WO2013009857A1

WO2013009857A1 - Translucent solar cell

Info

Publication number: WO2013009857A1
Application number: PCT/US2012/046234
Authority: WO
Inventors: Richard L. MCCLARY; Charles Y. TOMITA; Anthony C. VASKO
Original assignee: The University Of Toledo; Solar Spectrum, Llc
Priority date: 2011-07-11
Filing date: 2012-07-11
Publication date: 2013-01-17

Abstract

A solar cell includes a substrate layer formed from a transparent or partially transparent material, a first electrode layer engaging the substrate layer and formed from a transparent or partially transparent material, a first semiconductor layer engaging the first electrode layer and having a thickness in a range of from about 50 nm to about 120 nm that allows a predetermined amount of light to pass therethrough, and a second semiconductor layer engaging the first semiconductor layer and having a thickness in a range of from about 200 nm to about 700 nm that allows a predetermined amount of light to pass therethrough. A second electrode layer engages the second semiconductor layer and includes a first sheet having a generally uniform thickness that allows a predetermined amount of light to pass therethrough and a second sheet having a grid-like structure with at least one space or opening that allows light to pass therethrough.

Description

TITLE

TRANSLUCENT SOLAR CELL

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001 ] This application claims the benefit of United States Provisional Application No. 61/572,120, filed July 11, 2011, the disclosure of which is incorporated herein by reference.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was not made with any government support.

BACKGROUND OF THE INVENTION

[0003] The state of the art in photovoltaic technology results in solar panels that allow little to no visible light transmission therethrough. Part of this is due to the relatively large thicknesses of the various layers typically used in conventional solar cells. For example, as discussed in a paper by Wu et al. in the Proceedings Of The 17th European Photovoltaic Science and Engineering Conference (2001), a longstanding efficiency record (at 16.5% efficient) for a cadmium sulfide (CdS)/cadmium telluride (CdTe) device included a layer of CdTe that was approximately 10 microns in thickness. A CdTe layer having such a relatively large thickness is essentially opaque to sunlight.

[0004] From the National Renewable Energy Laboratory (0-7803-1220- 1/93 IEEE), there is a report of a 13.4% efficient CdTe solar cell in "The U.S. DOE/NREL

Polycrystalline Thin Film Photo voltaics Project" by Zweibel, K., Ullal, H.S., Mitchell, R.L., and Noufi, R. (Photovoltaic Specialists Conference, 1991, Conference Record of the Twenty Second IEEE, Oct. 7-11, 1991, Pages 1057-1061, vol. 2). Conventional CdTe solar cells having efficiencies of less than 16.5% have been also reported in "Perspectives And Opportunities In Polycrystalline Thin Film Photovoltaic Technologies" by Zweibel, K., Ullal, H.S., von Roedern, B.G., Noufi, R., Coutts, T.J., and Al-Jassim, M.M.

(Photovoltaic Specialists Conference, 1993, Conference Record of the Twenty Third IEEE, May 10-14, 1993, Pages 379-388). General trends have been recognized as stated by Akhlesh Gupta et al. in his paper on "Effect Of CdTe Thickness Reduction In High Efficiency CdS/CdTe Solar Cells" in Materials Research Society Proceedings, 668 , H6.4 doi: 10.1557/PROC-668-H6.4 (2001).

[0005] Additionally, to collect the photon energy in typical cell designs, at least one solid layer of highly conductive material is used to effectively collect the electrical current. However, such a solid layer of highly conductive material can make the solar cell opaque to the transmission of sunlight therethrough, regardless of the thickness of the semiconductor layer.

[0006] Attempts have been made to develop solar panels that are translucent to sunlight with limited success. For example, Parikh et al. (MRS 2007) published data on cells using transparent conducting oxide back contacts, but these cells were intended to be a top layer in a tandem structure and not a stand-alone translucent structure.

Moreover, it was discovered that the oxide layers could form a backward field opposing the main junction, an issue that might be difficult to overcome in industrial mass production.

[0007] As can be seen, there remains a great need to produce translucent solar cells at low cost with high efficiency that require small amounts of critical raw materials.

SUMMARY OF THE INVENTION

[0008] This invention is a multiple layer CdTe solar cell that produces electricity and is translucent. This invention optimizes translucent properties with high electrical efficiencies and fill factors. This invention uses several unique designs in the cell construction to solve the translucent, photovoltaic efficiency, and low cost parameters necessary for a useful translucent product. [0009] One approach of this invention, as described herein, is to provide a back contact for solar cell that combines a highly conductive grid layer that collects electrical current from a less conductive, uniform-thickness layer, wherein the less conductive, uniform- thickness layer is effective to not only make good ohmic contact with an absorbing semiconductor layer, but is sufficiently thin as to allow light to pass therethrough in an amount that is suitable for human viewing, even after it has been passed through (and partially absorbed) in the active semiconductor materials. An optimization range is provided that balances the electrical efficiency and the visual efficiency by adjusting the uniform conductive layer and the grid structure. This invention also contemplates deviating from a typical uniform-thickness conductive layer to a grid-like layer construction for increasing the amount of light that can pass therethrough.

[0010] This invention is particularly suited to applications such as solar energy- producing, translucent skylights for warehouses and homes, commercial building windows that allow light to pass into the building and produce energy at the same time, greenhouses that allow plant growth and provide their own electricity for climate control, and automobile windows for powering fans and air conditioners. It has been found that translucent solar cells according to this invention can obtain efficiencies of greater than 10 percent at the individual cell level and still allow light transmission therethrough.

