Open Access
Add to BibSonomyAbstract
Solution processable nanocrystal solar cells combine the advantages of low-cost printing and wide range of accessible absorber materials, however high trap densities limit performance and layer thickness. In this work we develop a versatile route to realize the infiltration of a photonic crystal, with copper indium diselenide nanocrystal ink. The photonic crystal allows to couple incident light into pseudo-guided modes and thereby enhanced light absorption. For the presented design, we are able to identify individual guided modes, explain the underlying physics, and obtain a perfect match between the measured and simulated absorption peaks. For our relatively low refractive index layers, a 7% maximum integrated absorption enhancement is demonstrated.
© 2017 Optical Society of America
1. Introduction
To target a lowering of costs in photovoltaic (PV) production, and open new markets, many novel materials and fabrication techniques have been proposed in recent years. Widely addressed targets are vacuum-free, low-temperature, and high-throughput processes on inexpensive substrates. Next to organic and dye sensitized solar cells, other solution based materials have come into focus. Particle- and precursor-based solutions of PV active materials promise to fulfill the targets [1]. A concern they have in common with other thin film technologies, such as amorphous silicon, is the increase of defect densities. The consequence is an increased in charge carrier recombination. Thin absorber layers are therefore not only a feature, but matching the extent of the depletion region is a necessity. One way for ensuring proper light absorption in these thin films is by using light-trapping strategies.
One of these new material systems are colloidal particles of I-III-VI2 chalcopyrite semiconductors [2]. They are attractive as they have a broadly tunable, direct band-gap and are long-term stable. In conventional thin film deposition techniques for chalcopyrite solar cells, such as copper indium (gallium) diselenide (CuIn(Ga)Se2), layers are co-evaporated in a time consuming multi-stage process, achieving the desired stoichiometry and film thickness is often a challenge [3]. Nanocrystals, on the other hand, are chemically synthesized with a controlled composition and size [4]. When dispersed in a solvent, nanocrystals can be applied as an ink. Possible new market applications include roll-to-roll printing and the use of flexible substrates, uniquely the here used CuInSe2 (CIS) nanocrystals hold the world record for solar cell directly fabricated on paper [5]. For standard devices on a glass substrate the best device performance was achieved with an absorber layer of only 150 nm average thickness, possessing a conversion efficiency (η) of 3.1% [6]. The device underwent mild heat treatment at 200 °C post-deposition, still compatible with most flexible substrates. It is worth highlighting that CIS nanocrystals are quite suitable for fabricating high efficiency (15.0%) solar cells [7], but these high efficiencies have only been achieved with high annealing temperatures (>500 °C) or selenization processes.
Just like comparable low-cost PV materials the CIS nanocrystals suffer from a high charge carrier trap densities. The comparably low efficiency achieved with these devices clearly indicates that improvements in film formation are of primary importance. Nevertheless, a key challenge will remain to ensure high light absorption and simultaneously efficient charge carrier extraction. As in many devices thick layers quickly reduce the short circuit current density (Jsc). Evident for this behavior are two devices with average CIS layer thicknesses of 140 nm and 280 nm. The 280 nm device, although thicker, affords a conversion efficiency of only about 1% while the 140 nm device achieves almost 2% (for measurement details see [8], Panthani et. al.). Figure 1(a) shows the current density-voltage (J-V) curve of the two devices. The optically relevant layers are indium tin oxide (ITO), zinc oxide (ZnO), cadmium sulfide (CdS), the CIS nanocrystal absorber layer and a gold (Au) back contact (see Fig. 1(b), for details on the deposition techniques see [8], Panthani et. al.), these devices did not undergo post-deposition heat treatment. For further illustrating the issue, we show in Fig. 1(c) and Fig. 1(d) the spectral irradiance (air mass 1.5 global (AM 1.5g) solar spectrum [9]) absorbed and converted to charge carriers in the CIS layers of 140 nm and 280 nm, respectively. The complete solar cell stacks are simulated, and the absorption in the sole CIS layers is reported. The basis for the conversion to charge carriers is the measured external quantum efficiencies (EQE). EQE curves of similar devices can be found in [8], Panthani et. al.. Area (I) is the difference between the light absorbed in the CIS layer and charge carrier extraction, it shows internal losses. This loss is dominant in the thicker device. Area (II) corresponds to light incident upon the cell, but not absorbed in the CIS layer. This optical loss is the main limiting factor in thin devices. These observations highlight the need to exploiting a light-trapping strategy to assist the absorption enhancement in nanocrystals-based solar cells. The benefits are here twofold: to keep the layer thicknesses low, and to improve the harvesting of the incident light.
Fig. 1 (a) J-V characteristics of nanocrystal solar cells with CIS layer thicknesses of 140 nm (solid) and 280 nm (dotted) under AM 1.5g illumination (red) and dark condition (black). (b) Stacked layer structure of the cells. Utilization of spectral irradiance in the 140 nm (c) and in the 280 nm (d) CIS layers. Spectral irradiance absorbed in the CIS layer (dark green) is retrieved from transfer matrix simulations of the complete layer stack. The spectral irradiance converted to charge carriers (light green) is plotted using EQE data. The total spectral irradiance of the AM 1.5g spectrum is also plotted (black). The Jsc available via these spectra is given in the legends.
