AVS2014 Session EN-TuP: Energy Frontiers Poster Session

Embedding of multi quantum well (MQW) structures in an absorption layer of GaAs solar cell is a promising idea for developing higher efficient solar cell devices. Extension of the absorbing region to longer wavelength side enhances the short-circuit current. When the mini-band forms between the QW, further developing of the conversion efficiency would be expected in term of a carrier tunneling without recombination of the photo excited carriers. However, recombination at the QW interface still leads to the degradation of the efficiency. Therefore, we fabricated a strain-balanced InGaAs/GaAsP layer in the intrinsic region of the GaAs p-i-n solar cell [1]. The mini-band formation for the samples with different thickness of the barrier region from 5.3 to 1.9 nm was studied by using a photoreflectance (PR) and a surface photovoltage (SPV) spectroscopy at room temperature.

The energies at the Brillion zone edges were detected from the Kramers-Kronig analysis of the PR spectra for estimating the mini-band width as a function of barrier thickness. When the mini-band formed, the ratio of carriers escaped thermally from each QW may change. SPV spectra were, then, measured to know the carrier escaping rate and how this component affects the absorption spectra of the solar cell structures.

The observed PR spectra for the sample with thinnest barrier width of 1.9 nm show two critical energies at 1.274 and 1.312 eV. The energy difference of these critical energies was around 0.038 eV and this energy difference reduces from 0.020 to 0.009 eV for the samples with barrier thickness of 2.7 and 3.6 nm, respectively. Supposing that such energy difference shows the mini-band width, the reduction of the energy with increasing the thickness of the barrier can be understood. The theoretical estimation using a Kronig-Penny model calculation for the width suggests that the mini-band width in the QW is 70 and 0 meV for e1 and topmost hh1 sub-bands. Therefore, the observed mini-band width is smaller than expected. The reason is not clear at present. For the SPV spectra, two broad peaks were observed at 1.29 and 1.32 eV. The energy difference also decreases with increasing the thickness of the barriers. Smaller critical energies for SPV may be explained that these signals detect the maxima energy of the joint-density of states which is larger than the band edge energies. Since the carrier escaping from the QW and radiative and non-radiative recombination rates may be strongly temperature dependent, estimation of the relevant activation energies for carrier escaping from these measurements were useful for further discussion.

In the last decades, the deposition of CIGSe thin films by sputtering has seldom been investigated. Besides, the sputtering of a single quaternary CIGSe target would be a real advantage for industrial development. Indeed, sputtering technique exhibits a good compatibility with industrial up-scaling and limits selenium use with respect to toxicity issues. In 1992, Hernandez et al¹ early published on CIGSe layers deposited by sputtering of a single quaternary CIGSe target.

The improvement of plasma discharges at lower working pressures in the last years had allowed news possibilities for CIGS thin films deposited by sputtering. In 2010, Frantz et al² succeeded to obtain a CIGSe solar cell with an efficiency of 8.9%.

At IMN laboratory, a dedicated chamber has been home-designed for CIGSe thin films deposition using one step sputtering. CIGSe thin films were deposited on SLG/Mo substrates by radio-frequency magnetron sputtering at ambient temperature. The deposition parameters like pressure or power density have been investigated. Then, CIGS thin films have been annealed at various temperatures in a furnace under Ar flow. The evolution of chemical composition and structural properties of CIGS thin films as-deposited and annealed will be presented.

References

1. Hernández Rojas, J. L. et al.. Appl. Phys. Lett. 60, 1875–1877 (1992)

2. Frantz, J. A. et al. Thin Solid Films 519, 7763–7765 (2010).

In the framework of an emerging Solar Fuel program the first and essential CO₂ dissociation step of CO₂ into CO and O₂ has been studied. The focus was thereby on high energy efficiency of the process using non-thermal (non-equilibrium) microwave plasmas without the use of rare materials. The plasma was generated by tangentially injecting CO₂ gas in a quartz tube placed inside a low loss 915 MHz TM₀₁₀ circular waveguide mode cavity. Powers were used up to 10 kW. The feed gas expanded supersonically after the cavity to quench the plasma and prevent vibrational-translational relaxation losses. These experiments resulted in power efficient conversion (more then 50%) of large CO₂ flows (up to 75 standard liter per minute) with 11% conversion yield. The product formation was measured downstream the plasma at the expansion in a calibrated mass spectrometer.

