on the active layer for coupling incident light at a first wavelength band into the active lay^¬ er . By a thin film photovoltaic cell structure is meant here a structure forming at least a part of a thin film photovoltaic cell. By a thin film photovoltaic cell, in turn, is meant here a photovoltaic cell in which at least the active layer where the actual pho- tovoltaic effect, i.e. the generation of free charge carriers via absorption of incident light energy, takes place is in the form of a layer having a thickness in the range from a few nanometers to some tens of micrometers, typically from tens of nanometers to a few micrometers. The active layer can be formed of any material and structure suitable in the form of a thin film for photovoltaic conversion. It can comprise e.g. p-doped amorphous silicon, or polycrystalline silicon or gallium arsenide.

The basic configuration of the thin film cell struc^¬ ture can be based on the conventional layer structure with superposed p- and n-type semiconductor layers for forming a pn-junction, and an active layer being formed between these two layers. However, the basic configuration of the cell can as well be the next gen^¬ eration lateral one having the p- and n-type semicon- ductor regions at the sides of an active region formed within the active layer.

Further, by a photovoltaic cell is meant here a func- tional unit which can form the core portion of one discrete photovoltaic device or component. Also, the photovoltaic cell comprising an active layer can be an integrated and inseparable portion of a larger photo^¬ voltaic assembly.

A plasmonic light concentrator arrangement is an ar^¬ rangement for coupling light into the active region by means of excitation of plasmonic resonance by the in^¬ cident light. As described in the background section, such light concentrators include randomly and regular^¬ ly arranged arrays of plasmonic nanoparticles as well as plasmonic nanoantennas.

According to the present invention, the thin film pho- tovoltaic cell structure further comprises a second dielectric layer on the plasmonic light concentrator arrangement. The second dielectric layer is formed of a material which is transparent at the first wave^¬ length band and at a second wavelength band. Moreover, the first dielectric layer, the active layer, the plasmonic light concentrator arrangement, and the se^¬ cond dielectric layer of the cell are together config^¬ ured to form a resonant cavity suitable for coupling incident light at the second wavelength band into a standing wave confined in the resonant cavity, prefer^¬ ably so that the maximum of this standing wave is lo^¬ cated inside the active layer.

Thus, it is one of the core principles of the present invention to combine into a single thin film cell structure both a plasmonic light concentrator arrange^¬ ment operating at a first wavelength band and a Fabry- Perot type standing wave resonant cavity operating at a second wavelength band. This combination of two dif^¬ ferent light coupling arrangements and thus two wave^¬ length bands of enhanced light coupling into the ac- tive layer enables an enhanced light capturing effi^¬ ciency over a broad wavelength range.

The enhanced overall light capturing efficiency of the photovoltaic cell structure according to the present invention is achieved not just by combining into a single device two separate light coupling arrange^¬ ments. Instead, it is the overall configuration of the entire cell structure which serves to provide the op^¬ erations of the two different light coupling arrange- ments. The second dielectric layer is the only addi^¬ tional element needed with respect to the known plas- mon-enhanced cells. On the other hand, those known cells also usually comprise some protective layer on the plasmonic light concentrator arrangement. In com- parison to that, in the invention this layer of the cell is just configured to operate, in addition to those protective purposes, also as a part of the reso^¬ nant cavity. It is, naturally, essential for the intended opera^¬ tions of those two light coupling arrangements that the optical performance of the photovoltaic cell structure and each element thereof is carefully ad^¬ justed. In other words, the materials and thicknesses of the different layers as well as the detailed con^¬ figuration of the plasmonic light concentrator arrangement must be selected properly according to the desired optical operation. No detailed parameters of the different layers can be defined here because of the great possibility of variations e.g. in the mate^¬ rials of the different layers and the desired wave^¬ length bands. However, designing the cell configura- tion according to the desired optical performance is kind of routine engineering for a person skilled in the art. In other words, given the basic principle of the invention, a person skilled in the art in the field of designing thin film photovoltaic cells is able properly to select the details of the cell struc^¬ ture based on the common general knowledge and the prior art disclosures available in the literature. The first and the second wavelength bands can be ad^¬ justed e.g. to the near infrared (NIR) and to the vis^¬ ible ranges of the electromagnetic spectrum, respec^¬ tively. This is an efficient embodiment particularly for solar applications. In some embodiments, it is preferable that the adjustment of those two wavelength bands is performed so that the local field enhancement within the active layer is maintained substantially continuously over a broad frequency range covering part of the visible range and part of the near infra- red range. Though thin-film solar cells are usually designed to operate in the visible range, and the near infrared solar radiation is wasted in them, it can be reasonable to complement in solar collectors the usual solar cells operating in the visible range by those operating in the near infrared. For example, this ap^¬ plies when the solar light is preliminarily concentrated by lens concentrators .