Photonic crystal (PhC) slabs [10], high-index waveguides with a lattice of two-dimensional (2D) periodicity, have received considerable attention among the different nanophotonic structures compatible with a thin-film stack. They can be designed to dramatically enhance the absorption in a targeted wavelength range [11]. The principle behind this approach relies on the coupling of the incoming light with pseudo-guided modes (slow Bloch modes) of the PhC, which results in additional absorption peaks [12,13]. Thus, successful implementations of a PhC slab in PV devices were reported for organic [14,15], inorganic [16–19], and organic-inorganic hybrid [20,21] PV active materials. Detailed studies of the light coupling to pseudo-guided modes and its potential to enhance light trapping in PV have been carried out numerically [22–24], and experimentally [25–27]. In practice, fabrication of periodic structures can be characterized as top-down or bottom-up methods. In top-down mostly an electron beam lithography [28], nanoimprint lithography [29], or interference lithography step [30], was followed by the transfer of the pattern into the active layer by etching. In bottom-up often vapor deposition is used either incorporating a mask [31], or a pre-textured substrate [32], or a vapor-liquid-solid (metal-catalyzed) growth is performed [33,34]. While absorption gains could be readily evidenced, especially for the top-down approaches the positive influence of the PhC on the resulting solar cell efficiency turned often out to be more challenging to demonstrate because of deleterious surface recombination effects [35–37].
In the present work, we propose to improve light-management in CIS nanocrystal-based PV absorbers using an infiltration approach, whereby the nanocrystal ink is directly infiltrated into a lower refractive index and patterned polymer template. Consequently, no additional passivation step is required and our method can be directly extended to more complex photonic structures, including quasi-random configurations [38]. For the sake of demonstration, we fabricate a PhC slab (square array of photoresist nanocones) by using direct laser writing (DLW) [39], which is one the most versatile nano- to microscale structuring techniques to date. This includes the possibility of implementing various kinds of complex 3D architectures that have been numerically shown to have higher, broadband absorption over a wider range of incidence angles compared to their 2D counterparts [40].
This array is directly formed onto a fused silica substrate. We exclude multiple layers in this simplified design to minimize interference effects and to unambiguously identify the creation of new absorption peaks originating from the PhC. For the same reason, we choose a thick absorber that allows several pseudo-guided modes to be coupled, with a view of demonstrating the validity of our approach.
Our study is organized as follows: We first determine the optical constants for CIS nanocrystal layers, measured using spectroscopic ellipsometry. This data is then used to numerically design and analyze the desired PhC that enables the generation of absorption peaks in the weakly absorbing region. Finally, we validate our approach experimentally and demonstrate an excellent agreement between our measured and simulated absorption spectra, proving that infiltrated PhC slabs can be used to tailor and to improve the optical properties of CIS nanocrystal-based solar cells.
2. Methodology
2.1 Optical simulations
The geometric optimization of the infiltrated PhC slab is based on 3D optical simulations that use the finite-difference time domain (FDTD) method available in a commercial software [41]. All simulations are performed in a wavelength range from 300 nm to 1300 nm using plane waves launched at normal incidence.
2.2 Fabrication: materials and techniques
The CIS nanocrystals used in this work are chemically synthesized by arrested precipitation, the full details being available in [8], Panthani et. al., the reaction temperature is 240 °C. The nanocrystals possess a diameter of 8.6 ± 1.9 nm – hence there is no quantum confinement of the band gap - and are capped with oleylamine ligands. Prior to their deposition, they are dissolved in toluene in a 100 mg/ml solution.
DLW is used for flexible photoresist patterning of the PhC lattice. In DLW, a laser beam is focused into a photoresist layer on a substrate. The photoresist is transparent for the laser wavelength, but, due to high intensities at the focal point, two-photon absorption occurs. The two-photon polymerization that follows is a non-linear process, making feature sizes below the diffraction limit possible. By moving the focal spot through the photoresist, 3D patterns and structures can be created. The commercially available DLW system used (Nanoscribe GmbH Photonic Professional GT) employs a frequency-doubled fiber laser emitting pulses at 780 nm. The laser beam is focused through an objective with 63 × magnification and a large numerical aperture (NA = 1.4) immersed directly into the photoresist. The photoresist used is IP-Dip (Nanoscribe GmbH) with the unexposed resist possessing a refractive index of 1.52 at 780 nm [42].
For pattern fabrication, the focal point is positioned above a fused silica substrate in such a way that half of the focal region exposes the photoresist and the other half resides in the substrate. The dimensions of the exposed region can be varied via slight adjustments of the focal point height, or more directly by adjusting the laser dose. The focal spot is moved to every lattice position in a 100 µm by 100 µm field using the positioning system based on pivoted galvo mirrors. For the large-area texturing realized here, it is necessary for multiple fields to be stitched together. To make optical measurements possible an area of 10 mm by 5 mm was structured. Unexposed resist is removed using the solvent propylene glycol methyl ether acetate (PGMEA), and the sample subsequently washed with isopropanol. The CIS nanocrystals are then infiltrated into the patterned samples using a doctor blading system (Zehntner ZAA 2300). This is realized by sweeping the nanocrystals ink over the sample at a distance of 200 µm from the surface and at a speed of 60 µm/s. All samples are dried under ambient conditions.