[0011] Various aspects of this invention will become apparent to those skilled in the art from the following detailed description of the preferred embodiments, when read in light of the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Fig. 1 is a schematic perspective view of a portion of a solar cell in accordance with this invention.

[0013] Fig. 2 is a schematic sectional view of the solar cell illustrated in Fig. 1.

[0014] Fig. 3 is an end elevational view of an alternative embodiment of the solar cell illustrated in Figs. 1 and 2. [0015] Fig. 4 is a chart illustrating a CIE pho topic luminous efficiency function, which quantifies the sensitivity of a human eye under well-lit conditions.

[0016] Fig. 5 is a chart illustrating the irradiance of the solar spectrum.

[0017] Fig. 6 is a chart illustrating a transparent figure of merit for the solar cell, shown in comparison with other common light sources.

[0018] Fig. 7 is a chart illustrating the effect of a 5 nm thick gold layer on the transparent figure of merit illustrated in Fig. 6.

[0019] Fig. 8 is a chart illustrating an estimate of the efficiency of the solar cell as a function of thickness of the CdTe layer in an individual section thereof.

[0020] Fig. 9 is a chart illustrating the total figure of merit for a solar cell that includes the effects of both transparency and power efficiency.

[0021] Fig. 10 is a chart illustrating an example of the use of a solar cell in accordance with this invention in combination with a compact fluorescent light (CFL) in the same room.

[0022] It will be understood that the figures are for the purpose of illustrating the concepts of the invention and may not be to scale.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] Referring now to the drawings, there is schematically illustrated in Figs. 1 and 2 a portion of a solar cell, indicated generally at 10, in accordance with this invention. As discussed above, the solar cell 10 is responsive to light energy for generating an electrical current. At the same time, however, the solar cell 10 of this invention allows some of the light that is not used to generate the electrical current to pass therethrough for the sake of illumination. As such, the solar cell 10 can be thought of as a window that allows light to pass therethrough to illuminate, for example, an interior of a structure on which the solar cell 10 is supported. The solar cell 10 of this invention may be either translucent or transparent to the passage of light therethrough, and the word "translucent" is used herein to indicate both of these characteristics. [0024] The illustrated solar cell 10 comprises seven layers of material having an overall thickness that is preferably in the range of from about 838.5 nm to about 1532 nm.

However, as will be explained below, the solar cell 10 may including a greater or lesser number of layers of material, and the overall thickness of the solar cell 10 may be greater or less than this range.

[0025] The illustrated solar cell 10 includes a substrate layer 11 that is preferably formed from a transparent or partially transparent material. The substrate layer 11 may, for example, be formed from at least one layer of a transparent glass including, but not limited to, commercial soda lime glass. The substrate layer 11 allows incident light to pass therethrough to the other layers of the solar cell 10 that are described below. The substrate layer 11 also provides structural stability to the solar cell 10. The thickness of the substrate layer 11 may be selected as desired to accomplish these and other functions.

[0026] The illustrated solar cell 10 also includes a layer 12 that is formed from a transparent conducting oxide (TCO) material. The TCO layer 12 functions as a first (negative) electrode that can be connected to a first electrical lead 12a (see Fig. 2) extending from the solar cell 10, while allowing incident light to pass therethrough to the other layers of the solar cell 10 described below. The TCO layer 12 is preferably formed from a material that exhibits good electrical conductance, high light transmission, and minimal visible light reflection. For example, the TCO layer 12 may be formed from indium tin oxide or fluorine-doped tin oxide, such as is commercially available under the trade name TEC Glass from Pilkington North America Inc. of Toledo, Ohio. The TCO layer 12 may, if desired, be comprised of a plurality of sub-layers of such material. In one embodiment of this invention, the TCO layer 12 can have a thickness in the range of from about 300 nm to about 900 nm.

[0027] Those having skill in the art will recognize that commercially available materials used to form the TCO layer 12 may include one or more intrinsic layers of a high resistance transparent (HRT) material. The properties of such HRT materials will be described in further detail below. Thus, in one embodiment of this invention, the TCO layer 12 can include at least one intrinsic layer of a HRT material. In another embodiment of this invention, the TCO layer 12 can be formed from silicon dioxide.

[0028] The solar cell 10 may additionally include a layer 13 of an HRT material, although such is not required. The HRT layer 13 primarily serves as a buffer between the semiconductor layers of the solar cell 10 (which are described in detail below) and other layers of the solar cell 10 to control diffusion and performance loss. The HRT layer 13 is preferably formed from a material that exhibits relatively high electrical resistance and good visible light transmission such as, for example, silicon dioxide, or intrinsic tin oxide or zinc oxide. In one embodiment of this invention, the HRT layer 12 is formed from zinc oxide having a thickness in the range of from about 80 nm to about 120 nm. The HRT layer 12 may also be formed from more than one sub-layer if desired.

[0029] The illustrated solar cell 10 further includes a first semiconductor layer 14 and a second semiconductor layer 15. Preferably, the first semiconductor layer 14 is preferably formed from an n-type heterojunction material that serves as a partner for the second semiconductor layer 15, which is preferably formed from a p-type photoelectric material. Thus, the first semiconductor layer 14 functions as a main junction effect between the TCO layer 12 and the second semiconductor layer 15. Preferably, the first semiconductor layer 14 is formed from CdS. The thickness of the semiconductor layer 14 depends upon the properties of the HRT layer 13 and the overall size of the solar cell 10. In one embodiment of this invention, the first semiconductor layer 14 has a thickness in the range of from about 50 nm to about 200 nm. Thinning of the first semiconductor layer 14 can result in a greater short circuit current. However, over-thinning can decrease the generation of voltage in the solar cell 10 and, thus, may negatively impact

manufacturing yield, especially in larger-area cells due to the probability of a cell containing a poor region of CdS.

