Here we demonstrate that 4-beam holographic lithography can be utilized to create plasmonic nanogaps that are 70 times smaller than the laser wavelength (488 nm). This was achieved by controlling phase, polarization, and laser beam intensity in order to tune the relative spacing of the two sublattices in the interference pattern of a compound-lattice in combination with the nonlinear resist response. Exemplarily, twin and triplet motive features were designed and patterned into polymer in a single exposure step and then transferred into gold nanogap arrays resulting in an average gap size of 22 nm and smallest features down to 7 nm. These results extend the utility of high-throughput, wafer-scale holographic lithography into the realm of nanoplasmonics.

COMMUNICATIONThe authors obtained strong and spectrally broad absorption in ultrathin (<25 nm) germanium fi lms on gold substrates due to the characteristic phase shifts induced by metals. Esfandyarpour et al. [ 14 ] have recently used a similar mechanism based on a metamaterial mirror to optimize the position of standing waves inside an organic solar cell. However, resonators with deep subwavelength dimensions, as required to obtain the spectrally broad enhancement reported by Kats et al., [ 13 ] were not subject of their work. It is still unclear how all the additional fi lms, required for electrical functionality of solar cells (e.g., doped layers, electrodes, absorber), can be integrated into a cavity having such dimensions.We answer this question by demonstrating the fi rst thin-fi lm solar cell taking advantage of the described effect. Specifi cally, we apply the concept to the case of amorphous germanium (a-Ge:H) solar cells. A short summary is given on how the phase shifts infl uence the resonance conditions in our particular solar cell stack. 1D optical simulations are carried out for thickness optimization and potential evaluation. A device with an absorber thickness of only 13 nm was fabricated, showing a remarkable short-circuit current density of 20 mA cm -2 and strong response in the infrared part of the spectrum.As stated above, the phase shifts play an important role in our solar cell design. Absorption enhancement is achieved in cases where a total phase difference of π arises between two incident waves refl ected at the fi rst and second mirror of the cavity, respectively. Both partial waves may undergo different phase shifts upon refl ection (φ 1 and φ 2 , respectively), and the difference Δφ = φ 2 − φ 1 directly contributes to the total phase difference. Restricting our calculations to -π≤Δφ<π, this leads to a required resonator length d max = (π− Δ φ)λ max /(4π n ) to obtain a lowest order maximum near wavelength λ max , where n is the real part of the refractive index of the absorber material. Similar arguments lead to d min = (2π−Δφ)λ min /(4π n ) for the lowest order minimum. As φ 1 and φ 2 are not necessarily equal to 0 or π for nonperfect materials, Δφ can take arbitrary values. Thus, d max can be reduced by adjusting Δφ to approach (but not reach) π. At the same time, the ratio λ min /λ max = (π−Δφ)/ (2π−Δφ) (calculated for fi xed d and neglecting dispersion) indicates that a broadening of the interference patterns is obtained in this case. The above ratio also evidences a reduced sensitivity of the resonance conditions to angle of incidence variations and surface roughness, as a non-normal incidence can be described by a reduced wave vector in normal direction, translating to an increased wavelength in our formulas. This effect has also been observed experimentally by Kats et al. [ 13 ] .Of course, the absorber needs to provide suffi cient single pass absorption at the low thicknesses required here. For example, experimental results by Kempa et al. [ 15 ] on hot

Plasmonic and photonic light trapping structures can significantly improve the efficiency of solar cells. This work presents an experimental and computational comparison of identically shaped metallic (Ag) and nonmetallic (SiO2) nanoparticles integrated to the back contact of amorphous silicon solar cells. Our results show comparable performance for both samples, suggesting that minor influence arises from the nanoparticle material. Particularly, no additional beneficial effect of the plasmonic features due to metallic nanoparticles could be observed.

No abstract