2.3 Measurement systems
Transmission electron microscopy (TEM, FEI Tecnai G2 Spirit BioTwin) images of the CIS nanocrystals are taken for determining particle size. The optical constants of an unpatterned nanocrystal thin film are determined using a rotating analyzer spectroscopic ellipsometer (J.A. Woollam Co. VASE®) running WVASE32 (J.A. Woollam, Ver. 3.774) analysis software. Details of this measurement will be discussed in section 3.1. The presented absorption, reflection and transmission spectra of samples are measured using a spectrometer (Perkin Elmer Lambda 950), equipped with a 100 mm integrating sphere accessory. Fabricated samples are investigated with a scanning electron microscope (SEM, Zeiss Supra 60 VP) and atomic force microscopy (AFM, Veeco Dimension Icon).
3. Results
3.1 Optical properties of the planar CIS layer
Four CIS “bulk” films of varying thickness (80 nm, 100 nm, 170 nm and 400 nm) were deposited via doctor blading on soda-lime glass and were measured by spectroscopic ellipsometry. Five angles of incidence from 35° to 75° were considered. An effort was made to describe the layers as an effective medium consisting of bulk CIS and oleylamine, but a match to the oleylamine oscillator model was not possible. This indicates that the optical properties are dominated by electronic transitions in CIS. Consistently the model that reproduces our system with the best fit assumes the layer as a single bulk, and consists of three Tauc-Lorentz oscillators, which relate to the three band transitions reported for CIS [43]. The oleylamine seems to mainly affect the oscillator strength. The resulting real (nCIS) and imaginary (kCIS, the extinction coefficient) parts of the complex refractive index (nCIS) of the CIS layer are shown in Fig. 2(a). In particular, it can be calculated that the refractive index difference between this layer and the photoresist used for patterning (nIP-DIP ≈1.5) is Δn ≈0.5 in the weakly absorbing regime. When comparing to the optical constants measured for bulk CIS [44], a high similarity in curve shape is apparent. However, the real part of the refractive index measured for bulk CIS is of the order of n ≈3, and the extinction coefficient roughly four fold higher.
Fig. 2 (a) Measured real (black) and imaginary (red) parts of the refractive index of the CIS nanocrystal layers. (b) The simulated (black solid) reflection, transmission and absorption spectra of the 400 nm thick CIS layer, obtained by implementing the measured CIS optical indices, are in good agreement with the experimental spectra (red crosses).
Figure 2(b) displays the measured transmission, reflection, and absorption spectra of the 400 nm thick CIS film on glass; and compares it to the calculated results using the transfer-matrix method [45]. An excellent agreement is obtained, showing that those indices can be used to predict the optical properties of our infiltrated absorbing PhCs.
When considering the extinction coefficient of the CIS layer, three different regions can be distinguished. For the typical layer thickness (H) of a few hundred nanometers absorption for wavelength (λ) below 500 nm is high and does not require the assistance of an additional light trapping structure (compare Fig. 2(b)). The situation differs in the 500 nm - 800 nm range, where absorption is weaker. Even though the condition for low absorption , needed for many approximations, is not strictly valid, absorption enhancement can have a significant impact in this range. For wavelengths above 800 nm, the true low absorption regime is reached. In the following, we will therefore design the PhC in such a way that new absorption peaks are introduced between 800 nm and 1300 nm.
3.2 Optical design and analysis of the PhC absorber layer
A key aspect of this work is to exploit a simple design to demonstrate that the absorption peaks introduced by our infiltrated PhC can be numerically predicted. Figure 3 shows a schematic of the sample structure investigated in this work. It consists of a single layer made of CIS nanocrystals infiltrated into a square array of photoresist nanocones on a fused silica substrate. The half-ellipsoidal shape of the PhC structure displayed here is given by the fabrication process, which was discussed in section 2.2 (an illustration of the process is shown in section 3.4). We have previously shown that various shapes of the PhC structure lead to absorption peaks in the same spectral position, as long as the volume filling fraction (VFF) of the nanostructures in the CIS layer is kept constant [46]. In general, a VFF of 25% showed a good compromise between efficient coupling of diffracted light and loss of PV active material to the polymer.
Fig. 3 Schematic of the investigated infiltrated PhC slab consisting of a 2D square array (lattice constant “a”) of photoresist nanocones embedded into a layer (thickness “H”) of CIS nanocrystals and standing on a fused silica substrate.
In order to determine the lattice constants (a) for which coupling to pseudo-guided modes in our devices is possible, we followed an approach based on the works by Tikhodeev et. al. and Ruppe et. al. [47,48]. For that purpose the PhC slab was first approximated by a homogeneous medium with an average complex refractive index . For a homogenous waveguide layer normal to the z-axis (compare Fig. 3) on a dielectric substrate, the eigenvalue equation for s-polarized (TE) modes is given by [49],
and for p-polarized (TM) modes by [38],where, , , and with kxy as the propagation constant in the waveguide plane, k0 = 2π/λ the free space wavenumber and ns the refractive index of the substrate. Modes can be found as wavelength dependent solutions for kxy with the condition for wave guiding being .The maximum number of absorption peaks that can be added to the spectrum, and therefore the effectiveness of the concept, depends on the number of pseudo-guided modes and of wave vectors under which coupling is possible. By solving Eqs. (1) and (2), it can be determined that the minimum thickness for at least one mode to be able to propagate is about 120 nm; for a patterned CIS layer with VFF = 25% and for wavelengths between 500 nm and 1300 nm. At least one higher order mode can propagate for thicknesses above 600 nm.