[0030] The second semiconductor layer 15 is preferably formed from a p-type photoelectric material that is compatible with the material of the first semiconductor layer 14 such as, for example, CdTe. The second semiconductor layer 15 serves as a light energy-absorbing layer and is optimized for transparency, while retaining good photoelectric efficiency. In one embodiment of this invention, the second semiconductor layer 15 has a thickness in the range of from about 100 nm to about 700 nm.

[0031] Next, the illustrated solar cell 10 includes an ohmic connection layer 16. The ohmic connection layer 16 provides an ohmic connection with the second semiconductor layer 15. Preferably, the ohmic connection layer 16 is formed from a material that is a good electrical conductor and has good adherence to the second semiconductor layer 15. The ohmic connection layer 16 may also have a suitable work function or Fermi level to form an ohmic connection to the second semiconductor absorber layer 15. For example, the ohmic connection layer 16 can be formed from a metallic material (such as, for example, gold, chromium, etc.) or from a combination of metal metallic materials. In one embodiment of this invention, the ohmic connection layer 16 is formed from a bi-layer of copper (having a thickness in the range of from about 0.5 nm to about 3.0 nm) and gold (having a thickness in the range of from about 3.0 nm to about 10.0 nm).

[0032] Lastly, the illustrated solar cell 10 includes a supplemental electrode layer 17 having a grid-like structure. The term "grid-like structure" is used herein to indicate any structure that includes one or more first portions that are separated from one or more second portions by one or more spaces or openings. For example, Figs. 1 and 2 illustrate a first embodiment of a grid-like structure for the supplemental electrode layer 17 that is defined by a plurality of discontinuous portions 17a that are generally rectangular in cross sectional shape and are separated from one another by spaces in both the length- wise and width- wise directions. Fig. 3 illustrates a second embodiment of a grid-like structure for the supplemental electrode layer 17 that is defined by a plurality of continuous portions 17b that are generally rectangular in shape (when viewing Fig. 3) and are separated from one another by openings in both the length- wise and width- wise directions.

[0033] The grid-like structure of the supplemental electrode layer 17 may be formed having any desired shape or combination of shapes including non-linear lines (curved or wavy lines, for example) and non-rectangular orientations. For reasons that will become apparent below, the sizes of the spaces or openings will preferably be relatively large in comparison with the sizes of the various portions of the grid-like structure. However, the sizes of the spaces or openings may have any desired size relative to the various portions of the grid-like structure. Similarly, the portions 17a and 17b of the grid-like structure of the supplemental electrode layer 17 may have any desired cross-sectional shape, such as square, rectangular, circular, and the like. The ohmic connection layer 16 and the supplemental electrode layer 17 together function as a second (positive) electrode that can be connected to a second electrical lead 16a (see Fig. 2) extending from the solar cell 10, while allowing incident light to pass therethrough.

[0034] As will be explained in greater detail below, the grid-like structure of the supplemental electrode layer 17 allows the thickness of the ohmic connection layer 16 to be sufficiently small as to allow at least some visible light to pass therethrough. Such light then can pass through the spaces or openings of the grid-like structure of the supplemental electrode layer 17. The thickness of the supplemental electrode layer 17 may vary in accordance with the sizes of the spaces or openings of the grid-like structure of the supplemental electrode layer 17 or any other desired parameter. Preferably, the supplemental electrode layer 17 is preferably formed from a material that is a good electrical conductor. For example, the supplemental electrode layer 17 can be formed from a metallic material such as, for example, gold, or from a combination of metal metallic materials. In one embodiment of this invention, the supplemental electrode layer 17 may be formed from gold having a thickness in the range of from about 25 nm to 100 nm.

[0035] The supplemental electrode layer 17 can be formed integrally with the ohmic connection layer 16 if the material used to form the supplemental electrode layer 17 is the same as the material used to form the ohmic connection layer 16. For example, a single layer of material (such as copper, gold, chromium, etc., as described above, or a combination thereof) may be applied to the second semiconductor layer 15 having a thickness that is equal to the desired combined thicknesses of the ohmic connection layer 16 and the supplemental electrode layer 17. Thereafter, portions of this uniform thickness layer may be removed so as to provide relatively small thickness portions (where the material has been removed) that define the ohmic connection layer 16 and relatively large thickness portions (where the material has not been removed) that define the grid-like supplemental electrode layer 17. The removal of the material can be accomplished by any desired method such as, for example, laser scribing or chemical etching using a conventional mask to cover the portions of the material that are not desired to be removed.

[0036] On the other hand, the supplemental electrode layer 17 can be applied to the ohmic connection layer 16 after the ohmic connection layer 16 has been applied to the second semiconductor layer 15. For example, the supplemental electrode layer 17 can be applied to the ohmic connection layer 16 by means of a conventional masking process, wherein the exposed surface of the ohmic connection layer 16 is covered by a mask having one or more openings formed therethrough. These openings are sized and shaped in accordance with the desired size and shape of the grid-like shape of the supplemental electrode layer 17. Then, the material used to form the supplemental electrode layer 17 is applied to the mask, but only the material in the regions of the openings is allowed to pass therethrough into contact with the ohmic connection layer 16. If desired, multiple layers of different materials may be applied to the ohmic connection layer 16 in this manner (for example, a stack of chromium, aluminum, and chromium layers).