The research at the plasma Solar Fuels facilities of DIFFER concentrates on optimization of power efficient selective vibrational CO dissociation. This can be achieved by means of several novel methods: like tuning the optimal kinetic energy range of the electrons by controlling the reduced electric field in the microwave cavity or make use of the relaxation time difference of translational and vibrational modes of CO controlled by modulating the microwave field. First results were obtained with a 2.45 GHz 1kW microwave plasma source. The source consists of a quartz tube with a tangential gas injection system placed inside a circular TEM mode coaxial cavity coupled to a TE₁₀ rectangular waveguide mode cavity. The configuration of the coaxial cavity allows control of the electric field. With an input of 7 standard liter per minute of CO₂ energy efficiencies 15% to 36% were obtained (based on CO₂ depletion) in the pressure range from 1Bar (atmospheric) till 0.2 Bar. An update on these new research developments will be shown.

The use of solution-based methods for semiconducting thin film deposition is of great importance for large area photovoltaics applications, enabling as well the deposition on flexible substrates. Chalcogenide semiconductor materials present p- and n-type characteristics that are attractive for photovoltaics applications. The optimization of device structure for a solar cell requires an atomistic understanding of the structure and chemical reactions taking place during deposition and post-deposition processes. In this regard, X-ray photoelectron spectroscopy (XPS) is uniquely suited for tracking chemical reactions occurring at the surface and interface of sub-nanometric layers. The chemical analysis and thermal stability for surface and interface reactions of binary and ternary chalcogenide semiconductors is presented. CuS, ZnS, SnS₂, CdS thin films and some of its combinations were deposited by the successive ionic layer absorption and reaction method (SILAR) under controlled environment next to the XPS analysis load-lock chamber. Step by step XPS analysis of the SILAR process reveals an incubation period that depends on temperature and ion concentrations. The stability of the heterostructures is discussed in terms of chemical reactivity and inter-diffusion.

We have recently reported that impurity-doped ZnO thin films appropriate for transparent electrode applications in thin-film solar cells were related more to the content rather than the kind of impurity doped into the films as well as to the deposition method used [1]. This paper describes the influence of the orientation and size in crystallite on the resulting optical absorption and the obtainable surface texture structure and carrier mobility in Al-doped ZnO (AZO) thin films prepared with different types of magnetron sputtering deposition (MSD) methods. The AZO thin films were prepared with a thickness of 500-3500 nm on OA-10 glass substrates at a sputter Ar gas pressure of 0.2-12 Pa and a temperature of room temperature (RT)-350^oC by d.c. MSD or r.f. power superimposed d.c. MSD. The surface texture formation was carried out as a result of MSD under the appropriate deposition condition or by wet-chemical etching (in a 0.1% HCl solution at 25^oC). For example, when the AZO films were prepared with increasing a sputter gas pressure from 0.4 to 12 Pa at a deposition temperature of 350^oC, the films exhibited the red shift with tailing in both the absorption edge and plasma edge as the sputter pressure was increased. With increasing the sputter gas pressure, the deposited films exhibited the increase of etching rate and the decrease of etch pit size obtained from the etching, whereas the pyramid type textured surface was formed at a pressure above approximately 6 Pa. It was also found that both in the c-axis orientation and the crystallite size as well as the carrier mobility in the deposited films were decreased by increasing the sputter gas pressure. It should be noted that the sputter gas pressure dependence described above was strongly dependent on the thickness of deposited films and MSD method used. In addition, the result described above was considerably affected by after heat-treatment with rapid thermal annealing (RTA) in air. Consequently, observed sputter gas pressure dependences such as the expanded tailing of absorption coefficient, etching characteristics and decrease of mobility were mainly correlated to the degradation of crystallinity as evidenced by the c-axis orientation and the crystallite size.

[1] T. Minami, T. Miyata, and J. Nomoto, Materials Science and Engineering, 34, 012001 (2012).

In a little over a year, there has been an unexpected breakthrough and rapid evolution of highly efficient solid-state hybrid solar cells based on organometal trihalide perovskite materials. This technology has the potential to produce solar cells with the very highest efficiencies while retaining the very lowest cost. We have measured the electronic density of states of CH₃NH₃PbI₃ using ultraviolet photoelectron spectroscopy (UPS), inverse photoemission spectroscopy (IPES), and x-ray photoelectron spectroscopy (XPS). The valence band maximum (VBM) and conduction band minimum (CBM) positions are obtained from the UPS and IPES spectra, respectively, by liner extrapolation of the leading edges. With the Fermi level close to the VBM, the sample is slightly p-type, with the E_VBM=0.76 eV, E_CBM=-0.9 eV, and transport energy gap 1.7 eV. The ionization potential (IP) of 6.16 eV can be obtained from the sum of the UPS measured E_VBM and the work function of 5.40 eV. The XPS spectra reveal an obvious deficiency of N is, whereas the concentrations of I and Pb are close to the expected values. The existence of O and excessive C indicate that the surface is contaminated, and the contamination is reduced by 12-20% by thermal annealing in vacuum. The interface between CH₃NH₃PbI₃ and TiO₂ is also investigated. As CH₃NH₃PbI₃ layer thickness increases, one sees a gradual shift of the vacuum level cut-off. At the CH₃NH₃PbI₃ thickness of 8.2 nm, the shift saturates, signaling the end of the interface dipole formation. The evolution of the VB is even more gradual, and it appears to mature only at the thickest layer. Given the UPS probing depth of ~1.5 nm and the typical interface formation in other systems, the saturation of the vacuum level cut-off at 8.2 nm in CH₃NH₃PbI₃/TiO₂ is surprisingly large. It points at the possibility that the films are not truly uniform. On the other hand, the interface dipole of 0.7 eV CH₃NH₃PbI₃ and TiO₂ is important for further understanding of the energy level alignment and charge transfer across the interface.