For determining the diffractive properties of the PhC pattern, we calculate its 2D power spectral density (2D-PSD). This approach can be applied not only to simple periodical patterns, but also to more complex photonic structures. The 2D-PSD is the squared absolute of the structures’ spatial Fourier transform. When discretizing a spatial pattern and denoting the height at a spatial position l and m with hlm, the 2D-PSD is given by [48]
where, νx and νy are the spatial frequency in the patterned plane, L is the length of the pattern and ΔL the discretization step. Mode coupling is possible where kxy is equal to a peak in the PSD. For a proof of concept sample we target thicknesses which allow for coupling incident light to at least two modes at the main diffraction order (1,0). This is valid for a layer thickness of 560 nm and a lattice constant of 650 nm.Figure 4(a) shows the dispersion relation of the TE and TM guided mode for a layer of 560 nm thickness and a VFF of 25%. Below the light line of the fused silica substrate, modes are no longer confined. Three TE modes and three TM modes are supported for kxy > 18 rad/µm, and two of each for kxy > 9 rad/µm. The 2D-PSD of a 2D PhC with a lattice constant of 650 nm was calculated according to Eq. (3). Figure 4(b) shows the 1D-PSD(νxy) with for the maximum of 2D-PSD(νx,νy). Four peaks, for which quasi-guided mode coupling is possible, can be found in the relevant range at angular spatial frequencies of 9.7 rad/µm, 13.7 rad/µm, 19.3 rad/µm and 21.6 rad/µm.
Fig. 4 (a) Dispersion relation for TE (blue) and TM (red) guided modes calculated for a 560 nm thick PhC structured CIS layer on fused silica, with a VFF of 25%, approximated as a homogenous layer of averaged refractive index. (b) One dimensional power spectral density (1D-PSD) calculated for the same PhC structured layer with a lattice constant of 650 nm. The numbers in brackets indicate the diffraction order. (c) Simulated absorption spectra of the infiltrated absorbing PhC (black) and of a planar reference of identical thickness (red) under normal incidence illumination. The three graphs ((a), (b), and (c)) combined show the relation between the spatial frequency of the PhC pattern and the resulting waveguide coupling condition (vertical green), and the wavelength at which absorption enhancement occurs (horizontal green). (d) Contour plot of the absorption enhancement as defined in Eq. (4) as a function of the lattice constant and CIS layer thickness. The VFF was kept constant at 25%.
The FDTD calculated absorption spectrum for this design, shown in Fig. 4(c), exhibits two absorption peaks for the structured sample, at 1110 nm and 960 nm. These coincide with the predicted coupling wavelength for the first order diffraction mode with the zeroth order and the first order waveguide modes, respectively. This correspondence is visualized by the green guides-to-the-eye in the graph. Outside the low absorption regime (λ ≤ 800 nm), another extended absorption peak is visible around the wavelength of 650 nm. Without deeper knowledge of the coupling strength of each mode, a direct correlation of this peak to a resonant mode is difficult, as absorption is significant enough at these wavelengths for many approximations to fail.
For a more complete picture, various combinations of the lattice constant and CIS layer thicknesses were investigated using FDTD simulations. We define absorption enhancement (AE) as the integrated absorption spectrum (A) weighted by the spectral irradiance (I) (AM 1.5g) and normalized to an unstructured CIS film of identical thickness:
The absorption enhancement as a function of the CIS layer thickness and the lattice constant is reported in Fig. 4(d). A clear tendency can be seen towards larger optimum lattice constants for increasing film thickness. As the wavelength range, where efficient absorption enhancement is possible, is moved to longer wavelength, lower spatial frequencies are needed for efficient coupling. The oscillations in the contour of the enhancement map are attributed to the dips in the AM 1.5g spectrum around 940 nm and 1130 nm.3.3 Fabrication and characterization
The targeted lattice constants of above 500 nm are well achievable with the DLW system used in this work to fabricate PhCs. How the pattern fabrication is done was described in detail in section 2.2, it is illustrated in Fig. 5(a) An SEM image of the resulting nanocone array with a lattice constant of 650 nm can be seen in Fig. 5(b). The exact dimensions of the nanocones were determined using AFM. The height profile of a single nanocone is shown in the inset of Fig. 5(b). As expected from the fabrication method the nanocones are of half-ellipsoidal shape. They possess a base diameter of 450 nm and a height of 520 nm. Therefore, they can be embedded completely within the targeted 560 nm thick CIS layer with a VFF of about 24%.
Fig. 5 (a) Schematic of the DLW process used for fabricating the square lattice of nanocones. (b) 30° tilted view SEM image of the fabricated PhC pattern with a lattice constant of 650 nm. Inset: AFM height profile of a single nanocone.