[0037] Alternatively, a layer of material (such as copper, gold, chromium, etc., as described above, or a combination thereof) may be applied to the ohmic connection layer 16 having a uniform thickness that is equal to the desired thickness of the supplemental electrode layer 17. Thereafter, portions of this uniform thickness layer may be removed so as to provide relatively small thickness portions (where the material has been removed) that expose the ohmic connection layer 16 and relatively large thickness portions (where the material has not been removed) that define the grid-like supplemental electrode layer 17 overlying the ohmic connection layer 16.

[0038] Regardless of how they are formed, the ohmic connection layer 16 and the supplemental electrode layer 17 function as a second, positive electrode for the solar cell 10. This second electrode includes first and second regions of different electrical resistivity. The first resistivity region is defined by the portions of the ohmic connection layer 16 that are not covered by the grid-like structure of the supplemental electrode layer 17. The second resistivity region is defined by the portions of the ohmic connection layer

16 that are covered by the grid-like structure of the supplemental electrode layer 17. Thus, it will be appreciated that the first resistivity regions of the second electrode are relatively thin in comparison to the second resistivity regions thereof.

[0039] For the sake of explanation, let:

Rl equal a first sheet resistance Rl defined by the first resistivity regions of the second electrode;

x equal the width of such first resistivity regions;

R2 equal a second sheet resistance R2 defined by the second resistivity regions of the second electrode;

y equal the width of such second resistivity regions.

In the illustrated embodiment, the grid-like structure of the supplemental electrode layer

17 is arranged perpendicular to the dividing line between the individual layers 11-17 of the solar cell 10, the second sheet resistance R2 is less than the first sheet resistance Rl, and the width of the second resistivity regions y is less than the width of the first resistivity regions x. However, as noted above, this invention may be practiced having other geometries for the grid-like structure.

[0040] The values of Rl and R2 can be adjusted in several ways. For example, the values of Rl and R2 can be adjusted using materials of different resistivity (such as TCO and aluminum materials, for example). Another way is to use materials of the same composition but having different thicknesses because, for a wide range of thicknesses, the overall resistivity of the sheet is equal to the intrinsic resistivity of the material divided by the thickness of the layer (in the direction perpendicular to the page in those drawings).

[0041] For the solar cell 10 to work efficiently, the grid-like structure should collect all of the electrical current from the first resistivity regions of the second electrode. More precisely, if R2 is low enough that losses in the grid-like structure may be neglected, the relative efficiency (1 being the case of no resistive losses) caused by the resistance R2 is (using an ideal diode model) can be calculated as: S'VQC

[0042] or even more precisely (using a semi-empirical correction and a more accurate diode model) as:

( 2- ^' ¾ϋ - VQC $ R 1 -JSC -X*

i

[0043] with V_pP being a parameter that is approximately the voltage at a maximum power point (note that if Vpp = Voc, the "precise" equations all reduce to the ideal equations).

[0044] These equations are valid in the case where the loss is due to decreasing maximum power point voltage and fill factor, but the current is still fully collected.

[0045] In the case where no grids are present, the limiting (minimum) voltage at maximum power point is Voc/2. However, with grids, the limiting maximum power point is 2/3· Voc (ideal diode) or (Voc+Vpp)/3 (more accurate case). This leads to: x < 2 · I ^{2 ! q}^ (ideal case)

[0046] or: x <. I · — -—— (more precise)

[0047] No non-zero value of Rl will result in zero loss of relative efficiency, but Rl can be chosen so that the resistive losses are small and acceptable.

[0048] For example, for CdTe, round estimates of Jsc = 20 mA/cm2 and Voc = 0.8V may be used and, for the ohmic connection layer 16 to be as transparent as desired, the sheet resistance may be 3000 ohm. If it is assumed that 90% relative efficiency is desired, then the ideal equation is solved as:

0.9 = 1-R * Jsc * x² / (8 * Voc) = 1 - 9.4x²/cm²

[0049] From that equation, x can be calculated as 0.1cm. Thus, for this example, the grid lines of the supplemental electrode layer 17 will be about 1 mm apart. The condition that:

[0050] is χ < 0.188 cm, which is satisfied by x = 0.1cm.

[0051] Referring back to Fig. 1, it can be seen that one or more first discontinuities 21 are formed in the illustrated TCO layer 12 and the HRT layer 13. The illustrated first discontinuities 21 extend generally perpendicular to the portions 17a that form the gridlike structure of the supplemental electrode layer 17, although such is not required. The illustrated first discontinuities 21 are filled with material from the first semi-conductor layer 14 although, again, such is not required. Similarly, one or more second

discontinuities 22 are formed in the illustrated first semi-conductor layer 14 and the second semi-conductor layer 15. The illustrated second discontinuities 22 also extend generally perpendicular to the portions 17a that form the grid-like structure of the supplemental electrode layer 17, although such is not required. The illustrated second discontinuities 22 are filled with material from the ohmic connection layer 16 although material from the supplemental electrode layer 17 may be used.

[0052] Lastly, one or more third discontinuities 23 are formed in the illustrated ohmic connection layer 16. The illustrated third discontinuities 23 also extend generally perpendicular to the portions 17a that form the grid-like structure of the supplemental electrode layer 17, although such is not required. A light diffraction blocking material 24 can be provided adjacent to the third discontinuities 23. In the illustrated embodiment, the third discontinuities 23 are filled with a light diffraction blocking material 24.

