Midgap Level Research Articles

On the investigation of electronic defect states in ZnO thin films by space charge spectroscopy with optical excitation

Electronic defect states in a n-type conducting zinc oxide thin film sample were investigated by means of space charge spectroscopy focussing on levels in the midgap region as well as on hole traps. To overcome the experimental difficulties arising from the wide bandgap and the lack of p-type conduction, optical excitation was employed to measure the emission of trapped charge carriers from these levels. Therefore – besides deep-level transient spectroscopy measurements – photo-capacitance, optically chopped photo-current, minority carrier transient spectroscopy, and optical capacitance–voltage experiments were conducted. In doing so, a midgap level labelled T4, and hole traps labelled TH1 and TH2 were detected. In the case of T4 and TH1 the photo-ionisation cross-section spectra were determined.

Photoluminescence imaging of electronic-impurity-induced exciton quenching in single-walled carbon nanotubes

The electronic properties of single-walled carbon nanotubes can be altered by surface adsorption of electronic impurities or dopants. However, fully understanding the influence of these impurities is difficult because of the inherent complexity of the solution-based colloidal chemistry of nanotubes, and because of a lack of techniques for directly imaging dynamic processes involving these impurities. Here, we show that photoluminescence microscopy can be used to image exciton quenching in semiconducting single-walled carbon nanotubes during the early stages of chemical doping with two different species. The addition of AuCl(3) leads to localized exciton-quenching sites, which are attributed to a mid-gap electronic impurity level, and the adsorbed species are also found sometimes to be mobile on the surface of the nanotubes. The addition of H(2)O(2) leads to delocalized exciton-quenching hole states, which are responsible for long-range photoluminescence blinking, and are also mobile.

Defect states of complexes involving a vacancy on the boron site in boronitrene

substitutional impurities are the most stable structures, irrespective of the changes in growth conditions. The high energies of formation of the boron vacancy complexes suggest that they are less stable, and their creation by ion bombardment would require high-energy ions compared to point defects. Using the relative positions of the derived midgap levels for the double vacancy complex, it is shown that the quasi-donor–acceptor pair interpretation of optical transitions is consistent with stimulated transitions between electron and hole states in boronitrene.

Nitrogen-DopedTiO2Photocatalyst Prepared by Mechanochemical Method: Doping Mechanisms and Visible Photoactivity of Pollutant Degradation

Nitrogen-doped TiO2(N/TiO2) photocatalysts were prepared using a mechanochemical method with raw amorphous TiO2as precursors and various nitrogenous compounds doses (NH4F, NH4HCO3, NH3·H2O, NH4COOCH3, and CH4N2O). The photocatalysts were characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), thermal gravimetric-differential thermal analysis (TG-DTA), and UV-Vis diffuse reflection spectra (UV-Vis-DRS). Their photocatalytic activities were evaluated with the degradation of p-nitrophenol and methyl orange under UV or sunlight irradiation. The catalysts had a strong visible light absorption which correspond to doped nitrogen and consequent oxygen deficient. The results of photocatalytic activity showed the visible light adsorption mechanisms, as the doped nitrogen species gave rise to a mid-gap level slightly above the top of the (O-2p) valence band, but not from the mixed band gap of the N-2p and O-2p electronic levels.

Photoconductivity Mapping of Semi-Insulating CdZnTe

Semi-insulating CdZnTe crystals were studied by photoconductivity mapping using both the contactless method and measurement with evaporated Au contacts. The contact quality was tested by measurement of voltampere characteristics by a three-point method. We evaluated the distribution of space charge analyzing the slope of lux-ampere characteristics. Mobility- lifetime (μτ) product maps were extracted fitting the Hecht relation to voltampere characteristics measured with a weak light at wavelength 750 nm. Correlation analysis of contactless resistivity and photoconductivity maps shows, that both these parameters are anticorrelated. This fact can be explained by a shift of Fermi level changing the average occupation of a midgap level. A decrease of occupation with an increasing resistivity results in an increase of electron trapping and decreased photoconductivity. This qualitative explanation is supported by simulating the dependence of the photocurrent density on the Fermi level position near the midgap using parameters of midgap levels assumed in state-of-the art radiation detectors. The map of electron mobility-lifetime product correlates with the maps of photocurrent intensity measured with and without the Au contact. The comparison of maps displaying slopes of lux-ampere characteristics with the map of μτ shows, that areas with a higher accumulation of space charge have a smaller μτ .