Thin CIS nanocrystal films are deposited and infiltrated into the patterned photoresist by doctor blading as illustrated in Fig. 6(a). A small drop of CIS nanocrystals ink is cast onto the substrate near the patterned area and then swept over the structures. Solvent remaining in the layer dries under ambient conditions, allowing the formation of compact layers over a few cm2. Two consecutive deposition steps are needed to achieve the targeted layer thickness of 560 nm. The result is shown in Fig. 6(b). It was obtained by cutting through the whole infiltrated PhC using FIB. We observe that the nanocones are perfectly embedded within the CIS layer, and that the latter is homogeneous and free of air voids. The inset to Fig. 6(b) shows a TEM image of the nanocrystals.
Fig. 6 (a) Schematic of the doctor blading step used for infiltrating the CIS nanocrystal ink into the nanocone array. (b) 54° tilted view SEM image of a FIB cut through the nanocones, buried in a dense CIS layer. Inset: TEM image of the CIS nanocrystals.
To prove that the additional absorption peaks predicted by optical simulation (see Fig. 4(c)) are also obtained in the fabricated samples, the absorption spectrum of the above-described infiltrated PhC is measured. As all of our simulations were conducted under normal incidence illumination, this condition has also to be fulfilled in the experimental setup for proper analysis. The spectra are taken in the center of an integrating sphere. Under normal incidence, the specularly reflected light escapes from the sphere and is therefore not accounted for. However, this configuration enables us to compare the number and position of the measured and simulated new absorption peaks. This is demonstrated in Fig. 7. The two main absorption peaks, arising from the PhC patterning and already predicted in Fig. 4, can be found at the wavelengths of 960 nm and 1110 nm in both the measured and simulated spectra. An additional peak at 650 nm also stands out from the absorption “background” in both cases, but has only a limited contribution to the integrated absorption increase.
Fig. 7 Simulated absorption spectra of the PhC structured CIS nanocrystal layer (solid black) of thickness 560 nm and of a planar reference (dashed black) of identical thickness (dashed black), shown together with the measured spectrum of the infiltrated PhC demonstrator (solid red). All spectra were obtained under normal incidence.
Thus, we show experimentally that new absorption peaks can be introduced in the targeted spectral range after tailoring the infiltrated PhC design numerically, and by considering the CIS nanocrystal optical indices measured by ellipsometry. Our approach leads to an increase in absorption, integrated over 300 nm to 1300 nm, of 3°% - 7% for layer thicknesses above 300 nm in the here investigated system (compare Fig. 4(d)). Essentially, this limited gain is attributed to the low refractive index of the CIS nanocrystals. Improvements should be possible by optimizing the PhC structure and moving from a square lattice to more profound architectures. However, this will not change the fact that only a modest refractive index contrast is obtained between the polymer template and the infiltrated CIS (Δn ≈0.5, see section 3.1), limiting the coupling efficiency. Additionally, the low refractive index of the CIS nanocrystals limits the number of modes that can propagate, and thereby the number of new absorption peaks that can be introduced in the spectrum. For these reasons, it can be anticipated that higher optical gains are at reach using higher refractive index absorbers. In the case of CIS nanocrystals, this could be achieved by either completely removing the oleylamine ligands, for example using thermal treatment. Another approach is performing a ligand-transfer, ideally employing ligands that have less effect in the CIS band-transitions (compare section 3.1). An increase from n ≈2 to n ≈3 should be possible, in the best case. Assuming an increase to n = 2.6 is possible, our model already predicts an absorption enhancement of 40% for a 300 nm layer on fused silica structured into a PhC of 500 nm lattice constant. Lastly we note that in a complete solar cell stack, where the infiltrated PhC is sandwiched between a mirror and a transparent conductive oxide layer, the mode confinement will be more efficient than in the present configuration, and so will be the coupling of light with pseudo-guided modes.
4. Conclusion
In this work, we introduced a light-trapping design for improving absorption in thin PV absorbers made of CuInSe2 nanocrystals. Unlike previous approaches involving the top-down structuration of an absorber by dry etching techniques, we proposed to form an absorbing PhC by infiltrating the CIS nanocrystal ink into a periodical array of photoresist nanocones.
Prior to that, the optical indices of the CIS layers were measured by spectroscopic ellipsometry and revealed weakly absorbing properties between 500 nm and 1300 nm in the required thin films, calling for the assistance of a light-trapping structure. Using optical simulations based on the FDTD method, we numerically tailored the geometrical parameters of the infiltrated PhC to enhance its absorption in the targeted spectral range. In addition, we demonstrated that an analytical model based on the homogenous layer approximation can be used to determine the frequencies where the incoming light will couple to the pseudo-guided modes within the PhC, and hence the position of the new absorption peaks. This was achieved by deriving the guided modes dispersion relation and by analyzing the 1D-PSD of the PhC pattern.