Alternatively, the third discontinuities 23 can remain unfilled, and a layer of a light diffraction blocking material (not shown) may be applied to the substrate layer 11 opposite the illustrated third discontinuities 23. The purpose of the light diffraction blocking material is to eliminate relatively bright regions of the solar cell 10 that would otherwise be created by the absence of material caused by the third discontinuities 23 in the ohmic connection layer 16. Such relatively bright regions might be considered undesirable from an aesthetic standpoint by a person looking through the solar cell 10. The use of the light diffraction blocking material (either within the third discontinuities 23 as illustrated or on the substrate layer 11 opposite the illustrated third discontinuities 23) has been found to ameliorate these relatively bright regions and does not significantly affect the overall translucency of the solar cell 10. However, it will be appreciated that the use of the light diffraction blocking material for this purpose is optional.

[0053] The discontinuities 21, 22, and 23 are provided to essentially divide the solar cell 10 into a plurality of relatively small electrical-current producing regions that are connected in series with one another. As a result, the solar cell 10 produces a relatively large amount of voltage, but a relatively small amount of current. The total amount of electrical current that is generated by the solar cell 10 is equal to the smallest amount of electrical current that is produced in any one of the electrical-current producing regions thereof. Typically, it is desirable that the total amount of electrical current that is generated by the solar cell 10 be relatively small in order to reduce the amount of electrical resistance losses. However, the overall electrical power output of the solar cell 10 can still be large, however, because the electrical voltages that are produced by the electrical-current producing regions are added across the entire solar cell 10. Because of the geometry of the solar cell 10, if low resistance bus bars (not shown) are used on the opposite ends thereof, electrical current will generally flow in one direction, which is perpendicular to the dimension connecting the electrical-current producing regions.

Therefore, low sheet resistance is desirable in that dimension.

[0054] The operation of the solar cell 10 will now be described. Referring to Figs. 1 and 2, light from a source (not shown) is incident on the substrate layer 11, passing through the substrate layer 11, the TCO layer 12, and the HRT layer 13 to the first and second semiconductor layers 14 and 15, respectively. Some of the incident photons from the light source strike atoms of the CdTe material contained in the second semiconductor layer 15, freeing electrons therefrom in a known manner. The freed electrons travel through to the ohmic connection layer 16 and the supplemental electrode layer 17, where they flow through the second electrical lead 16a to a load (not shown) to do work. The electrons then return from the load through the first electrical lead 12a to the TCO layer 12, the HRT layer 13, and back to the first and second semiconductor layers 14 and 15, respectively, thereby completing an electrical circuit.

[0055] Some of the photons of the incident light do not, however, free electrons on the initial pass continue through the solar cell 10. Because of the translucency of the solar cell 10, those photons pass through the ohmic connection layer 16 and the supplemental electrode layer 17, thereby providing some light to pass completely through the solar cell 10 to the side opposite from the light source.

[0056] TEST DATA

[0057] Test data on experiment cells were collected on several cell variants to establish the efficiency and translucency of the design. One embodiment of this invention has been found to achieve up to 10.9 percent electrical efficiency with a fill factor of 67 percent. This was accomplished with the unique conducting layer design of this invention that was optimized for maximum conductivity for efficiency, while still providing maximum translucency.

[0058] 1. Individual Cells With Grids

[0059] Work was done on solar cells of different CdTe absorber layer thickness, but with a grid-like structure (i.e., the supplemental electrode layer 17).

[0060] Measured Test Results for 0.5μπι CdTe cells:

where Voc is the open circuit voltage, Jsc is the short circuit current, FF is the fill factor, and Eff is the electrical efficiency. [0061] 2. One Micron CdTe Translucent Cell Test Cells With 5 nm/50 nm Grids:

[0062] Measured Test Results:

Best Average

Eff. 10.89 10.18

Jsc 22.34 22.07

Voc 0.74 0.72

FF 66.92 63.95

[0063] This invention introduces a ultra-thin film technology that is specifically designed to allow maximum light transmission with high energy conversion efficiency. To achieve maximum efficiency and translucent properties, this invention uses two technological breakthroughs to address the efficiency of the cell, the translucent properties, and the low cost produceability of the cell. To achieve translucent properties, the thicknesses for the CdTe, the uniform copper/gold layer, and the conductivity enhancing uniform gold layer are optimized. It is well known that as CdTe thickness is increased, the efficiency of a constant conducting backplane cell increases the fill factor and overall efficiency.

[0064] To achieve the design for this invention, a Figure Of Merit study in efficiency and fill factor versus CdTe, gold, copper, and CdS thickness as a function of transparency was conducted. Based on these theoretical and experimental data, this Figure Of Merit Study showed that the CdTe thickness should be in the order of 500 nm.

[0065] The conductive material may be any electrically-conducting material or combination of materials including (but not limited to) copper, platinum, gold, or other metals. CdTe films of thickness less than about 1000 nm become increasingly

transparent to the human eye. At the same time, however, solar cells with CdTe thickness of about less than 1000 nm begin to lose power conversion efficiency. Finding an optimum between these two competing effects is, therefore, important. These two effects can be quantified as follows:

FOM = Eff ^■ TFOM with FOM is the total figure of merit, Eff is the power conversion efficiency of the solar cell, and TFOM is the transparent figure of merit, which quantifies the cell's transparency and remains to be defined.

[0066] The effect of the power conversion efficiency on the total figure of merit is this: it causes the total figure of merit to be zero when the solar cell 10 produces no power, and causes the total figure of merit to increase as the output power of the solar cell 10 increases, and the scaling is natural. For the purpose of this aspect of the invention, it is desired that the transparent figure of merit to have similar properties.

[0067] It would be possible to define the transparent figure of merit as a function of the optical transmittance of the CdTe film (or of the total solar cell 10), namely, it is zero when the solar cell 10 cannot be seen through and it increases as the solar cell 10 becomes easier to see through. However, the scaling does not correspond to human vision because humans do not see with double beam spectrometers. Moreover, there is the question of how transmittance, which is in general a function of wavelength, should be reduced to a single number.