Mechanistic Study of Visible-Light-Induced Photodegradation of 4-Chlorophenol by TiO2−xNxwith Low Nitrogen Concentration

TiO<sub >2&#x2212;<i>x </i></sub>N<i >_x</i> powders with low N-doping concentrations (<svg style="vertical-align:-0.30096pt;width:117.825px;" id="M1" height="11.0625" version="1.1" viewBox="0 0 117.825 11.0625" width="117.825" xmlns="http://www.w3.org/2000/svg"> <g transform="matrix(1.25,0,0,-1.25,0,11.0625)"> <g transform="translate(72,-63.15)"> <text transform="matrix(1,0,0,-1,-71.95,63.5)"> <tspan style="font-size: 12.50px; " x="0" y="0">0</tspan> <tspan style="font-size: 12.50px; " x="6.2515001" y="0">.</tspan> <tspan style="font-size: 12.50px; " x="9.3772497" y="0">0</tspan> <tspan style="font-size: 12.50px; " x="15.62875" y="0">2</tspan> <tspan style="font-size: 12.50px; " x="21.880251" y="0">1</tspan> <tspan style="font-size: 12.50px; " x="31.595081" y="0">&lt;</tspan> <tspan style="font-size: 12.50px; " x="43.635471" y="0">𝑥</tspan> <tspan style="font-size: 12.50px; " x="53.987953" y="0">&lt;</tspan> <tspan style="font-size: 12.50px; " x="66.028343" y="0">0</tspan> <tspan style="font-size: 12.50px; " x="72.279846" y="0">.</tspan> <tspan style="font-size: 12.50px; " x="75.405594" y="0">0</tspan> <tspan style="font-size: 12.50px; " x="81.657089" y="0">4</tspan> <tspan style="font-size: 12.50px; " x="87.908592" y="0">9</tspan> </text> </g> </g> </svg>) were prepared by annealing commercial TiO<sub >2</sub> (P-25) under an NH<sub >3</sub> flow at 550&#xb0;C. Regardless of UV or visible case, the photoactivities of the samples decreased as <i >x</i> increased, and TiO<sub >1.979</sub>N<sub >0.021</sub> showed the highest activity for the 4-chlorophenol (4-CP) decomposition under the visible-light irradiation. The visible-light response for N-doped TiO<sub >2</sub> could arise from an N-induced midgap level, formed above the valence band (O 2p). Electron spin resonance (ESR) measurements and the radical scavenger technologies gave the combined evidence that the active species (<svg style="vertical-align:-0.17554pt;width:30.862499px;" id="M2" height="10.9125" version="1.1" viewBox="0 0 30.862499 10.9125" width="30.862499" xmlns="http://www.w3.org/2000/svg"> <g transform="matrix(1.25,0,0,-1.25,0,10.9125)"> <g transform="translate(72,-63.27)"> <text transform="matrix(1,0,0,-1,-71.95,63.5)"> <tspan style="font-size: 12.50px; " x="0" y="0">•</tspan> <tspan style="font-size: 12.50px; " x="6.5390692" y="0">O</tspan> <tspan style="font-size: 12.50px; " x="15.566235" y="0">H</tspan> </text> </g> </g> </svg> and <svg style="vertical-align:-3.13504pt;width:33.787498px;" id="M3" height="14.6" version="1.1" viewBox="0 0 33.787498 14.6" width="33.787498" xmlns="http://www.w3.org/2000/svg"> <g transform="matrix(1.25,0,0,-1.25,0,14.6)"> <g transform="translate(72,-60.32)"> <text transform="matrix(1,0,0,-1,-71.95,63.5)"> <tspan style="font-size: 12.50px; " x="0" y="0">O</tspan> </text> <text transform="matrix(1,0,0,-1,-62.92,60.37)"> <tspan style="font-size: 8.75px; " x="0" y="0">2</tspan> </text> <text transform="matrix(1,0,0,-1,-58.05,63.5)"> <tspan style="font-size: 12.50px; " x="0" y="0">•</tspan> </text> <text transform="matrix(1,0,0,-1,-51.51,68.5)"> <tspan style="font-size: 8.75px; " x="0" y="0">−</tspan> </text> </g> </g> </svg>) are responsible for the photodecomposition of 4-CP over TiO<sub >2&#x2212;<i>x</i></sub>N<i >_x</i> under the visible irradiation. A possible photocatalytic mechanism was discussed in detail.