In the present work, we synthesized and systematically characterized two novel graphene-based nanomaterials, a reduced graphene oxide sponge and a microwave-exfoliated graphene oxide. Their textural properties were determined by N₂ adsorption/desorption at 77 K, while additional characterization techniques were employed in order to elucidate further their structure, surface chemistry and morphology such as X-Ray Diffraction (XRD), Fourier-Transform Infrared Spectroscopy (FT-IR), Field-Emission Scanning Electron Microscopy (FE-SEM) and High-Resolution Transmission Electron Microscopy (HR-TEM). Both nanomaterials were additionally evaluated for their H₂ storage performance and were critically compared to commercially available carbons (e.g. graphene nanoplatelets, carbon nanotubes) based on systematic H₂ adsorption/desorption measurements at 77K between 0-1 bar. Maximum H₂ gravimetric capacities ~0.5 wt% and ~0.7 wt% were recorded at 77 K and 1 bar for the reduced GO sponge and the microwave-exfoliated GO, respectively. A linear relationship was found between the H₂ uptake values and the BET specific surface area of the materials included in this study.

The high efficiency which some organisms perform photosynthesis and its abundance around the globe, coupled with high world energy demand in the coming years and the low competitiveness of current photovoltaic devices in comparison with other forms of energy, inspired us to build a Bio-photovoltaic device composed of a heavily p-doped silicon substrates and a tandem of protein complex called Photosystem 1 (PS1) found in the thylakoid of leaves.

The process extraction of PS1 complexes from spinach essentially consists of two general steps: the extraction of thylakoid membranes from the spinach leaves and the isolation and purification of PS1 complexes from the thylakoids.

For the deposition of PS1 films onto silicon substrates we use a simple and straightforward method. The procedure entails depositing a volume (100µL) of an aqueous PS1 suspension on to a silicon surface modified with 3-aminopropyl-triethoxysilane (APTS) and applying a negative pressure (~70 mTorr) to remove the solvent from the PSI solution via vacuum deposition process. The thermodynamic driving force for the formation of these films arises from hydrophobic interactions between neighboring protein complexes in the film and by tuning the silicon Fermi energy by doping. This promotes the alignment of silicon bands with redox active sites of PS1, which leads to the formation of dense and resilient films.

Iterative depositions were similarly performed and for two deposition steps, a thickness of ~600nm was achieved, thereby increasing the efficiency of the bio-hybrid cell.

These results represent significant progress toward affordable, biologically-inspired renewable energy conversion platforms.

Scanning capacitance microscopy (SCM) is a powerful method for mapping dopant variation in semi-conductor and PV devices. In SCM a conducting AFM tip is used to scan the surface in contact mode. The tip and sample surface form a metal–insulator–semiconductor structure. Simultaneous with the topography data collection, an AC bias is applied to the sample while the tip is grounded. The resulting oscillation of carriers near the tip, leads to a modulated capacitance (dC/dV), which is measured by a capacitance-sensing circuit. The capacitance measured by the SCM sensor varies as the carriers move towards (accumulation) and away from (depletion) the probe. When the sample is fully depleted the measured capacitance is that of the oxide plus the depletion layer. When carriers are accumulated at the surface, the measured capacitance is that of the oxide layer. The magnitude of the change in capacitance (dC) for a given change in voltage (dV)depends on the carrier concentration. For heavily doped materials the carriers do not move far. Hence, the measured capacitance variation between accumulation and depletion is small. The opposite is true for lightly doped semiconductors which yield a large capacitance change. The sign of the measured dC/dV signal changes between n-type and p-type. In p-type semiconductors, as the sample voltage becomes more negative relative to the tip, the width of the surface depletion layer will increase. Hence, the total capacitance will decrease. If the sample voltage is positive, accumulation will occur, and the capacitance measured will be that of the oxide layer only. For n-type samples the effects are opposite.