[0068] The actual unit that represents the human eye's response to light is not percent (transmittance) or watt (of light energy received) but lux. Lux takes into account the human eye's response to light as a function of frequency (or wavelength), by weighing a source's power density (in watts per square meter) with the luminous efficiency function. For example, as shown in Fig. 4, the human eye is most sensitive to light around 550 nm (which is green), and the luminous efficiency function peaks at that region.

[0069] However, lux alone may not be sufficient suitable as the transparent figure of merit because of the enormous dynamic range of the human eye. The light of one sun is about 10⁵ lux, while the human eye is capable of detecting light as dim as 2x10^~9 lux. For a more common example, the lux of lighting in a typical office is 5xl0² lux (about two hundred times less than the light from the sun), yet most people would not describe the light in a typical office as being dim. Therefore, it is proposed herein to measure the transparent figure of merit in decibels as follows: Lux

TFOM = 10Jfi log 40

( LuxO

[0070] For sound, the factor Io is defined to be the threshold of human hearing, so a corresponding factor for light LuxO could be defined as the minimum threshold for human vision. However, for humans to be able to detect light as dim as 2x10^~9 lux, it is typically necessary to be acclimated in total darkness for about a half an hour, a situation that is unlikely to be of interest. Moreover, human light detection in the retina is the result of both (1) the cones in the human eye, which have color sensitivity, are in the center of the retina (the fovea), and operate at a relatively high light level; and (2) the rods in the human eye, which are insensitive to color, surround the fovea, and operate at low light level. Vision that relies on detection by cones is called photopic vision, while vision that relies on detection by rods is called scoptic vision (there is also an

intermediate range, mesopic vision, in which both cones and rods are active). Vision becomes pure scoptic vision at a level of 0.034 lux. Therefore, for the purpose of this illustration, it will be assumed that the value of LuxO is 0.034 lux. In practice, however, the value of LuxO may be somewhat larger, perhaps as large as 400 lux in a room that is illuminated by conventional electric lighting.

[0071 ] The total figure of merit for a trans arent cell can then be defined as:

with Lux being the lux transmitted through the solar cell 10 with a standard (AM 1.5) one sun on the other side.

[0072] The lux through a solar cell can be calculated as the integral over wavelength as follows:

where T is the optical transparency of the device as a function of wavelength, L is the photopic luminosity function, and I is the AM 1.5 solar spectrum (see Fig. 5, which illustrates that the irradiance is significant to about 2.5 microns - however, the sensitivity of the human eye is essentially zero past about 800 nm). Τ(λ) can be calculated from the absorption coefficients and thicknesses of the cell's component layers.

[0073] The transparent figure of merit can be based on a transmission function that is derived from the simple product of the transmission of each layer. In this analysis, reflection at interfaces and constructive and destructive interference effects are ignored.

[0074] The effect of CdS can be simplified by considering two limits, a thin limit (in which the CdS absorbs no light whatsoever), and a thick limit (in which all light of wavelength shorter than 515 nm (corresponding to the CdS bandgap of 2.4 eV) is completely absorbed but all light of longer wavelength is completely transmitted).

[0075] Fig. 6 illustrates the transparent figure of merit of the solar cell 10

(corresponding to perfect transparency of the substrate and a perfectly transparent back contact). As shown therein, the thickness of the CdS layer 14 has almost no effect on the transparent figure of merit. For CdTe layers 15 that are thicker than 250 nm, the thickness of the CdS layer 14 has significant no effect at all. This is because CdTe strongly absorbs 515 nm (and shorter) wavelength light, so even a modestly thin CdTe layer 15 will absorb all the light that CdS layer 14 could absorb.

[0076] It can be seen from Fig. 6 that the relationship of the transparent figure of merit with thickness of the CdTe layer 15 is nearly linear. Furthermore, the deviation from linearity is that the slope decreases slightly with increasing thickness of the CdTe layer 15. The transmittance through a film has an exponential dependence on film thickness (Beer's Law), while the logarithmic nature of the transparent figure of merit transforms exponential decay back to a linear decrease. A complication is that the absorption coefficient of CdTe is a function of wavelength. An average effect (which the

transparent figure of merit measures) would be expected to most heavily weigh the lowest absorption coefficients, since strongly absorbed light barely contributes to the transmitted lux. The lowest absorption coefficient is at the bandgap, corresponding to 825 nm light. On the other hand, the photopic response most heavily weighs light at about 550 nm. In fact, a linear fit to the transparent figure of merit in Fig. 6 gives a slope corresponding to the absorption coefficient of the CdTe layer 15 at about 605 nm, which is indeed between the two intuitive limits. As the CdTe layer 15 becomes thicker, the weighing effect of the photopic response becomes less important compared to the effect of the lesser absorption at the bandgap, causing the slope to decrease in absolute magnitude slightly.

[0077] The transparent figure of merit as shown corresponds to subjective experience of the transparency of thin CdTe films. It could be considered that a CdTe layer 15 of about 2000 nm would be "impossible" to see through. Glasses for safely viewing a solar eclipse appear to be completely opaque until they are held directly facing the sun. The effect a CdTe layer 15 that is much thicker than 1000 nm is similar. On the other hand, a CdTe layer 15 that is about 500 nm is easily seen through, especially when looking directly through it at outdoor light, while a CdTe layer 15 that is about 150 nm seems to merely add a slight yellow tint.