To test our method experimentally, we developed a fabrication route by employing DLW to structure a polymer layer into a square array of nanocones with a lattice constant of 650 nm, following the design rules established numerically. The subsequent CIS ink infiltration into this array formed the absorbing PhC consisting of photoresist nanocones completely embedded into a compact, homogeneous nanocrystal layer. The measured absorption spectrum of the infiltrated PhC indicated the presence of new absorption peaks in the desired spectral region, with a good agreement with both the optical simulations and the analytical predictions. A maximum overall integrated absorption enhancement of 3% - 7% was determined for CIS layer thickness above 300 nm. We identified the low refractive index of the CIS nanocrystals as the main limiting factor in the present design, as it leads to both a moderate coupling efficiency and a restricted number of pseudo-guided modes.
Consequently, we advocate the use of higher refractive index nanocrystals, which under experimentally relevant assumptions, should achieve a substantial absorption enhancement of 40%. A complementary solution consists in enriching the PSD spectrum by designing more complex infiltrated photonic structures. Owing to the versatility of the DLW technique, such configurations are directly accessible using the approach presented herein.
Funding
Helmholtz Association via the i) Helmholtz Young Investigator Group of U. Paetzold; the ii) the Helmholtz Postdoc Program (for G. Gomard); iii) the Professorial Recruitment Initiative for B. S. Richards; as well as iv) the Helmholtz program Science and Technology of Nanosystems (STN).
Acknowledgments
Portions of this work were presented at the OSA Light, Energy and the Environment Congress in 2016, paper PW3B.2. Patrice Brenner, from the Center for Functional Nanostructures (CFN), is acknowledged for his assistance with the FIB cut and subsequent SEM observations of the samples. Furthermore this work was supported by the Karlsruhe Nano-Micro Facility (KNMF), the Karlsruhe School of Optics & Photonics (KSOP), and the Young Investigator Network (YIN) of the Karlsruhe Institute of Technology (KIT).
References and links
1. H. Azimi, Y. Hou, and C. J. Barbec, “Towards low-cost, environmentally friendly printed chalcopyrite and kesterite solar cells,” Energy Environ. Sci. 7(6), 1829–1849 (2014). [CrossRef]
2. C. Eberspacher, C. Fredric, K. Pauls, and J. Serra, “Thin-film CIS alloy PV materials fabricated using non-vacuum, particles-based techniques,” Thin Solid Films 387(1–2), 18–22 (2001). [CrossRef]
3. M. Powalla and B. Dimmler, “Scaling up issues of CIGS solar cells,” Thin Solid Films 361, 540–546 (2000). [CrossRef]
4. M. G. Panthani, V. Akhavan, B. Goodfellow, J. P. Schmidtke, L. Dunn, A. Dodabalapur, P. F. Barbara, and B. A. Korgel, “Synthesis of CulnS2, CulnSe2, and Cu(InxGa(1-x))Se2 (CIGS) nanocrystal “inks” for printable photovoltaics,” J. Am. Chem. Soc. 130(49), 16770–16777 (2008). [CrossRef] [PubMed]
5. V. R. Voggu, J. Sham, S. Pfeffer, J. Pate, L. Fillip, T. B. Harvey, R. M. Brown Jr, and B. A. Korgel, “Flexible CuInSe2 Nanocrystal Solar Cells on Paper,” ACS Energy Lett. 2(3), 574–581 (2017). [CrossRef]
6. V. A. Akhavan, M. G. Panthani, B. W. Goodfellow, D. K. Reid, and B. A. Korgel, “Thickness-limited performance of CuInSe2 nanocrystal photovoltaic devices,” Opt. Express 18(19), A411–A420 (2010). [CrossRef] [PubMed]
7. S. M. McLeod, C. J. Hages, N. J. Carter, and R. Agrawal, “Synthesis and characterization of 15% efficient CIGSSe solar cells from nanoparticle inks,” Prog. Photovolt. Res. Appl. 23(11), 1550–1556 (2015). [CrossRef]
8. M. G. Panthani, C. J. Stolle, D. K. Reid, D. J. Rhee, T. B. Harvey, V. A. Akhavan, Y. Yu, and B. A. Korgel, “CuInSe2 Quantum Dot Solar Cells with High Open-Circuit Voltage,” J. Phys. Chem. Lett. 4(12), 2030–2034 (2013). [CrossRef] [PubMed]
9. National Renewable Energy Laboratory, “ASTM G173-03 Reference Spectra,” http://rredc.nrel.gov/solar/spectra/am1.5/ASTMG173/ASTMG173.html.
10. S. Fan and J. D. Joannopoulos, “Analysis of guided resonances in photonic crystal slabs,” Phys. Rev. B 65(23), 235112 (2002). [CrossRef]
11. G. Gomard, R. Peretti, E. Drouard, X. Meng, and C. Seassal, “Photonic crystals and optical mode engineering for thin film photovoltaics,” Opt. Express 21(9), A515–A527 (2013). [CrossRef] [PubMed]
12. A. Chutinan and S. John, “Light trapping and absorption optimization in certain thin-film photonic crystal architectures,” Phys. Rev. A 78(2), 023825 (2008). [CrossRef]
13. K. R. Catchpole and M. A. Green, “A conceptual model of light coupling by pillar diffraction gratings,” J. Appl. Phys. 101(6), 063105 (2007). [CrossRef]
14. D. H. Ko, J. R. Tumbleston, L. Zhang, S. Williams, J. M. DeSimone, R. Lopez, and E. T. Samulski, “Photonic crystal geometry for organic solar cells,” Nano Lett. 9(7), 2742–2746 (2009). [CrossRef] [PubMed]
15. D. Duché, E. Drouard, J. J. Simon, L. Escoubas, P. Torchio, J. Le Rouzo, and S. Vedraine, “Light harvesting in organic solar cells,” Sol. Energ. Mat. Sol. 95, S18–S25 (2011). [CrossRef]
16. G. Gomard, X. Meng, E. Drouard, K. E. Hajjam, E. Gerelli, R. Peretti, A. Fave, R. Orobtchouk, M. Lemiti, and C. Seassal, “Light harvesting by planar photonic crystals in solar cells: the case of amorphous silicon,” J. Optics-Uk 14(2), 024011 (2012). [CrossRef]
17. S. Basu Mallick, M. Agrawal, A. Wangperawong, E. S. Barnard, K. K. Singh, R. J. Visser, M. L. Brongersma, and P. Peumans, “Ultrathin crystalline-silicon solar cells with embedded photonic crystals,” Appl. Phys. Lett. 100(5), 053113 (2012). [CrossRef]
18. E. Garnett and P. Yang, “Light trapping in silicon nanowire solar cells,” Nano Lett. 10(3), 1082–1087 (2010). [CrossRef] [PubMed]
19. J. Bhattacharya, N. Chakravarty, S. Pattnaik, W. D. Slafer, R. Biswas, and V. L. Dalal, “A photonic-plasmonic structure for enhancing light absorption in thin film solar cells,” Appl. Phys. Lett. 99(13), 131114 (2011). [CrossRef]
20. U. W. Paetzold, W. Qiu, F. Finger, J. Poortmans, and D. Cheyns, “Nanophotonic front electrodes for perovskite solar cells,” Appl. Phys. Lett. 106(17), 173101 (2015). [CrossRef]
21. S. Jeong, E. C. Garnett, S. Wang, Z. Yu, S. Fan, M. L. Brongersma, M. D. McGehee, and Y. Cui, “Hybrid silicon nanocone-polymer solar cells,” Nano Lett. 12(6), 2971–2976 (2012). [CrossRef] [PubMed]
22. G. Gomard, E. Drouard, X. Letartre, X. Q. Meng, A. Kaminski, A. Fave, M. Lemiti, E. Garcia-Caurel, and C. Seassal, “Two-dimensional photonic crystal for absorption enhancement in hydrogenated amorphous silicon thin film solar cells,” J. Appl. Phys. 108(12), 123102 (2010). [CrossRef]
23. A. Bozzola, M. Liscidini, and L. C. Andreani, “Photonic light-trapping versus Lambertian limits in thin film silicon solar cells with 1D and 2D periodic patterns,” Opt. Express 20(6), A224–A244 (2012). [CrossRef] [PubMed]
24. A. Naqavi, F.-J. Haug, K. Söderström, C. Battaglia, V. Paeder, T. Scharf, H. P. Herzig, and C. Ballif, “„Angular behavior of the absorption limit in thin film silicon solar cells,” Prog. Photovolt. Res. Appl. 22(11), 1147–1158 (2014). [CrossRef]
25. U. W. Paetzold, S. Lehnen, K. Bittkau, U. Rau, and R. Carius, “Nanoscale observation of waveguide modes enhancing the efficiency of solar cells,” Nano Lett. 14(11), 6599–6605 (2014). [CrossRef] [PubMed]
26. U. W. Paetzold, S. Lehnen, K. Bittkau, U. Rau, and R. Carius, “Nanoscale investigation of polarization-dependent light coupling to individual waveguide modes in nanophotonic thin-film solar cells,” IEEE J. Photovolt. 5(6), 1523–1527 (2015). [CrossRef]
27. A. Naqavi, K. Söderström, F.-J. Haug, V. Paeder, T. Scharf, H. P. Herzig, and C. Ballif, “Understanding of photocurrent enhancement in real thin film solar cells: towards optimal one-dimensional gratings,” Opt. Express 19(1), 128–140 (2011). [CrossRef] [PubMed]
28. C. Lin, L. J. Martínez, and M. L. Povinelli, “Experimental broadband absorption enhancement in silicon nanohole structures with optimized complex unit cells,” Opt. Express 21(18), A872–A882 (2013). [CrossRef] [PubMed]
29. M. Meier, U. W. Paetzold, M. Ghosh, and R. van Erven, “Periodic nano-textures enhance efficiency in multi-junction silicon thin-film solar cells,” Phys. Status Solidi 212(1), 30–35 (2015). [CrossRef]
