Field of the Invention

The present invention relates to a thin film solar cell. In particular, the present invention relates to a thin film solar cell that has one or more diffractive optical supercell structures with the purpose of enhancing light trapping.

Background of the Invention

Solar panels are made of a series of solar cells. These are used to capture sunlight and convert it into electrical power. With the on-going move towards renewable energy, solar panels are becoming of increasing interest to consumers and businesses alike. However, the cost of solar panels is still relatively high and the efficiency relatively low. To make solar panels more accessible, there has been a drive to reduce costs and improve efficiency. To this end, thin film solar cells are increasingly being used. These devices have an active layer that has a reduced thickness. This means that processing and material costs are reduced, while the overall device efficiency can be maintained. To maintain the efficiency of thin film devices, light trapping structures are used to effectively couple light into guided modes of the thin film, thus facilitating efficient absorption.

Recent work has shown that the use of diffractive optical structures provides a promising light trapping technique for thin-film solar cells. Diffractive structures or gratings can be used to excite quasi-guided modes in the absorbing film, providing strong absorption enhancement at the resonant wavelength. As an example, a simple grating has been used in a solar cell to act as a surface coupler. The grating has a period corresponding to the wavelength of light, i.e. 500-1000 nm for silicon-based solar cell applications. Such a grating relies on few diffraction orders, mainly the first, to couple light into the thin film. This gives a very narrow-band and strongly angle- dependent enhancement, so only a limited scope for light trapping.

To increase the bandwidth and the number of angles that can be coupled into guided modes, a larger period grating can be used. In this case, higher orders of the grating are used to excite multiple modes in the thin film. Simple gratings diffract most of the energy into the first diffracted order. However, this cannot excite a guided mode if the period is larger than the wavelength. Hence, most of the incoming power is lost and not successfully diffracted into a guided mode. Summary of the Invention

According to the present invention, there is provided a solar cell that has a thin film of active material on which is defined a periodic series of diffraction cells, wherein each diffraction cell has a non-periodic series of diffraction features and is adapted to suppress at least one low order diffraction mode.

By designing each diffraction cell to suppress lower order diffraction modes, which generally cannot be coupled into the thin film, more energy is transferred into the higher order modes, which can couple into the thin film. This allows multi-mode coupling, and maximises the excitation of guided modes into the thin film.

Typically, the period of the diffraction cells is two or more optical wavelengths, at the wavelength of interest. Preferably, the period is three to five optical wavelengths.

Preferably, the first order mode is suppressed. In addition, the second order mode may be suppressed. Further higher order modes may be suppressed, for example the third and optionally the fourth mode.

The diffraction cells may be provided in a one dimensional array. Alternatively, the diffraction cells may be provided in a two dimensional array.

The diffraction cells may all be the same.

The thin film of active material may have a thickness in the range of 100 nm to 100 μηι, the range between 100 nm and 1 μηι being the most promising, depending on material.

The thin film of active material may comprise silicon, but the method can also be successfully applied to other materials such CdTe, CIGS and the different forms of silicon, such as amorphous, microcrystalline and single crystalline silicon. According to a second aspect of the present invention, there is provided a solar panel that includes a plurality of solar cells according to the first aspect of the invention.

According to yet another aspect of the invention, there is provided a solar panel system that includes a plurality of solar panels according to the second aspect of the invention.

Brief Description of the Drawings

Various aspects of the invention will now be described by way of example only and with reference to the accompanying drawings, of which:

Figure 1 (a) is a plan view of a solar cell with a diffractive optical supercell;

Figure 1 (b) is a plan view of a solar cell with conventional periodic gratings;

Figure 2(a) is a cross section through the solar cell of Figure 1 (a), in which light coupling into the thin film is shown;

Figure 2(b) is a cross section through the solar cell of Figure 1 (b), in which light coupling into the thin film is shown;

Figure 3 is a plan view of a supercell solar cell that is specifically designed to suppress the first and second diffraction orders;

Figure 4 shows the Fourier series for the cell of Figure 3;

Figure 5 illustrates the design principle of a supercell grating that consists of a superposition of multiple low filling factor gratings with the same period, where the gratings are spatially offset from one another;

Figures 6(a) to (c) show a comparison between the performances of a conventional subwavelength grating and a super cell;

Figure 7 is a Scanning Electron Microscope (SEM) image of a supercell that was fabricated in Silicon On Insulator (SOI) material;

Figure 8 is a plot of absorption versus wavelength for experimental data and calculated data for a SOI wafer with supercell gratings;

Figure 9 is a plan view of a fabricated 2D supercell;

Figure 10 shows a plan view of a 2-D supercell lattice on a thin film solar cell (a), and the corresponding diffraction orders (b), also known as Fourier components;

Figure 1 1 shows an absorption spectrum for a supercell and an absorption spectrum for a standard 2D grating with an optimised period of 350nm;

Figure 12 is a cross section through another example of supercell solar cell, and

Figure 13 is a cross section through yet another example of supercell solar cell. Detailed Description of the Drawings

Figure 1 (a) shows a supercell based solar cell and Figure 1 (b) shows a conventional grating based solar cell.