[0078] Reasonable efficiencies are possible with an ohmic contact layer 16 of a thin (5 nm) layer of gold, so long as a grid of thicker gold is also used for current collection. With an ohmic contact layer 16 of only 3 nm of gold, current collection is less desirable, presumably because the thinner layer of gold does not completely wet the surface (as opposed to the overall resistivity of the sheet calculation mentioned above). Regardless, a gold ohmic contact layer 16 having a thickness of 5 nm is not perfectly transparent, but blocks about half of all incident light. The effect of a 5 nm layer of gold is shown in Fig. 7. The effect is not strong for most wavelengths of light. One noticeable effect is the decrease (in the absolute magnitude) of the slope with increasing thickness of the CdTe layer 15 is less. This may be explained by noting that the transparency of a thin layer of gold is highest in the green region of the spectrum, where the photopic response is near maximum. This increases the weighting of the CdTe absorption in that region, lessening the effect of the lesser CdTe absorption near the bandgap.

[0079] As efficiency enters into the total figure of merit for transparent cells, we need a way to calculate it as a function of CdTe thickness. Fig. 8 shows an estimate of achievable efficiency. The exact mathematical form of the curve in Fig. 8 is not derived from physical principles but is merely a good approximate match to some obtained data and expected trends.

[0080] With estimates or calculations of the efficiency and transparent figure of merit for CdTe solar cells, it is possible to obtain the total figure of merit by multiplying the two together. The result is shown in Fig. 9, which is a chart illustrating the total figure of merit for a transparent CdTe solar cell that weighs the effects of transparency and power efficiency. As shown therein, in all cases (regardless of whether a transparent ohmic contact layer 16 or 5 nm gold ohmic contact layer 16 is used, and almost regardless of the thickness of the CdS layer 14), the total figure of merit reaches a maximum at 0.6 microns of the thickness of the CdTe layer 15.

[0081] Not shown in Figs. 8 and 9 is the effect of the thickness of the CdS layer 14 on efficiency. Thinning of the CdS layer 14 increases short circuit current, but can decrease yield and open circuit voltage. These negative effects may be ameliorated with a proper highly resistive and transparent (HRT) buffer layer. These complications make it difficult to predict the effect of a very thin CdS layer 14 on performance. However, so long as the general shape of the curve of Fig. 8 remains constant, the thickness of the CdTe layer 15 for the optimum total figure of merit is expected to remain the same.

[0082] As apparent from the above discussion, several choices have been made as to how the transparent figure of merit has been defined, in particular the choice of reference lux. The reference lux was chosen on the basis of the subjective level of transparency of a CdTe device. A different application, however, could shift the level of transparency and, thus, change the value of LuxO. For example, solar eclipse glasses are transparent, in that the sun can be seen through them. However, virtually nothing else can be seen through them, including articles that are illuminated by the sun.

[0083] The question then is are these transparent devices intended to (1) let light pass therethrough to illuminate an otherwise unlighted area (such as, for example, a skylight in an residential hallway that is unlit during the day), (2) supplement electrical lighting, or (3) let so much light through that they act as a window? For each of these specific applications, a different minimum transparent figure of merit might be necessary. For the first example, the desired light level might be about that of a typical living room, and any thickness of the CdTe layer 15 that is below about 750 nm may provide sufficient transparency. For the second example, any amount of light that is less than the provided electrical light would not be noticed (similar to using a flashlight in broad daylight).

Therefore, the reference lux value to consider would be that of a typical office, resulting in a shift in the maximum total figure of merit to 300 nm of CdTe. For the third example, the thickness of the CdTe layer 15 would have to be slightly thinner yet.

[0084] Another issue is the robustness of the figure of merit in light of possible improvements to CdTe device efficiency. For example, Fig. 8 shows efficiency asymptotically reaching about 13.5%, which is the case for sputtered CdTe cells on a TEC 15 substrate. However, CdTe deposited in other ways on other substrates can reach higher efficiencies. The current world record performance is over 17%, and the theoretical limit is approximately 30%.

[0085] However, the graphs in Figs. 6 and 7 illustrate nearly straight lines, and the efficiency graph in Fig. 8 is nearly a straight line for thin CdTe (before the curve rolls over and saturates). If the efficiency is:

E = n t

and the transparent figure of merit is:

TFOM = a - b t;

then their product has a maximum at

t = a/(2b)

[0086] In other words, it is independent of n and, thus, is independent of the exact value the efficiency has for any given thickness (so long as there is a nearly linear relation between efficiency as thickness). This is true so long as efficiency improvements do not move the saturation point in the efficiency curve to much lower thicknesses.

[0087] Another issue to consider is the exact form of the transparent figure of merit. For the purpose of this discussion, the transparent figure of merit has been treated as a product of two values, the assumption being that that power generation and transparency are equally important. Notwithstanding this, however, it is contemplated that a CdTe solar panel may be used as a skylight and also connected to a compact fluorescent light (CFL) in the same room. In such an instance, it can be assume the CFL produces 60 lumens per watt. A lumen is equal to a lux times a square meter. If the CdTe skylight is one square meter in size, then the lux that it lets through is equal to the lumens it lets through. The result of this calculation is shown in Fig. 10. The maximum total lumens is at a thickness of the CdTe layer 15 of zero (a window is much better at lighting a room than a CFL, at least during a sunny day). However, a CdTe layer 15 of zero thickness also produces no energy. Because generic energy production (not just for powering lights) is as important, this justifies the transparent figure of merit definition used herein over one that only attempts to maximize lumens.