30. X. Meng, G. Gomard, O. El Daif, E. Drouard, R. Orobtchouk, A. Kaminski, A. Fave, M. Lemiti, A. Abramov, P. Roca i Cabarrocas, and C. Seassal, “Absorbing photonic crystals for silicon thin-film solar cells: Design, fabrication and experimental investigation,” Sol. Energy Mater. Sol. Cells 95, S32–S38 (2011). [CrossRef]
31. X. Sheng, J. Liu, I. Kozinsky, A. M. Agarwal, J. Michel, and L. C. Kimerling, “Design and non-lithographic fabrication of light trapping structures for thin film silicon solar cells,” Adv. Mater. 23(7), 843–847 (2011). [CrossRef] [PubMed]
32. S. Jeong, L. Hu, H. R. Lee, E. Garnett, J. W. Choi, and Y. Cui, “Fast and scalable printing of large area monolayer nanoparticles for nanotexturing applications,” Nano Lett. 10(8), 2989–2994 (2010). [CrossRef] [PubMed]
33. M. D. Kelzenberg, S. W. Boettcher, J. A. Petykiewicz, D. B. Turner-Evans, M. C. Putnam, E. L. Warren, J. M. Spurgeon, R. M. Briggs, N. S. Lewis, and H. A. Atwater, “Enhanced absorption and carrier collection in Si wire arrays for photovoltaic applications,” Nat. Mater. 9(3), 239–244 (2010). [PubMed]
34. J. Wallentin, N. Anttu, D. Asoli, M. Huffman, I. Åberg, M. H. Magnusson, G. Siefer, P. Fuss-Kailuweit, F. Dimroth, B. Witzigmann, H. Q. Xu, L. Samuelson, K. Deppert, and M. T. Borgström, “InP Nanowire Array Solar Cells Achieving 13.8% Efficiency by Exceeding The Ray Optics Limit,” Science 339(6123), 1057–1060 (2013). [CrossRef] [PubMed]
35. V. Depauw, X. Meng, O. El Daif, G. Gomard, L. Lalouat, E. Drouard, C. Trompoukis, A. Fave, C. Seassal, and I. Gordon, “Micrometer-thin crystalline-silicon solar cells integrating numerically optimized 2-D Photonic crystals,” IEEE J. Photovolt. 4(1), 215–223 (2014). [CrossRef]
36. C. Trompoukis, O. El Daif, P. P. Sharma, H. S. Radhakrishnan, M. Debucquoy, V. Depauw, K. Van Nieuwenhuysen, I. Gordon, R. Mertens, and J. Poortmans, “Passivation of photonic nanostructures for crystalline silicon solar cells,” Prog. Photovolt. Res. Appl. 23(6), 734–742 (2015). [CrossRef]
37. M. Smeets, K. Bittkau, F. Lentz, A. Richter, K. Ding, R. Carius, U. Rau, and U. W. Paetzold, “Post passivation light trapping back contacts for silicon heterojunction solar cells,” Nanoscale 8(44), 18726–18733 (2016). [CrossRef] [PubMed]
38. E. R. Martins, J. Li, Y. Liu, V. Depauw, Z. Chen, J. Zhou, and T. F. Krauss, “Deterministic quasi-random nanostructures for photon control,” Nat. Commun. 4, 2665 (2013). [CrossRef] [PubMed]
39. H. B. Sun and S. Kawata, “Two-photon photopolymerization and 3D lithographic microfabrication,” Adv. Polym. Sci. 170, 169–273 (2006). [CrossRef]
40. G. Demésy and S. John, “Solar energy trapping with modulated silicon nanowire photonic crystals,” J. Appl. Phys. 112(7), 074326 (2012). [CrossRef]
41. Lumerical Solutions Inc, “Lumerical,” https://www.lumerical.com.
42. H. Nanoscribe Gmb, “Nanoscribe,” http://www.nanoscribe.de/files/1814/0662/4393/IP-Resist_IP-Dip_web.pdf, (acquired 03/2016).
43. M. I. Alonso, K. Wakita, J. Pascual, M. Garriga, and N. Yamamoto, “Optical function and electronic structures of CuInSe2, CuGaSe2, CuInS2, and CuGas2,” Phys. Rev. B 63(7), 075203 (2001). [CrossRef]
44. P. D. Paulson, R. W. Birkmire, and W. N. Shafarman, “Optical characterization of CuIn1-xGaxSe2 alloy thin films by spectroscopic ellipsometry,” J. Appl. Phys. 94(2), 879–888 (2003). [CrossRef]
45. G. Yin, C. Merschjann, and M. Schmid, “The effect of surface roughness on the determination of optical constants of CuInSe2 and CuGaSe2 thin films,” J. Appl. Phys. 113(21), 213510 (2013). [CrossRef]
46. S. Dottermusch, A. Quintilla, G. Gomard, U. W. Paetzold, and B. S. Richards, “Simulation of absorption enhancement and optical modes in CIS nanocrystal embedded photonic crystal designs,” in Light, Energy and the Environment, OSA Technical Digest (online) (Optical Society of America, 2016), paper PTu2B.3.
47. S. G. Tikhodeev, A. L. Yablonskii, E. A. Muljarov, N. A. Gippius, and T. Ishihara, “Quasiguided modes and optical properties of photonic crystal slabs,” Phys. Rev. B 66(4), 045102 (2002). [CrossRef]
48. C. Ruppe and A. Duparré, “Roughness analysis of optical films and substrates by atomic force microscopy,” Thin Solid Films 288(1), 8–13 (1996). [CrossRef]
49. M. Adams, An Introduction to Optical Waveguides (John Wiley and Sons Inc., 1981).