Characterization and Photocatalytic Activity of N/TiO&lt;sub&gt;2&lt;/sub&gt; Prepared by a Mechanochemica1 Method Using Various Nitrogenous Compounds

Nitrogen doped TiO2 (N/TiO2) photocatalysts were prepared using a mechanochemica1 method with raw amorphous TiO2 as precursors and various nitrogenous compounds doses (NH4F, NH4HCO3, NH3·H2O, NH4COOCH3, and CH4N2O). The photocatalysts were characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and UV-Vis diffuse reflection spectra (UV-Vis-DRS). Their photocatalytical activities were evaluated with the degradation of p-nitrophenol and methyl orange under UV or sunlight irradiation. The catalysts had a strong visible light absorption which corresponding to doped nitrogen and consequent oxygen deficient. Concentration of the nitrogen substances also affected the visible light adsorption of the TiO2. The XPS results indicated the doped nitrogen in the TiO2 may exist in the formation of O2Ti−N−H and/or O2Ti−N, nitrogen was in form of interstitial atom. The results of photocatalytic activity showed the visible light adsorption mechanisms, as the doped nitrogen species gave rise to a mid-gap level slightly above the top of the (O-2p) valence band, but not from the mixed band gap of the N-2p and O-2p electronic levels.

Tailoring the surface properties and carrier dynamics inSnO2 nanowires

We report a study of the role of mid-gap defect levels due to surface states inSnO2 nanowires on carrier trapping. Ultrafast pump–probe spectroscopy provides carrierrelaxation time constants that reveal the nature and positions of various defect levels dueto the surface states which in turn provide details on how the carriers relax after theirinjection. The effect of oxygen annealing on carrier concentration is also studied throughXPS valence band photoemission spectroscopy, a sensitive non-contact surfacecharacterization technique. These measurements show that charge transfer associated withchemisorption of oxygen in different forms produces an upward band bending and leads toan increase in the depletion layer width by approximately 70 nm, thereby decreasing surfaceconductivity and forming the basis for the molecular sensing capability of the nanowires.

Determination of bandgap states in p-type In0.49Ga0.51P grown on SiGe/Si and GaAs by deep level optical spectroscopy and deep level transient spectroscopy

The presence and properties of traps in p-type In0.49Ga0.51P grown on low dislocation density, metamorphic Ge/SiGe/Si substrates and GaAs substrates were determined using deep level transient spectroscopy (DLTS) and deep level optical spectroscopy (DLOS) leading to the quantification of trap behavior throughout the entire 1.9 eV bandgap of the In0.49Ga0.51P material as a function of substrate. Thermal emission-based DLTS revealed a single hole trap at Ev + 0.71 eV for growth on both lattice matched and mismatched substrates with similar concentrations. Complementary, optical emission-based DLOS measurements detected bandgap states at Ev + 1.18 eV, Ev + 1.36 eV, and Ev + 1.78 eV for p-type In0.49Ga0.51P grown on both substrate types. The total concentration of the DLOS-detected states was found to comprise approximately 80% of the entire trap concentration in p-type In0.49Ga0.51P bandgap. This relatively high concentration of above midgap levels may be of great significance for minority carrier devices that utilize p-type In0.49Ga0.51P (such as high efficiency III–V multijunction solar cells) since their position in the bandgap and high concentrations suggest that strong minority carrier electron trapping behavior can be expected. The primary effect of substituting the GaAs substrate by Ge/SiGe/Si is to increase the concentration of these states by a factor of 2–3, with no additional levels detected due to the replacement by the Si-based substrates, indicating that all detected traps are native to the epitaxial In0.49Ga0.51P material (regardless of the substrate), but whose concentrations appear to be influenced by dislocation density.

Electrical characterization of as-grown and H plasma treated CdTe epitaxial layers

In the present work deep level transient spectroscopy (DLTS) and high-resolution Laplace DLTS were employed to study the electrical properties of defects in as-grown and H-plasma treated CdTe thin films grown by the molecular beam epitaxy technique. We demonstrate that two dominant deep states in as-grown CdTe have the activation energy close to the midgap level and, therefore, could act as recombination centres in these crystals. Both traps are donor-like defects. However, their concentration was found to be too small to influence the electrical properties of devices based on this semiconductor. On the other hand, another dominant midgap trap was observed in H-plasma treated CdTe. The question as to the origin of these defects will be addressed.

Dislocation-induced electronic levels in semi-insulated CdTe

We studied deformation-induced defects in semi-insulating CdTe and CdZnTe by infrared photoluminescence (PL) and compared our data with earlier results. We confirmed the direct correlation between Y-emission and dislocation density in both compounds. The Y-band intensified near an indenter deformation or near a scribing line, but was barely visible in low-dislocation areas (etch pit density <2×10 5 cm −2). Plastic deformation also increased the concentrations of grown-in defects, namely, those of an important midgap level E C−0.74 eV in CdTe and Cd 1− x Zn x Te ( x<0.1), the materials of choice in today’s detector technology. Our findings demonstrate that dislocation-induced defects can degrade charge collection in radiation detectors.

Highly Visible Light Active TiO2−xNx Heterojunction Photocatalysts

Nitrogen doped anatase-rutile heterojunctions are successfully synthesized through an ethylenediaminetetraacetic acid (EDTA) modified sol−gel process. An FT-IR study of EDTA modified TiO2 gel confirms the existence of an ionic intermediate (as indicated by a Δν value of 233 cm−1). Differential scanning calorimetry (DSC), X-ray diffraction (XRD), and Raman spectroscopy are employed to study the phase evolution, phase purity, and crystallite size of samples. Formations of O−Ti−N and N−Ti−N bonds in calcined samples are confirmed using XPS and FT-IR spectroscopy. All EDTA modified samples show significantly higher visible light photocatalytic activity than the unmodified sample. The most active nitrogen doped heterojunction obtained at 400 °C exhibits 9-fold visible light activity in comparison to the standard photocatalyst Degussa P-25. It is proposed that the photo excited electrons (from the visible midgap level) are effectively transferred from the conduction band of anatase to that of rutile causing eff...

Photoluminescence and electric spectroscopy of dislocation-induced electronic levels in semi-insulated CdTe and CdZnTe

We studied deformation-induced defects in semi-insulating CdTe and CdZnTe by infrared photoluminescence (PL), contact less photoconductivity and resistivity. Plastic deformation increased the concentrations of grown-in defects, namely, those of an important midgap level E C−0.74 eV in CdTe and Cd 1− x Zn x Te ( x<0.1), the materials of choice in today’s X-ray and gamma ray detector technology. We confirmed the direct correlation between Y-emission and the dislocation density in both compounds. The Y-band intensified near an indenter deformation or near a scribing line, but was barely visible in low-dislocation areas (etch pit density <2×10 5 cm −2). Our results correlate with recent findings that dislocation-induced defects and their clusters degrade charge collection in radiation detectors. Photoluminescence of midgap levels can serve as a tool to identify areas of degraded performance in semi insulated CdTe and CdZnTe crystals for radiation detectors.

A New SiGeC Vertical MOSFET: Single-device CMOS (SD-CMOS)

AbstractA new type of silicon-based Vertical MOSFET concept is presented - Single-Device CMOS (SD-CMOS) - in which the same structure can be operated as NFET or as PFET, depending on the biasing conditions [1]. SD-CMOS offers new possibilities for CMOS integration schemes that are simpler - requiring only 5 masks for the “Front- End” - and less costly to manufacture, than any integration scheme requiring the fabrication of two devices with opposite polarities.In epitaxially grown Vertical MOSFETs, the source, channel and drain can have atomically sharp interfaces, well controlled doping, and channel length controlled by the epitaxial process rather than by lithography and ion-implantation. With epitaxial growth, it is straightforward to do bandgap engineering by incorporating films such as Si1-xGex, and/or Si1-yCy and/or Si1-x-yGexCy, into any of the aforementioned regions. Suitable band offsets at the source/channel interface [2] can suppress DIBL, which is a key limitation to CMOS scaling. These advantages of Vertical MOSFETs are crucial for future CMOS technology nodes, such as 22nm and below.SD-CMOS has a metallic drain region with work-function close to the mid-gap energy of the channel material, which can be a homogenous material, a random alloy, or a superlattice. The source region has a very narrow bandgap, achievable with (Si1-yCy)m-(Si1-xGex)n superlattices, whose mid-gap level is aligned with that of the channel region, and with band offsets with the channel region that are nearly symmetric for the conduction and valence bands. The source contact is a metal with a work-function close to the mid-gap level of the source and channel regions, which in turn are aligned with the work-function of the drain. The potential barrier, for electrons and holes from the source contact to the source region, is required to be just a few KT. For operation as NMOS and PMOS with nearly symmetric threshold voltages, the work-function of the gate electrode is also aligned with the mid-gap energy level of the channel.SD-CMOS is unique due to its band alignments: symmetric band edges, from source to drain, with respect to the mid-gap energy (the “mirror” line). Such configuration can only be obtained in the absence of doping, which if present would immediately break that symmetry. The conduction and valence band edges are required to be asymmetric with respect to a cross-section line crossing the channel through the middle of the gate.The conduction (valence) band offset between source and channel sets the barrier height for electrons (holes) in the OFF condition for NMOS (PMOS), while applying a voltage at the gate leads to the accumulation of electrons (holes) at the source/channel interface, thereby pushing the Fermi-Level in the source above (below) the conduction (valence) band edge of the channel for the ON condition. Band diagrams and a CMOS fabrication flow for SD-CMOS will be presented. [1] US Patent 6,674,099 [2] US Patent 5,914,504

Doping single-walled carbon nanotubes through molecular charge-transfer: a theoretical study

We study the effect of the molecular charge transfer on the electronic structure of metallic (5,5) and semiconducting (8,0) single-walled carbon nanotubes (SWNTs) induced by surface adsorption of various organic donor-acceptor molecules of different affinities using ab initio density functional theory. Our results, obtained from first-principles spin-polarized calculations show that the adsorption of molecules with different affinities reflects the difference in interaction strength that measure the overall energy of adsorption. Moderate values of the binding energy of these surface adsorbed molecular charge-transfer complexes suggest that the nature of interaction is in the physisorption regime, and mainly governs by Coulombic forces. We also find that the large band gap of semiconducting (8,0) SWNT can be tuned through the surface adsorption of selective organic molecules which gives rise to mid-gap localized molecular levels near the Fermi energy with tuning of band gap region. Interestingly, we find that the metallic (5,5) SWNT and semiconducting (8,0) SWNT turn into semiconducting and metallic nanotubes respectively in presence of selective surface adsorbed molecules, corroborating recent experimental findings. We also suggest that these charge transfer effect can be probed through optical conductivity measurement, as the low-frequency profiles are affected by charge transfer.

Studies of the electronic and vibrational signatures of the unusual bonding geometries in melt-quenched amorphous silicon

Tight-binding molecular dynamics simulations have been performed to investigate the effect of quenching rate of the Si melt on the resulting local structure of amorphous silicon. Different quenching rates were used to cool liquid silicon in the simulations to demonstrate that the choice of quenching rates significantly influences the resulting local structure. The calculated pair correlation functions show that the local structure is sensitive to the thermal processing of the liquid silicon melt. The use of cooling rates higher than 10−13 K s−1 appears to prevent the activation of the required structural re-arrangements necessary to stabilise the networks, causing unexpected bonding geometries to develop. The electronic signatures of the defects show that only the triangular defect structure contributes resonance states to the conduction band tail. Also, the vibrational signature of the triangular structure shows a high energy transverse optical mode at 95 meV, indicating that the defect is likely to be unstable at 300 K, although both defects contribute minimal states to the mid-gap level.

Materials, Devices, and Circuits of Transparent Amorphous-Oxide Semiconductor

This paper presents the following recent investigations of transparent amorphous-oxide semiconductors (TAOS) from materials to devices and circuits. 1) Composition of metals in TAOS are widely explored with the aim of seeking semiconductors suitable for the channel layers of thin-film transistors (TFTs) composing backplanes for flat-panel displays. It is found in combinatorial approaches to the materials exploration that indium-based ternary TAOS (In-X-O) and their TFTs show the properties and the performance as good as those of the most popular material of amorphous In-Ga-Zn-O (alpha-IGZO) when X = Zn or Ge. 2) Defects and impurities in TAOS are investigated by theoretical approaches. The first-principle calculation of the electron states reveals that charge-neutral oxygen vacancy or interstitial forms the density of states around mid-gap level and does not generate carriers directly, while hydrogen impurity raises the Fermi level beyond the conduction-band minimum and acts as a donor in TAOS. 3) Device structures of TAOS-TFTs are also investigated extensively for better performance and stability. It is found in channel-etch type TFTs with bottom-gate inverse-stagger structures that the TFT characteristics and stability are significantly improved by chemically removing the back-channel layer in a wet-etching process. Coplanar homojunction (CH) structure is proposed as a novel device structure where conductive alpha-IGZO regions work as the source and drain electrodes to the channel region of semiconductor alpha-IGZO. The CH TFTs show excellent characteristics and stability, with low series resistance without any difficulty in making good electrical contact between metals and TAOS. 4) Circuits using TAOS-TFTs are demonstrated. A ring oscillator composed of fifteen-stage inverters with a buffer circuit operates as designed by circuit simulation with a TFT model for hydrogenated amorphous Si TFTs. Pixel circuits composed of three TFTs and one transparent capacitor successfully drive organic light-emission diode cells without a planarization layer on a 2-in diagonal panel having 176 times144 times 3 pixels.

Compensation and Photosensitivity in CdTe Doped With Indium

To better our knowledge of the characteristics of semi-insulated cadmium telluride (CdTe) doped with indium (In), we explored the role of deep levels in compensation and trapping. We assessed the defects and their distribution across a wafer in several ways; by measuring dark resistivity and photosensitivity maps, photoluminescence, Photo-Induced Current Transient Spectroscopy (PICTS), and Thermoelectric Effect Spectroscopy (TEES). We determined that electron trapping to a near midgap level in CdTe:In begins when the Fermi-level lies above this level. We demonstrated first that a small movement (ap 1divide2 kT) of the Fermi-level downward significantly increases electron trapping. PICTS and TEES measurements confirmed the presence of a positively charge electron trap at <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">E</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</sub> -0.65 eV (plusmn 0.05 eV) with a high capture cross-section. This level transforms into a neutral one when the Fermi-level moves above it. Photoluminescence measurements detected this energy level that, when positively charged, was responsible for a 0.68-eV emission, while in a neutral state, it was accountable for an emission peak located at 0.87 eV. We discuss the nature of the deep donors, considering the latest ldquoab initiordquo calculations: also, the Te anti-site is compared to complex defects, such as H-O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Cd</sub> .

2D device modelling and finite element simulations for thin-film solar cells

Thin-film heterojunction solar cell devices are modelled in two dimensions, from fundamental material parameters, using the finite element method. The electrostatic potential is solved for, together with the quasi-Fermi levels, while optical absorption is calculated from n and k values of the materials used. In this implementation, all material parameters can be input as functions of spatial coordinates, which makes it very flexible when applied to materials with inhomogeneities. As an example of the model in use, it is applied to a thin-film solar cell based on a structure with a CIGS absorber layer, a CdS buffer layer and a ZnO/ZAO transparent front contact. The effects of spatial inhomogeneities in the band gap energy and in the mid-gap trap level density on device performance are simulated.

Impurity photovoltaic effect in GaAs solar cell with two deep impurity levels

The impurity photovoltaic (IPV) effect has been extensively investigated for silicon solar cells with indium impurities. A small positive effect on the efficiency was theoretically predicted, provided that efficient light trapping schemes are applied. Since IPV effects are predicted to be more pronounced in wider band gap materials and possibly also by introducing more than one impurity level, we carried out a numerical study on GaAs solar cells with copper impurities. Indeed, copper introduces multiple acceptor type mid-gap levels in GaAs, two of which are investigated here: one at 0.14 eV and another at 0.40 eV above the valence band edge. We used a solar cell device simulator (SCAPS) specially adapted to include the IPV effect. We varied the impurity concentration and the light trapping parameters of the device, and calculated the occupation probability of the impurity levels, the generation and recombination through these levels (thermal and optical) and the solar cell characteristics: quantum efficiency, QE( λ), short-circuit current J sc, open-circuit voltage V oc and efficiency η. A significant IPV effect with extended response in the infrared is calculated when efficient light trapping is present. When an internal reflection coefficient R=0.99 can be reached at both the front and the back surface of the solar cell, an increase of the short-circuit current densities 2 mA/cm 2 can be obtained. The consequences of two levels instead of one are calculated and discussed, leading to a trade-off between J sc improvement and a V oc decrease.