[0088] Numerous references, including various publications, are cited and discussed in the description of this invention. They are presented merely to clarify the description of this invention and are not an admission that any such reference is "prior art" to this invention. All references cited and discussed in this specification are incorporated herein by reference in their entirety and to the same extent as if each reference were individually incorporated by reference.

[0089] Although this invention has been described in detail with reference to certain preferred embodiments, such embodiments are merely illustrative of the invention. Other versions are possible to those skilled in the art without departing from the scope of the invention. Therefore, the spirit and scope of the invention should not be limited solely to the versions described above.

Claims

What is claimed is:

1. A solar cell comprising:

a substrate layer formed from a transparent or partially transparent material;

a first electrode layer engaging the substrate layer and formed from a transparent or partially transparent material;

a first semiconductor layer engaging the first electrode layer and having a thickness that allows a predetermined amount of light to pass therethrough;

a second semiconductor layer engaging the first semiconductor layer and having a thickness that allows a predetermined amount of light to pass therethrough; and

a second electrode layer engaging the second semiconductor layer, the second electrode layer including a first sheet having a generally uniform thickness that allows a predetermined amount of light to pass therethrough and a second sheet having a grid-like structure with at least one space or opening that allows light to pass therethrough.

2. The solar cell defined in Claim 1 wherein the grid-like structure is defined by a plurality of discontinuous portions that are separated from one another by spaces.

3. The solar cell defined in Claim 2 wherein the discontinuous portions extend in a length-wise direction.

4. The solar cell defined in Claim 2 wherein the discontinuous portions extend in both a length-wise direction and a width-wise direction.

5. The solar cell defined in Claim 2 wherein the discontinuous portions are generally rectangular in cross sectional shape.

6. The solar cell defined in Claim 1 wherein the grid-like structure is defined by a plurality of continuous portions that are separated from one another by openings.

7. The solar cell defined in Claim 6 wherein the continuous portions extend in a length-wise direction.

8. The solar cell defined in Claim 6 wherein the continuous portions extend in both a length-wise direction and a width-wise direction.

9. The solar cell defined in Claim 6 wherein the continuous portions are generally rectangular in cross sectional shape.

10. The solar cell defined in Claim 1 wherein one or more discontinuities are formed in the first sheet of the second electrode layer, and wherein a light diffraction blocking material is provided adjacent to the one or more discontinuities.

11. The solar cell defined in Claim 10 wherein the light diffraction blocking material is provided within the one or more discontinuities.

12. The solar cell defined in Claim 10 wherein the light diffraction blocking material is provided on the substrate layer.

13. A solar cell comprising:

a substrate layer formed from a transparent or partially transparent material;

a first semiconductor layer engaging the first electrode layer and having a thickness that is in a range of from about 50 nm to about 120 nm;

a second semiconductor layer engaging the first semiconductor layer and having a thickness that is in a range of from about 200 nm to about 700 nm; and

a second electrode layer engaging the second semiconductor layer, the second electrode layer including a first sheet having a thickness that is in a range of from about

0.5 nm to about 10 nm and a second sheet having a grid-like structure with at least one space or opening that allows light to pass therethrough.

14. The solar cell defined in Claim 13 wherein the grid-like structure is defined by a plurality of discontinuous portions that are separated from one another by spaces.

15. The solar cell defined in Claim 14 wherein the discontinuous portions extend in a length-wise direction.

16. The solar cell defined in Claim 14 wherein the discontinuous portions extend in both a length-wise direction and a width- wise direction.

17. The solar cell defined in Claim 14 wherein the discontinuous portions are generally rectangular in cross sectional shape.

18. The solar cell defined in Claim 13 wherein the grid-like structure is defined by a plurality of continuous portions that are separated from one another by openings.

19. The solar cell defined in Claim 18 wherein the continuous portions extend in a length-wise direction.

20. The solar cell defined in Claim 18 wherein the continuous portions extend in both a length-wise direction and a width-wise direction.

21. The solar cell defined in Claim 18 wherein the continuous portions are generally rectangular in cross sectional shape.

22. The solar cell defined in Claim 13 wherein one or more discontinuities are formed in the first sheet of the second electrode layer, and wherein a light diffraction blocking material is provided adjacent to the one or more discontinuities.

23. The solar cell defined in Claim 22 wherein the light diffraction blocking material is provided within the one or more discontinuities.

24. The solar cell defined in Claim 22 wherein the light diffraction blocking material is provided on the substrate layer.

25. A method of manufacturing a solar cell comprising the steps of:

(a) providing a substrate layer formed from a transparent or partially transparent material;

(b) applying a first electrode layer to the substrate layer, wherein the first electrode layer is formed from a transparent or partially transparent material;

(c) applying a first semiconductor layer to the first electrode layer, wherein the first semiconductor layer has a thickness that allows a predetermined amount of light to pass therethrough;

(d) applying a second semiconductor layer to the first semiconductor layer, wherein the second semiconductor layer has a thickness that is a function of a transparent figure of merit; and

(e) applying a second electrode layer to the second semiconductor layer, the second electrode layer including a first sheet having a generally uniform thickness that allows a predetermined amount of light to pass therethrough and a second sheet having a grid-like structure with at least one space or opening that allows light to pass

therethrough.

26. The method defined in Claim 25 wherein step (d) is performed by applying a second electrode layer having a generally uniform thickness that is determined from a total figure of merit calculation.

27. The method defined in Claim 26 wherein the total figure of merit calculation is performed by multiplying a power conversion efficiency of the solar cell by a transparent figure of merit calculation.

28. The method defined in Claim 27 wherein the transparent figure of merit calculation is at least:

29. The method defined in Claim 27 wherein the transparent figure of merit calculation is equal to: