AVS2015 Session EM+EN-ThA: Materials for Light Management

Semiconductor nanowires are filamentary crystals with a tailored diameter between few to few hundred nanometer. Their special shape and dimensions render them especially interesting for photonic applications. In my talk I will discuss several photonic applications of nanowires. I will start by showing how to modify light absorption and emission of nanowires by coupling them to plasmonic elements [1]. I will then follow by explaining the use of III-V nanowires for solar cell applications. I will show how, by choosing the adequate diameter and length, it is possible to obtain absorption cross-sections much larger than the nanowire physical size. This concentration effect can be used to increase the efficiency of nanowire-based solar cells and to reduce considerably the use of materials [2].

References

[1] A. Casadei et al , Scientific Reports 5, 7651 (2015)

[2] P. Krogstrup et al, Nature Photon. 7, 306 (2013) ; M. Heiss et al, Nanotech. 25, 014015 (2014)

GaAs nanowires (NWs) are expected to applying to optoelectronic devices. However, material properties have not been understood yet due to a presence of impurity and defect level related to an involved catalyst which was inevitable for growing nanowires. Recently, we succeeded in fabricating the catalyst-free GaAs NWs on a (111) Si substrates using a combination of molecular beam epitaxy and vapor-liquid-solid method [1]. The NW had two kinds of crystalline phases, a zinc-blend (ZB) and a wurtzite (WZ) structures [2]. We found the amount of WZ phase increased with increasing the amount of Si-doping by using high-resolution X-ray diffraction and transmission electron microscope [3]. In this study, we investigate the electronic band structure of Si-doped GaAs NWs by using a photoreflectance (PR) and a photoluminescence (PL) techniques and discuss the effect of Si-doping on the optical properties.

Three kinds of samples with different Si cell temperatures at 1015, 1065, and 1150ºC were grown. The average diameter and length of NWs were 60 nm and 35 mm, respectively. The amount of Si doping was evaluated by a carrier concentration estimated from a Hall measurement. The lowest hole concentration was 5.5 ×10¹⁷ cm^-3 for the sample grown at 1015 ºc and increased about an order by increasing the cell temperature. PR and the PL emission light were carried at 4 K.

The crystallographic investigations showed that the amount of secondary WZ phase increased with increasing the Si-doping. Three critical energies were observed at 1.51, 1.49 and 1.43 Ev in the PR. The first two signals were observed for all samples and attributed to band to band and band to Si acceptor transitions, respectively. Since the signal at 1.43 Ev appeared only in high Si-doping sample with high amount of WZ phases, this is due to the band to band transition at the interface between the two different crystalline phases. In the PL spectra, three other emission peaks were observed at 1.46, 1.42 and 1.37eV. The intensities of these peaks changed for the samples with different cell temperature, these may be due to impurity and defect levels in nanowires. Si dopant itself as well as different crystal structure affect the electron transition. Since the interface transition observed at 1.43 Ev becomes dominant, emissions through such impurities were hidden for the sample with highest Si-doping. Temperature dependences of the PR and PL spectra will be discussed for further understanding the effect of Si-doping.

[1] J. H. Paek et al., Phys. Stat. Sol. (c) 6, 1436 (2009).

[2] D. Spirkoska et al., Phys. Rev. B 80, 245325 (2009).

[3] A. Suzuki et al., Jpn. J. Appl. Phys. 54, 035001 (2015).

In this regard, the nanowire (NW) geometry offers potential advantages in the solar energy conversion process. Due to their richness in structure and morphology combined with doping and bandgap engineering, NWs provide an opportunity for new charge separation mechanisms.

In this work we propose a new IB-based solar cell design that is advantageously benefited from nanostructuring. We propose the use of core-shell heterostructures in order to reduce the optical coupling between the different band states in a three-level IBSC. Taking advantage of the intrinsic anisotropy of the nanowire geometry, the fundamental idea here is that excitons are separated by the heterostructure along the radial direction, while carrier extraction is performed along the axial direction. As a result, mid-gap recombination rate is reduced by several orders of magnitude.

On the other hand, it is known that optical resonances in NWs result into light self-concentrating effects that allow high absorption with reduced material. What’s more, light resonances in NWs leave a very specific spatial distribution of light inside the nanostructure. By tuning the geometrical parameters of the NW, one can guide light around different regions in the NW depending on the photon energy. A combination of both electrical and morphological engineering, can lead to high efficiency PV.

Insertion of a multiple quantum-well (MQW) structure into the absorbing layer of solar cells is promising for accomplishing higher conversion efficiency. Recently, a MQW with a very thin barrier structure has been proposed to enhance the conversion efficiency [1]. The coupling of the wave functions between adjacent quantum wells causes a mini-Brillouin zone along the growth direction, which results in the formation of miniband. We have discussed carrier escaping mechanism in MQW by using photoreflectance (PR) and photothermal spectroscopy (PPT) and found that internal electric field in the QW region might affect the recombination probability [2]. In this study, we investigate the effect of internal electric filed on the miniband width.

Three kinds of MQW samples were grown by a metal-organic vapor phase epitaxy technique. Two GaAs p-i-n solar cell structure samples with MQWs in the i-layer were prepared. The doping levels in the p- and n-type layer were changed and this induced the different strength of the electric field in the absorbing layer. Another sample had GaAs n-n structures without any electric field in the absorbing layer. The thicknesses of the well barrier were changed from 2 to 6 nm for discussing the detailed miniband formation. PR and PPT measurements were carried out at room temperature. The miniband width was estimated from the difference of the peak energies of the PR modulus spectra. The lower and higher energy peak correspond to the energies of bottom and top of the miniband, respectively. The PPT method is used to detect the heat generated by non-radiative recombination of photo generated carriers. The miniband width was also calculated from the Gaussian decomposition technique of the observed PPT spectra [3].

[1] Y. Wang, et al., J. Cryst. Growth 352, 194 (2012).

[2] T. Aihara et al., J. Appl. Phys. 116, 044509 (2014).

[3] T. Aihara et al., J. Appl. Phys. 117, 084307 (2015).

Luminescent solar concentrators (LSCs) offer many advantages over traditional concentrator geometries. As opposed to concentrators that rely on tracking, LSCs operate under both direct and diffuse illumination and require the solar cell to only be efficient at a small range of wavelengths. In practice, most LSCs suffer losses from reabsorption of emitted light, non-unity quantum yields, and incomplete light guiding to the solar cell. Here we combine tunable lumophores based on quantum dot heterostructures with photonic designs to improve the concentration efficiency of LSCs and study light propagation within the device.

To achieve high concentration ratios it is critical to have high effective Stokes shifts, high quantum yields, and to reduce escape cone losses. We synthesized a series of core-shell nanocrystal lumophores that exhibit tunable emission. We show how the narrow emission bandwidth of these nanocrystals enables the use of a 1D photonic mirror on the top surface of the device, designed to admit incident sunlight and trap luminesced light from the nanocrystal. In combination, the concentration ratios from these devices exceed the performance of dyes with higher quantum yields but broader emission. Another approach to photonic LSCs is to restrict the angle of emission from the lumophores to promote coupling to the total internal reflection modes of the LSC. This talk will discuss designs for the latter approach and compare achievable concentration factors.

The second portion of the talk will discuss light management strategies for solid-state lighting, and the incorporation of plasmonic nanostructures to enhance light outcoupling from solid-state devices. This section of the talk will compare and contrast plasmonic structures for solar cells and solid-state lighting, and discuss ways that design rules should be adjusted for different materials systems.