The supercell solar cell of Figure 1 (a) has a thin film of active material, for example silicon, on the surface of which is defined a series of supercells, which form a series of non-periodic steps in the external surface of the solar cell. Each supercell has non- periodic features, in this case simple line features. In the example shown, the width of the line features within each supercell is different. In addition, the spacing between the line features is different. The period of the supercell is a multiple of the target wavelength (typically 3-5 wavelengths).

The solar cell of Figure 1 (b) has a thin film of active material, for example silicon, on the surface of which is defined a conventional periodic diffraction grating. In this case, the line features of the grating are all of the same width and the spacing between each line feature is the same.

The supercell grating is designed to couple higher diffracted orders into guided modes and to suppress modes that cannot be coupled, typically the first and second order modes. As a rule of thumb, if the supercell is m wavelengths long, then the m-th and higher diffracted orders couple into a guided mode while the m-1 ^st and lower order modes are diffracted into air, so need to be suppressed. For example, for a wavelength of λ=600 nm and m=3, the supercell would be a=1800 nm long, and the fine structure would be designed to suppress the 1 st and 2nd diffracted order while enhancing the third and higher orders.

The supercell of Figure 1 (a) is designed to suppress the first and second order modes. The Fourier series for the cell of Figure 1 (a) is illustrated in the graph on the right hand side of Figure 1 (a). From this, it can be seen that the first and second orders are substantially suppressed, and most of the incident energy is in the third and fourth diffraction orders. In contrast, for the conventional grating of Figure 1 (b), much of the incident energy is transferred into the first and second diffracted orders. Figure 2 illustrates the difference in coupling between the different diffracted orders in case of the supercell grating and the regular grating. This shows the advantages of suppressing the lower diffraction orders, which do not couple into the thin film. By ensuring that as much energy as possible is transferred into the higher diffraction orders, coupling efficiency for the supercell solar cell is better than for the simple grating, where the lower diffractions orders that do not couple into the thin film have relatively high energies.

For the sake of completeness, it should be noted that in the example of Figure 1 , and indeed all of the supercell solar cells described herein, the zero order exhibits the highest amplitude. This order is not affected by the supercell geometry. Hence, it needs to be suppressed by a separate optimisation involving the depth and fill-factor of the grating. Techniques for doing this are known in the art and are not the subject of the present invention.

Figure 3 shows an example of a supercell that is designed to suppress the first two diffraction orders for enhancing absorption on a thin-film crystalline silicon solar cell, typically 400nm thick. The absorbing region spans the wavelength range from 300nm to 1 .2μηι and the grating is designed to trap light mainly in the region between 500nm and 1 .2μηι, where the absorption length is much larger than the material thickness. This supercell has a period of 1 .8 μηι. This means that the first two diffracted orders contribute to diffraction into air. Following the design rule, these diffracting orders are suppressed by the supercell. Figure 4 shows the Fourier series of the supercell of Figure 3.

Suppression of the first and second orders is caused by interference. The supercell Fourier series comprises multiple top-hat functions Fourier series. Because each top- hat function is displaced in space from each order, their Fourier series are phase shifted by a factor of exp(i2nna/ ). Here, / ^' is the pure complex number, n is the order number, a is the displacement length and is the period. Since this phase shift depends on the order number, it is possible to get constructive interference for higher orders and destructive interference for the lower orders when multiple top-hat functions are added together. By suitably designing the supercell, it is possible to ensure that the lower diffraction orders destructively interfere and so are suppressed. To design the cell the Fourier spectrum of the grating is tailored. This process is illustrated in Figure 5 and is based on the well-known property of the Fourier series that a spatial shift of the function induces a phase shift of the Fourier series, as shown in equation 1 :

2π

f (x - x ₀ ) <^ F (m) exp(-j— mx ₀ ) (equation. 1 )

a where F(m) is the Fourier series of f (x), m is an integer corresponding to the diffraction order, a is the period, and x ₀ is the spatial translation. Since the phase shift depends on x ₀ and on m, the spatial shift can be chosen to provide destructive interference for lower orders and constructive interference for higher orders. In this example, the aim is to couple light mainly in the wavelength region above 600 nm, because this is the spectral region where the absorption in silicon is very low. Once the wavelength region of interest is defined, the highest diffractive order is given by the ratio between the period and the wavelength. As an example, the period is selected to be 1800 nm, which gives a ratio of 3; i.e., the m = 3 order couples into air for A < 600 nm, and only into the slab for A > 600 nm, while orders 1 and 2 always diffract into air and must therefore be suppressed. The next step is to superimpose multiple gratings in order to reduce orders 1 and 2. The selection of the required spatial shifts is done using a simple binary search. In the binary search, an array of pixels is defined. Pixels defined as 1 represent ridges and pixels defined as 0 represent grooves. The pixel size is arbitrary, and so the fabrication limitations were chosen as the constraint criteria, i.e. 1 12 nm. Therefore, there are 32 pixels for the chosen period of 1800 nm, with a minimum block size of 2 pixels.

A binary search is performed by swapping the value of each pixel and calculating the Fourier transform of the total pixel array each time a pixel is changed. The resulting supercell consists of 10 blocks, with the following sequence of number of pixels: 4, 2, 3, 4, 2, 2, 3, 5, 4, 3. This sequence, with bold representing ridges, is shown schematically in Figure 5.

The performance of the supercell can be compared to conventional sub-wavelength gratings using the Rigorous Coupled Wave Analysis (RCWA), which is a standard semi-analytical method to calculate diffraction problems. As an example, the performance of an optimised simple sub-wavelength grating (with 600nm period) was compared to the supercell shown in Figure 3. For the purposes of the calculations, it was assumed that the silicon slab was 400nm thick. The integrated absorption was calculated assuming the solar photon density. The enhancement factor is a measure of the total absorption relative to an un-patterned slab.

Figure 6 shows comparative simulations for a supercell based solar cell and a simple grating. Figure 6 (a) shows calculated enhancement over un-patterned slab for TE polarization for both a super cell solar cell and a simple grating. Figure 6 (b) shows calculated enhancement over un-patterned slab for TM polarization for both a super cell solar cell and a simple grating. Figure 6 (c) shows calculated enhancement over un-patterned slab for un-polarized light for both a super cell solar cell and a simple grating. As can be seen in Figure 6, the supercell outperforms the subwavelength grating over a wide range of incidence angles. This is advantageous for solar cell applications.

Figure 7 shows a Scanning Electron Microscope (SEM) image of a supercell that was fabricated in a Silicon On Insulator (SOI) wafer, which has 220nm of silicon on top of 2μηι silicon dioxide, sitting on a 500 μηι substrate. The absorption of the SOI with the grating on the top was measured. The contribution of the grating is seen as resonances in the absorption spectrum. Figure 8 shows a comparison between experimental and calculated absorption on a SOI wafer with the supercell gratings. As can be seen from this, there is a very good match between the experiment and the calculations.

The supercell concept can be extended to 2-D structures. An example of a fabricated 2D supercell is shown in Figure 9. This 2D supercell is achieved by simply repeating the 1 D supercell of Figure 7 in the x and y directions. Another option is to create a superlattice that suppresses low diffraction orders, as is shown in Figure 10. Figure 10(a) shows a superlattice and Figure 10(b) shows the corresponding Fourier components of the superlattice showing suppressed low diffraction orders. In this case, the 2D structure was created using a numerical optimisation routine based on the concepts described above. The benefits of the supercell approach are higher in the 2D case than in the 1 D case, as 2-D patterning allows control of more degrees of freeedom. Numerical simulations have shown that the 2D supercell can substantially outperform standard 2D gratings.

Figure 1 1 shows an absorption spectrum for a supercell and an absorption spectrum using a standard 2D grating with an optimised period of 350nm. As is clear from the spectra, the supercell can excite much more resonances than the standard grating, resulting in a 46% efficiency enhancement. Numerical simulations also show that the supercell is capable of providing, in a 400nm thick silicon slab, the same integrated absorption that a standard optimised 2D grating can provide in a 1 um thick slab. Therefore, the supercell is capable of reducing the cost associated with more than half of the absorbing slab thickness.

Figure 12 shows a solar cell that includes the diffractive supercells of the invention. In this case, the supercell grating is provided in the interface between an externally facing transparent conductor and the active layer. To make the cell of Figure 12, the active layer may be patterned on an upper surface and coated with a layer that is transparent to light. The rear surface of the active layer is coated with a metal or another conductive layer. Optionally, the rear surface is also transparent. Alternatively, an upper substrate is first patterned and then coated with a transparent layer. In this case, the pattern is automatically transferred to the transparent conductor. Finally the active layer is deposited. Another option is to pattern the transparent conductor and then deposit the active layer.

Figure 13 shows an alternative arrangement. In this case, the interface between the active layer and the externally facing transparent layer is substantially flat, and the diffractive supercell grating is provided at the interface between the rear conductor and the active layer. In this example, the structure may be made starting from the transparent conductor, then the active layer is deposited and patterned, and finally the metal on the rear surface is deposited. Another option would be to start from the metal on the rear surface, pattern it, then deposit the active layer and finally deposit the transparent conductor.

The present invention provides a simple and effective solution to the technical problem of how to maximise coupling efficiency in thin film solar cells. Advantageously, the thin film solar cells of the present invention can be manufactured using conventional semiconductor processing techniques, for example photolithography, nanoimprint litography and etching to form the diffractive optical structures.

A skilled person will appreciate that variations of the disclosed arrangements are possible without departing from the invention. For example, although the solar cells described above have periodic arrays of supercells, this is not essential, and the supercells may be non-periodically spaced, provided the overall arrangement and interaction of the supercells is such as to suppress at least one low order diffraction mode. Accordingly, the above description of the specific embodiment is made by way of example only and not for the purposes of limitation. It will be clear to the skilled person that minor modifications may be made without significant changes to the operation described.