In a preferred embodiment, the plasmonic light concen- trator arrangement comprises a nanoantenna for concen^¬ trating the incident light at the first wavelength band into a spot extending into the active layer, preferably so that as large a portion as possible of the spot lies within the active layer. As described above in the background section, nanoantennas provide significant advantages over the randomly or regularly distributed nanoparticles in the less strict design and manufacturing tolerances of the nanostructures . By the spot into which the nanoantenna scatters the inci^¬ dent light energy at the resonance wavelengths is meant here a sub-wavelength spatial region whose ef- fective thickness is equal to or less than the thick^¬ ness of the active photovoltaic layer. In the art, this spot is commonly called a "hot spot".

In the nanoantenna based approach, the plasmonic light concentrator arrangement comprises preferably a nano^¬ antenna which is configured to provide multiple plas- mon resonance wavelengths within the first wavelength band . The conventional multi-frequency plasmonic nanoanten- nas are based on a combination of several separate an^¬ tenna units, one for each desired frequency, forming a unit cell of the antenna. This unit cell is then re^¬ peated to form a larger antenna array. The separate antenna units of the unit cell can comprise antenna units of the same type with different sizes, antenna units of different types, or antenna units differing from each other both in their types and sizes. For example, a unit cell of a multi-frequency antenna can be formed of a combination of dimer type and bow-tie an^¬ tenna units .

The unit cell of several separate antenna units neces^¬ sarily has a rather large area. When the size of the unit cell becomes large in comparison to the wave^¬ length of incident light, parasitic reflection from the periodic antenna array to two Fraunhofer maxima takes place, thus making the coupling efficiency of the antenna array sensitive to the direction of the incident light. In other words, the antenna array then works efficiently for perpendicular incidence, i.e. for a zero angle of incidence only. Moreover, even for this normal incidence, an antenna having a width sub^¬ stantially larger than the wavelength of the incident light forms a hot spot with a thickness easily exceed^¬ ing the typical thickness of the photovoltaic layer in a thin-film solar cell. Then, the transmitted wave ex^¬ tends outside the photovoltaic layer, and a signifi^¬ cant part of the incident light energy will be lost in the substrate layer (s) below the photovoltaic layer. To avoid the above problems, according to a second as^¬ pect, the present invention is focused on a nanoanten- na with multiple plasmon resonance wavelengths, which nanoantenna is suitable for use in a plasmonic light concentrator arrangement of a thin film photovoltaic cell structure as described above. However, the nano^¬ antenna of the present invention is not limited to be used as a part of a photovoltaic cell structure ac^¬ cording to the first aspect of the present invention only. They can also be used in other plasmonic light concentrator arrangements. The nanoantenna comprises a unit cell comprising two pairs of first sub-wavelength metal patches placed around a central gap separate from each other, the two first patches of each pair lying on opposite sides of the central gap centrally located on a straight connection line crossing the central gap, the connection lines of the two pairs of sub-wavelength metal patches being directed substantially perpendicularly to each other. In other words, in this embodiment the unit cell comprises at least four first sub-wavelength metal patches located so as to surround a central gap. The two pairs of patches serve for at least two resonance frequencies. In the conventional dimer type and bow-tie nanoantennas , there is only one useful plasmon resonance. For exam- pie, in the bow-tie geometry the useful resonance is excited by polarization of incident light along the axis of symmetry connecting the two triangles. There is actually also another plasmon resonance excited by polarization perpendicular to said axis. However, this second resonance occurs at a frequency significantly higher than that of the main antenna resonance and does not result in the formation of the hot spot in the central gap of the antenna. Moreover, at this res^¬ onance of a bow-tie nanoantenna, the local field is concentrated inside the triangular nanopatches and is mainly dissipated. With the two pairs of patches ac- cording to this embodiment of the present invention, the two resonant frequencies are by nature close to each other and can also overlap, thereby forming ef^¬ fectively a single broad resonance band. The basic principle of the nanoantenna described above is to implement one single unit cell capable of pro^¬ ducing multiple resonant frequencies. Thereby, the need for separate unit cells for different resonant frequencies is avoided, and the unit cell can be made substantially smaller than in the prior art solutions. Generally, the overall size of the nanoantenna should be less than or in the range of the wavelength of the incident light. The width of the unit cell comprising the two pairs of sub-wavelength metal patches, as de- fined in the direction of any of the connection lines, is preferably less than or equal to 1 ym. This limita^¬ tion of the maximum unit cell size ensures that the above-described problems of parasitic reflection and transmission of light energy into the substrate can be avoided in the visible and near IR wavelengths cover^¬ ing the most important wavelength range in solar energy applications.

The central gap width as defined in the directions of any of the connection lines is preferably in the range of 50 - 300 nm. Manufacturing of nanoantenna unit cells with a gap width of below 50 nm could necessi- tate expensive high resolution equipment and processes like e.g. electron beam lithography. On the other hand, with central gap widths of over 300 nm, the patches of an antenna designed for visible and near infrared wavelengths no more operate properly together so as to form the hot spot in the area of the gap.

In one preferred embodiment, each of the first sub- wavelength metal patches has substantially a shape of an isosceles trapezoid, the shorter base of the trape^¬ zoid facing towards the central gap. This shape has been found a good alternative to provide plasmonic resonance frequencies in the near infrared region. In a preferred embodiment optimized for operation around 1 ym and providing efficient field enhancement in a hot spot in the area of the central gap, each of the first sub-wavelength metal patches is formed of gold and has a thickness of 50 - 100 nm; the width of the central gap along any of the connection lines is 150 - 250 nm; the lengths of the shorter and longer bases of the trapezoid are 75 - 150 nm and 200 - 450 nm, re^¬ spectively; and the height of the trapezoid is 300 - 500 nm. In an alternative preferred embodiment, each of the first sub-wavelength metal patches has an elongated shape having a longitudinal axis of symmetry which is directed substantially perpendicularly with respect to the connection line on which the first patch lays. Elongated patches positioned perpendicularly with re^¬ spect to the fictitious connection lines are particu^¬ larly suitable for operation in the visible wavelength range. In an even more preferred embodiment, the unit cell further comprises, for each of the first sub- wavelength metal patches, a second elongated sub- wavelength metal patch lying centrally located on the same connection line on which the first sub-wavelength metal patch lies on the side of the first patch oppo^¬ site to the central gap, and having a longitudinal ax^¬ is of symmetry which is directed perpendicularly with respect to this connection line. This configuration thus comprises a total of at least eight sub- wavelength patches placed around the central gap of the unit cell.

In one particularly efficient embodiment of the unit cell comprising the first and the second elongated sub-wavelength metal patches, each of the first and the second sub-wavelength metal patches is formed of gold and has a thickness of 20 - 50 nm; the width of the central gap along any of the connection lines is 100 - 150 nm; the lengths of the first and the second sub-wavelength metal patches along the longitudinal axis of symmetry thereof are 100 - 150 nm and 200 - 250 nm, respectively; and the widths of the first and the second sub-wavelength metal patches in a direction perpendicular to the longitudinal axis of symmetry thereof are 25 - 50 nm and 50 - 100 nm, respectively. With this kind of unit cell configuration, wherein each of the first and the second sub-wavelength metal patches further has substantially an elliptical shape, it is possible to implement a nanoantenna where the plasmon resonances produce a broad combined resonance band covering the majority of the visible wavelength range . In a preferred embodiment of the thin film photovolta^¬ ic cell structure according to the present invention comprising a multi-resonance nanoantenna in the plas- monic light concentrator arrangement, the nanoantenna is a nanoantenna of any of the types and geometries described above. Preferably, the second dielectric layer of the thin film photovoltaic cell structure according to the pre^¬ sent invention is formed of plastic. The main ad^¬ vantages of plastic used in the second dielectric lay- er relate to the manufacturing aspects. Different plastics with suitable optical properties are availa^¬ ble with reasonably low prices, thus enabling cost- efficient manufacturing. Moreover, plastics are avail^¬ able in the form of large sheets suitable for simulta- neous manufacturing of a large number of individual solar cells. Suitable plastic materials are e.g. poly^¬ ethylene, polyamide, polymethyl methacrylate, polytet- rafluorethylene, and polystyrene. A third aspect of the present invention relates to a method for manufacturing a thin film photovoltaic cell structure. The method comprises the steps of: provid^¬ ing a first dielectric layer on a substrate; forming an active layer on the first dielectric layer for gen- erating free charge carriers via a photovoltaic ef^¬ fect; and forming a plasmonic light concentrator ar^¬ rangement on the active layer for coupling incident light at a first wavelength band into the active lay^¬ er. In forming the active region, principles and pro- cesses as such known in the field of thin film photo- voltaics can be used. The basic configuration of the thin film cell can be based on the conventional layer structure with superposed p- and n-type semiconductor layers forming a pn-junction and an active layer lo- cated between these two layers. Also, the basic con^¬ figuration of the cell can be a lateral one having the p- and n-type semiconductor regions at the sides of an active region formed within the active layer. Conventionally, plasmonic light concentrators are formed directly on the preliminary cell structure com^¬ prising the active layer. This means that the active layer and other elements of the preliminary cell structure have to be protected during the formation of the plasmonic light concentrator arrangement. Accord^¬ ing to the present invention, instead, the step of forming the plasmonic light concentrator arrangement on the active layer comprises forming the plasmonic light concentrator arrangement on a film of a dielec^¬ tric material which is transparent at the first wave^¬ length band and at a second wavelength band; and at- taching the plasmonic light concentrator arrangement together with the film on the active layer such that the film forms a second dielectric layer on the plas^¬ monic light concentrator arrangement. Thus, the plas^¬ monic light concentrator is first fabricated separate- ly from the actual photovoltaic core structure of the solar cell, and then attached to it. As a great ad^¬ vantage, said protection of the active layer and other elements of the preliminary cell structure is not needed .

Further, when fabricating the plasmonic light concentrator arrangement separately, very efficient manufac^¬ turing processes enabling low unit costs can be used. It is possible to print the plasmonic light concentra- tor arrangements of a great number of photovoltaic cells on a flexible film. Printing can be performed even in a roll-to-roll process. The principles of printing an array of metallic nanoantenna structures on flexible films or sheets are discussed e.g. by Kotter et al . in "Solar Nantenna Electromagnetic Col^¬ lectors", Proceedings of the 2nd International Confer^¬ ence on Energy Sustainability ASME August 10-14, 2008, Jacksonville, Florida, USA, paper ES 2008-54016; and by Kotter et al. in "Theory and Manufacturing Process- es of Solar Nanoantenna Electromagnetic Collectors", Journal of Solar Energy Engineering, (2010) 132: 011014. As another essential feature of the method according to the present invention, the first dielectric layer, the active layer, the plasmonic light concentrator ar^¬ rangement, and the second dielectric layer are config^¬ ured to form a resonant cavity for coupling incident light at the second wavelength band into a standing wave confined in the resonant cavity.

Preferably, forming the plasmonic light concentrator arrangement on the film of a dielectric material com^¬ prises forming a nanoantenna of any of the types and geometries described above.

The principles of, the advantages achieved by, as well as the preferred details of a photovoltaic cell struc^¬ ture thereby manufactured are discussed above in the context of the first aspect of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

Preferred embodiments of the present invention are de^¬ scribed in more detail in the following with reference to the accompanying figures, wherein

Figure 1 illustrates a thin film photovoltaic cell structure according to the present in^¬ vention ;

Figure 2 shows a plasmonic light concentrator arrangement of the thin film photovoltaic cell structure of Figure 1 ;

Figure 3 shows a nanoantenna forming a part of the plasmonic light concentrator arrange^¬ ment of Figure 2 ;

Figure 4 represents simulation results of a nanoantenna according to Figure 3;

Figure 5 shows a modified version of the nanoantenna geometry of Figure 3;

Figure 6 shows another nanoantenna suitable for a plasmonic light concentrator arrange- ment according to the present invention; and

Figure 7 illustrates manufacturing of a thin film photovoltaic cell structure according to the present invention. The figures are not in scale.

Figure 1 shows as a cross section a photovoltaic cell structure 1 comprising a wafer carrier 2 serving as a substrate and mechanical support for the photovoltaic cell structure, a first dielectric layer 3 on the wa^¬ fer carrier 2, and an active layer 4 on the first dielectric layer 3. The active layer 4 of the photovolta^¬ ic cell structure 1 can be formed, for instance, of p- doped silicon or p-doped gallium arsenide according to the known principles in the field of thin film photo- voltaics. Respectively, the material of the wafer car^¬ rier 2 can be any known substrate material suitable for the materials of the first dielectric layer and the active layer formed thereon.

In addition, there is a plasmonic light concentrator arrangement 5 on the active layer 4. The plasmonic light concentrator arrangement 5 is configured to op^¬ erate at a first wavelength band for coupling incident light at this wavelength band into the active layer 4. The photovoltaic cell structure 1 of Figure 1 further comprises a second dielectric layer 6 formed of a die^¬ lectric material, e.g. plastic, which is transparent at said first wavelength band to allow light at this wavelength band to enter the structure, and at a se- cond wavelength band.

According to a core principle of the present inven^¬ tion, the materials, layer thicknesses, and other fac^¬ tors affecting the optical performance of the overall layer structure of Figure 1, including the second die^¬ lectric layer 6, are selected to form a Fabry-Perot type standing wave resonant cavity for coupling inci^¬ dent light at a second wavelength band range differing at least partially from the first wavelength band into a standing wave confined in the resonant cavity. Thus, the photovoltaic cell structure 1 of figure 1 compris^¬ es two different arrangements for increasing the effi^¬ ciency of capturing light into the active layer 4. This combination of two different light coupling ar- rangements and thus two wavelength bands of enhanced light coupling into the active layer enables an en^¬ hanced light capturing efficiency over a broad wave^¬ length range. The actual selection of said materials, layer thicknesses, and other factors affecting the op- tical performance of the overall layer structure so that such a resonant cavity is formed can be per- formed, in general, according to common design princi^¬ ples known in the art.

In the following, the selection of the material of the first dielectric layer 3 is shortly discussed a bit more closely. As known for those skilled in the art of thin film optics, one key material parameter in opti^¬ cal layer structures is the permittivity ε. The needed permittivity and thickness of the first dielectric layer depends on the choice of the wafer carrier 2. For example, if the wafer carrier is prepared using the SOI technology (silicon on insulator) , then the wafer carrier is formed of low-quality amorphous sili^¬ con whose relative permittivity s_r in the frequency region corresponding to the wavelength range of 0.8 to 1.3 micron is close to 9 (see R Swanepoel, J. Phps . E: Sci. Instrum. Vol. 16, 1983), and the attenuation co^¬ efficient is relatively high (up to 1/cm) due to the lack of homogeneity. Simulations have shown that the optimum operation of the standing wave resonant cavity, producing the maximum of the field located inside the active layer 4, at wavelengths of 0.8 to 1 micron is achieved with this type of a wafer carrier when the relative permittivity of the first dielectric layer 3 is s_r=4, and the thickness thereof is 200-220 nm. Ma^¬ terials having such relative permittivity include e.g. Si02 and alumina.

If the target is to provide a large-area thin film so- lar cell with a flexible substrate formed of e.g. some polymer, the typical relative permittivity of such substrates of s_r=2...3, naturally, changes the optimum first dielectric layer material and thickness from the above examples.

It is a very important issue in the present invention that the broadband standing-wave resonator cavity can be obtained with just two simple and cheap dielectric layers with sub-micron thicknesses, i.e. the second dielectric layer 6 and the first dielectric layer 3, whereas in the prior art approaches reported in the literature either a pair of semi-transparent mirrors (nano-polished silver nano-film) or a pair of Bragg mirrors (periodic multilayers) is used. This simple solution of the present invention is possible because the photovoltaic active layer 4 between those two die- lectric layers has a very high relative permittivity (typically over 9) , and its optical contrast with the second and the first dielectric layers is high.

Concerning the word "layer" in the context of the first dielectric layer, it has to be noted that in the optical sense, the first dielectric layer operating as a part of the resonant cavity is actually not a true "layer". There is no internal light reflection from the bottom interface of this layer, i.e. from the in- terface between the first dielectric layer and the substrate. In that sense, the first dielectric layer can be considered as a semi-infinite medium.

The thin film photovoltaic cell structure 1 of Figure 1 represents a portion of a lateral cell configuration where the p- and n-type semiconductor regions (not shown in Figure 1) forming the pn-junction of the device are located laterally apart from each other and are in contact with the same side of the active layer 4. The basic principle of the present invention, i.e. the combination of those two types of light coupling arrangements, can be utilized in the more conventional vertical cell configuration as well, where the n- and p-type regions forming the pn-junction are arranged on different sides of the active layer. Generally, the plasmonic light concentrator arrange^¬ ment 5 of Figure 1 can be formed according to the known principles of plasmonic light concentrators. In the example of Figure 1, the plasmonic light concen- trator comprises an array of nanoantennas 7. Figure 1 is limited to present a portion of a photovoltaic cell structure substantially corresponding to the width of a unit cell 12 of one of the nanoantennas 7. As known in the art, the purpose of a nanoantenna integrated into a photovoltaic cell is to concentrate incident light energy into a hot spot 8 in the antenna gap 9. Due to the difference in the complex permittivity be^¬ tween the gap 9 and the active layer 4 of the photo^¬ voltaic cell, the center of the hot spot is displaced from the center of the gap towards the active layer.

Figure 2 shows an array of nanoantennas 7 formed on a flexible plastic sheet 10 made of, for instance, poly^¬ ethylene. The nanoantennas are formed by printing gold patches 11 on the plastic sheet 10.

Figure 3 shows a unit cell 12 of the nanoantennas 7 of the arrangement of Figure 2. The unit cell 12 compris^¬ es four gold patches 11 arranged as two pairs located around a central gap 9. Each pair lies on a connection line 13a, 13b crossing the central gap 9 so that the two gold patches 11 thereof are located centrally on opposite sides of the gap. The two imaginary connec^¬ tion lines 13a, 13b are directed oppositely to each other.

Each of the gold patches 11 has a shape of an isosce^¬ les trapezoid and is formed so as to have the shorter base of the trapezoid facing towards the central gap. The thickness d_A of the gold patches 11 can be 50 - 100 nm, e.g. about 75 nm. The width w_A of the square- form central gap 9 can be about 200 nm, for example. Suitable lengths for the shorter and the longer bases bi, b_∑ are 100 - 150 nm and 200 - 300 nm, respectively. The height h of the trapezoid preferably lies in the range of 300 - 500 nm. Thus, the overall width i¾ of the unit cell 12 thereby formed can be 800 - 1000 nm.

With the geometry of Figure 3, the nanoantenna produc^¬ es multiple resonance frequencies at near infrared wavelengths. A key factor to enable the multi- frequency operation is the presence of two pairs of patches 11 in the unit cell 12 of the antenna 7. The shape of the patches 11, in turn, contributes to the resonant operation at infrared wavelengths. Figure 4 shows simulated local field intensity en^¬ hancement in the gap of a nanoantenna according to Figure 3. Graph a) represents the simulated results at 850 nm, whereas graph b) shows the corresponding re^¬ sults for 980 nm. Simulations were performed in the HFSS package (High-Frequency Structure Simulation) for normal incidence of light. Figures a) and b) clearly prove that the nanoantenna operates as intended, i.e. by effectively enhancing the intensity in the gap. For those two wavelengths, the intensity distributions are similar but enhancement factors are different. Maximal intensity enhancement factor in the structure is achieved at 980 nm and is close to 225 (amplitude en^¬ hancement 15) . The resonance frequencies according to the nanoantenna 7 of Figure 3 can be adjusted by modifying the dimen^¬ sions of the nanopatches 11 and the central gap 9. On the other hand, for each material of the patches 11, the dimensions required for a particular resonant fre- quency are unique. The sharp edges of the patches 11 of Figure 3 are not necessary, but the edges can be smoothed as well with^¬ out substantially deteriorating the optical perfor^¬ mance of the nanoantenna 7. Moreover, it has been found that by making the sides of the trapezoids con^¬ vex or concave, the resonant frequencies are shifted towards infrared or visible wavelengths, respectively. As an example of the latter, Figure 5 presents a nano^¬ antenna geometry modified from that of Figure 3.

The geometry of Figure 5 differs from that of Figure 3 in that the sides of the trapezoids are recessed to make the shape of the trapezoid concave. For one par^¬ ticular geometry with the gap width w_A of 200 nm, the lengths of the shorter and the longer bases bi, b_∑ of 100 nm and 400 nm, respectively, and the width bj of the trapezoid at the midpoint between the shorter and the longer bases of 300 nm, simulations showed a shift of the central operational frequency of the nanoanten- na 7 of from 270 - 300 THz to 200 - 240 THz, depending on the material of the film 10 on which the nanoantenna 7 is formed. This corresponds to a shift in the wavelength of from 1.00 - 1.11 to 1.25 to 1.50 ym. In the simulated geometry, the unit cell width i¾ was about 800 nm. This high operation wavelength with such a low size of the nanoantenna is a very interesting combination not achieved, according to the best knowledge of inventor, by any prior art solution. As an alternative to the two pairs of trapezoids of the unit cell 12 of Figure 3, Figure 6 shows a nanoan^¬ tenna 14 where each of the gold patches 151, 152 have the form of an ellipse. The nanoantenna 14 of Figure 6 is an example of nanoantenna geometry providing multi- pie resonances in the visible wavelength range. As in the geometry of Figure 3, also the elliptical gold patches 151, 152 of Figure 6 are arranged as pairs around a central gap 16, each pair lying on one of two fictitious, orthogonally directed connection lines 13a, 13b crossing the central gap 16. The longi^¬ tudinal axis of symmetry 171, 172 of each patch 151, 152 is directed perpendicularly with respect to the connection line on which the patch lies. Instead of two, there are four pairs of patches 151, 152 in the nanoantenna 14 of Figure 6. The patches 151, 152 have different sizes so that patches 151 of a first size lie adjacent to the central gap 16, thereby forming an inner circle around it, whereas patches 152 with a larger second size lie behind the patches 151 of the first size.

The elliptical gold patches 151, 152 of Figure 6 can have a height d_B of 20 - 50 nm. Suitable lengths li, 1_∑ of the smaller and larger patches are 100 - 150 nm and 200 - 250 nm, respectively. The widths m_lr iri2 of the smaller and larger patches can be in the ranges of 25 - 50 and 50 - 100, respectively. With these dimen^¬ sions, the overall width W_B of the unit cell 18 can be in the range of 500 - 600 nm, for example.

The elliptical shape of Figure 6 is only one example of possible shapes of elongated patches for a nanoan^¬ tenna operating at visible wavelengths. Also other elongated shapes having a longitudinal axis of sym- metry could be used.

It is essential in the nanoantenna geometries of Fig^¬ ures 3, 5, and 6 that the different patches 11, 151, 152 of a unit cell 12, 18 are truly separated from each other. The plasmonic light concentrator arrangement 5, 7 ac^¬ cording to Figures 1 and 2, as well as plasmonic light concentrator arrangements based on, for example, nano- antennas 7, 14 according to Figures 5 and 6, can be manufactured cost-efficiently as large-area sheets. Printing the gold patches 11 forming the nanoantennas 7 can be performed e.g. in a roll-to-roll process. To be used as a part of a thin film photovoltaic cell structure, this sheet with the nanoantennas 7 can then be cut into suitably sized elements according to the size of individual photovoltaic cells or of an array of cells.

Figure 7 illustrates the principle of incorporating the plasmonic light arrangement 5, 7 thereby formed as a part of a photovoltaic cell structure 1. As in Fig^¬ ure 1, also the drawings of Figure 7 are limited to present portions of a photovoltaic cell structure (1) corresponding to the width of a unit cell of one of the nanoantennas 7 only.

In the embodiment of Figure 7, the volumes 19 between the adjacent patches of the nanoantenna 7 are left empty (air-filled) , whereby they form thin cavities in the final photovoltaic cell structure. It is also pos^¬ sible to fill those volumes with some suitable materi^¬ al, if appropriate in some particular embodiment.

In the process illustrated in Figure 7, a first die- lectric layer 3 made of a dielectric material is first formed on a wafer carrier 2, and an active layer 4 is formed on the first dielectric layer. The plastic sheet 10 having the nanoantennas 7 thereon is flipped over and placed on top of the active layer 4 upside down, i.e. with the side of the nanoantennas 7 towards the active layer 4. When the superstrate of the plas^¬ tic sheet 10 and the nanoantennas 7 thereon is placed on the underneath layer stack by slightly pressing it against the active layer, self-adhesion occurs between the superstrate 10, 7 and the active layer 4. The thickness of the nanoantennas 7, which can be e.g. in the range of 50 to 100 nm, is roughly in the range of one tenth of the thickness of the plastic sheet 10. Therefore, the effect of the air gaps in the volumes 19 on the adhesion between the superstrate 10, 7 and the active layer 4 is, in most cases, insignificant. If appropriate in some particular embodiment, it is also possible to use some intermediate adhesive mate^¬ rial between the active layer 4 and the superstrate 10, 7 to ensure sufficient adhesion therebetween. Moreover, the plastic sheet 10 can be additionally fixed to the underneath active layer 4 along the pe^¬ rimeter of the cell area.

As a great advantage of separate fabrication of the nanoantennas 7 on the plastic sheet 10, no protective intermediate layer is needed to protect the active layer as is the case if growing or depositing plas- monic light concentrators directly on the active lay^¬ er . In the thereby formed photovoltaic cell structure 1, the plastic sheet 10 forms the second dielectric lay- er6 of the structure. After attaching the plastic sheet 10 having the nanoantennas 7 on the active layer 4, the thickness thereof can be reduced to optimize the operation of the resonant cavity. Suitable thick^¬ ness of the second dielectric layer formed e.g. of polyethylene can be e.g. in the range of 300 - 800 nm. In many embodiments, the most suitable thickness is 600 - 800 nm. The embodiments of the present invention are not lim^¬ ited to examples discussed above but can freely vary within the scope of the claims.

Claims

1. A thin film photovoltaic cell structure (1) com^¬ prising :

- a substrate (2);

- a first dielectric layer (3) on the sub^¬ strate (2 ) ;

- an active layer (4) on the first dielectric layer (3) ; and

- a plasmonic light concentrator arrangement (5, 7, 14) on the active layer (4) for cou^¬ pling incident light at a first wavelength band into the active layer (4);

characteri zed in that the thin film photovolta^¬ ic cell structure (1) further comprises a second die- lectric layer (6) formed of a material which is trans^¬ parent at the first wavelength band and at a second wavelength band on the plasmonic light concentrator arrangement (5, 7, 14); the first dielectric layer (3), the active layer (4), the plasmonic light concen- trator arrangement (5, 7, 14), and the second dielec^¬ tric layer (6) being configured to form a resonant cavity for coupling incident light at the second wave^¬ length band into a standing wave confined in the reso^¬ nant cavity.

2. A thin film photovoltaic cell structure (1) as de^¬ fined in claim 1, wherein the first and the second wavelength bands are adjusted to the near infrared and to the visible ranges of the electromagnetic spectrum, respectively.

3. A thin film photovoltaic cell structure (1) as de^¬ fined in claim 1 or 2, wherein the plasmonic light concentrator arrangement (5) comprises a nanoantenna (7, 14) for concentrating the incident light at the first wavelength band into a spot (8) extending into the active layer (4) .

4. A thin film photovoltaic cell structure (1) as de^¬ fined in claim 3, wherein the nanoantenna (7, 14) is configured to provide multiple plasmon resonance wave- lengths within the first wavelength band.

5. A nanoantenna (7, 14) suitable for a plasmonic light concentrator arrangement (5) of a thin film pho^¬ tovoltaic cell structure (1) as defined in claim 4, wherein the nanoantenna (7, 14) comprises a unit cell (12, 18) comprising two pairs of first sub-wavelength metal patches (11, 151) placed around a central gap (9, 16) separate from each other, the two first patch^¬ es of each pair lying on opposite sides of the central gap (9, 16) centrally located on a straight connection line (13a, 13b) crossing the central gap (9, 16), the connection lines (13a, 13b) of the two pairs of first sub-wavelength metal patches (11, 151) being directed substantially perpendicularly to each other.

6. A nanoantenna (7, 14) as defined in claim 5, where^¬ in the unit cell width (i¾, W_B) as defined in the di^¬ rection of any of the connection lines (13a, 13b) is less than or equal to 1 ym.

7. A nanoantenna (7, 14) as defined in claim 5 or 6, wherein the central gap width (w_Ar w_B) as defined in the direction of any of the connection lines (13a, 13b) is 50 - 300 nm.

8. A nanoantenna (7) as defined in any of claims 5 to 7, wherein each of the first sub-wavelength metal patches (11) has substantially a shape of an isosceles trapezoid, the shorter base of the trapezoid facing towards the central gap (9) .

9. A nanoantenna (7) as defined in claim 8, wherein the thickness (d_A) of each of the first sub-wavelength metal patches (11) is formed of gold and has a thick^¬ ness (d_A) of 50 - 100 nm; the width (w_A) of the central gap (9) along any of the connection lines (13a, 13b) is 150 - 250 nm; the lengths (bi, b_∑) of the shorter and longer bases of the trapezoid are 75 - 150 nm and 200 - 450 nm, respectively; and the height (h) of the trapezoid is 300 - 500 nm.

10. A nanoantenna (14) as defined in any of claims 5 to 7, wherein each of the first sub-wavelength metal patches (151) has an elongated shape having a longitu^¬ dinal axis of symmetry (171) which is directed sub- stantially perpendicularly with respect to the connec^¬ tion line (13a, 13b) on which the first sub-wavelength metal patch (151) lays.

11. A nanoantenna (14) as defined in claim 10, wherein the unit cell (18) further comprises, for each of the first sub-wavelength metal patches (151), a second elongated sub-wavelength metal patch (152) lying centrally located on the same connection line (13a, 13b) on which the first sub-wavelength metal patch (151) lies on the side of the first sub-wavelength metal patch (151) opposite to the central gap (16), and hav^¬ ing a longitudinal axis of symmetry (172) which is di^¬ rected perpendicularly with respect to this connection line .

12. A nanoantenna (14) as defined in claim 11, wherein each of the first and the second patches (151, 152) is formed of gold and has a thickness (d_B) of 20 - 50 nm; the width (w_B) of the central gap (16) along any of the connection lines (13a, 13b) is 100 - 150 nm; the lengths (li, 1_∑) of the first and the second sub- wavelength metal patches (151, 152) along the longitu- dinal axis of symmetry (171, 172) thereof are 100 - 150 nm and 200 - 250 nm, respectively; and the widths (m_lr ΠΙ2) of the first and the second sub-wavelength metal patches (151, 152) in a direction perpendicular to the longitudinal axis of symmetry (171, 172) there^¬ of are 25 - 50 nm and 50 - 100 nm, respectively.

13. A nanoantenna (14) as defined in claim 12, wherein each of the first and the second sub-wavelentgh metal patches (151, 152) has substantially an elliptical shape .

14. A thin film photovoltaic cell structure (1) as de^¬ fined in claim 4, wherein the nanoantenna (7, 14) is a nanoantenna as defined in any of claims 5 to 13.

15. A thin film photovoltaic cell structure (1) as de^¬ fined in any of claims 1 to 4 and 14, wherein the se^¬ cond dielectric layer (6) is formed of plastic (10) .

16. A method for manufacturing a thin film photovoltaic cell structure (1) comprising the steps of:

- providing a first dielectric layer (3) on a substrate (2 ) ;

- forming an active layer (4) on the first dielectric layer (3) ; and

- forming a plasmonic light concentrator arrangement (5, 7, 14) on the active layer (4) for coupling incident light at a first wave- length band into the active layer (4); ch a r a c t e r i z e d in that the step of forming the plasmonic light concentrator arrangement (5, 7, 14) on the active layer (4) comprises forming the plasmonic light concentrator arrangement (5, 7, 14) on a film (10) of a dielectric material which is transparent at the first wavelength band and at a second wavelength band; and attaching the plasmonic light concentrator arrangement (5) together with the film (10) on the ac^¬ tive layer (4) such that the film (10) forms a second dielectric layer (6) on the plasmonic light concentra^¬ tor arrangement (5, 7, 14); wherein the first dielec^¬ tric layer (3), the active layer (4), the plasmonic light concentrator arrangement (5, 7, 14), and the se^¬ cond dielectric layer (6) are configured to form a res^¬ onant cavity for coupling incident light at the second wavelength band into a standing wave confined in the resonant cavity.

17. A method according to claim 16, wherein forming the plasmonic light concentrator arrangement (5, 7, 14) on a film (10) of a dielectric material which is transparent at the first wavelength band and at a se^¬ cond wavelength band comprises forming a nanoantenna (7, 14) according to any of claims 3 to 13.