Carrier Capture Cross Sections Research Articles

Investigation of silicon surface passivation by sputtered amorphous silicon and thermally evaporated molybdenum oxide films using temperature- and injection-dependent lifetime spectroscopy

Crystalline silicon (c-Si) surface passivation has been investigated by sputtered hydrogenated intrinsic amorphous silicon (S-i-a-Si:H) and thermally evaporated molybdenum oxide (MoOx) thin films. The temperature- and injection-dependent lifetime spectroscopy technique has been adopted for analyzing the passivation quality of the c-Si surface, using parameters such as the minority carrier effective lifetime (τeff), the activation energy of surface/interface defect states (ΔE), and the electron to hole carrier capture cross-section ratio (k) at the interface. With S-i-a-Si:H films, a τeff of ∼70 µs and ΔE of ∼51 meV have been observed in comparison to a τeff of ∼110 µs and ΔE of ∼109 meV from the MoOx films. These entirely different parameters are an indication of the relatively strong carrier recombination with dense interface/surface states from the S-i-a-Si:H passivation layers. The S-i-a-Si:H layers are unable to minimize the c-Si surface trap states with the chemical passivation for reducing carrier recombination due to the generation of additional surface defect states by the sputtering damage. However, the MoOx layers show better c-Si surface passivation due to the reduction of majority carriers by the carrier inversion (field-effect passivation) and chemical passivation. This effect is clearly reflected with the opposite trend in the carrier capture analysis from S-i-a-Si:H and MoOx layers.

Comparison of three IBSCs in terms of photo-filling, capture cross section of carriers, the concentration of IB states and overlapping of absorption coefficients in the wide range of band gaps including GaAs

Intermediate-band solar cells, IBSCs, with single IB provide better utilization of the solar spectrum than single junction solar cells. In this study, Model−B IBSC was proposed, investigated and compared with previously used two models named here as Model−A and Model−C respectively. Detailed Balance Model was applied to satisfy Current Balance Equations with the optimum photo filling of IB states in Model−B. Optimum design parameters were obtained using 104 cm−1 of band-to-band absorption coefficient, α, in the wide range of band gap materials, overlapping absorption, photo filling, f, of IB states, capture cross-section ratio of carriers, concentration of IB states, and IB layer thickness, w, under 1–46,000 sun concentration, X. Maximum limiting efficiency value of 63.2% has been obtained with 1.95 eV of band gap, for fαw≥5 and 46,000X, under non-overlapping absorption. The peak efficiency of Model−B was 16% over the Model−A and 8% over the Model−C, for equal capture cross sections under large overlapping. The peak efficiency of Model−B was 35% over Model−A, at large non-equal capture cross-sections and large overlapping. Results predicted that the materials in the wide range of band gaps could be good candidates for fabricating IBSCs using Model−B with an efficiency of 50% or above at lower sun concentration, 10,000X, even under large overlapping.

Theoretical simulation of the degradation on GaAs sub-cell with different defects induced by 1MeV electron irradiation

The degradation on the GaAs sub-cell of GaInP/GaAs/Ge triple-junction solar cells caused by different types of defects induced by 1 MeV electron irradiation is studied by numerical simulation. The simulation model is introduced that includes geometric structure, material, doping concentration and electron irradiation defects. The degradation results of short-circuit current, open-circuit voltage, maximum power, conversion efficiency and external quantum efficiency versus different defects have been investigated, and the degradation mechanism is analyzed. The results show that when the defect concentration and carrier capture cross-section is the same, the related parameters degradation of GaAs sub-cell is more serious when the defect energy level is closed to the intrinsic Fermi energy level. Meanwhile, the degradation of related parameters induced by E2 (Ec-0.14 eV) and E5 (Ec-0.96 eV) defects are more serious than the degradation of related parameters induced by other defects. The research will provide a method to analyze the detailed degradation mechanism induced by electron irradiation and a reference to analyze the degradation of solar cells induced by space particles.

How Much Physics is in a Current–Voltage Curve? Inferring Defect Properties From Photovoltaic Device Measurements

Defect-assisted recombination processes are critical to understand, as they frequently limit the photovoltaic (PV) device performance. However, the physical parameters governing these processes can be extremely challenging to measure, requiring specialized techniques and sample preparation. And yet the fact that they limit performance as measured by current–voltage ( JV ) characterization indicates that they must have some detectable signal in that measurement. In this work, we use numerical device models that explicitly account for these parameters alongside high-throughput JV measurements and Bayesian inference to construct probability distributions over recombination parameters, showing the ability to recover values consistent with previously reported literature measurements. The Bayesian approach enables easy incorporation of data and models from other sources; we demonstrate this with temperature dependence of carrier capture cross-sections. The ability to extract these fundamental physical parameters from standardized, automated measurements on completed devices is promising for both established industrial PV technologies and newer research-stage ones.

Explosion Dependence of the Photoconductivity of Nanostructured CdZnS Films on the Excitation Time

The kinetics of photoconductivity excitation in pyrolytic CdZnS films is studied. It is established that the time dependences of the photocurrent in the investigated structures are of an explosion character. To explain these features, a model is proposed that considers the centers of localization for which the charge carrier capture cross section is functionally related to the optical modulation of the surface potential barrier.

Charge carrier trapping by dislocations in single crystal diamond

Charge carrier trapping in diamond crystals containing well-defined concentrations of dislocations was investigated by several complementary techniques. Samples with dislocation densities ndis between <1 × 107 and ≈1 × 109 cm−2 were grown heteroepitaxially on Ir/YSZ/Si(001). In optical pump–probe experiments, ambipolar diffusion coefficients were determined from the decay of light-induced transient free carrier gratings. Modeling their variation with excitation density yielded trapping cross sections σ of 29 and 10 nm for the dislocations and a stress-field-induced reduction in exciton binding energies from 80 to 73 and 60 meV at ndis = 1 × 108 and 1 × 109 cm−2, respectively. The lifetime measured by induced absorption scaled proportional to 1/ndis with absolute values ranging from 0.1 to 10 ns. In the electrical measurements on two sets of detector slices, electron–hole pairs were excited by α-particles and transport was measured separately for electrons and holes. Both types of carriers showed fast transient current signals. The time constant of the additional slow component exclusively seen for holes was in agreement with the activation energy of boron acceptors. Their concentration of ≈0.5 ppb yielded σ = 1.77 × 10−13 cm2 for charged point traps. Schubweg and carrier lifetime due to deep trapping roughly reproduced the 1/ndis trend. For electrons at 3 V/μm, a value σ = 40 nm was deduced. Cross sections for holes were significantly smaller. Differences in hole trapping between the samples are attributed to charging of dislocations controlled by chemical impurities. Increase in lifetime at high voltages is explained by reduced capture cross sections for hot carriers.

Impacts of carrier capture processes in the thermal quenching of photoluminescence in Al–N co-doped SiC

High concentrations of aluminum (Al) and nitrogen (N) dopants of 6H SiC have been achieved by a fast sublimation growth process. The Al–N co-doped 6H-SiC layer exhibits a strong light-blue photoluminescence emission at low temperatures due to emissions from DI centers and donor acceptor pairs (DAP). The photoluminescence quenching mechanisms of those emissions are different. The decrease of free carrier capture cross-section as temperature increases according to the cascade capture process causes quenching of the photoluminescence emission form DI centers. Emission from Al–N DAP centers exhibits an exponential quenching with activation energy of (95 ± 10) meV. This is attributed to a competing hole capture by non-radiative defect in a multiphonon emission process.

Effective and Noneffective Recombination Center Defects in Cu2ZnSnS4: Significant Difference in Carrier Capture Cross Sections

By combining the electron–phonon coupling effect and the static coupling formalism, we calculate, through the first-principles methods, the carrier capture cross sections of the three possible nonr...

Effects of oxygen related thermal donors on the performance of silicon heterojunction solar cells

Absract The influence of oxygen-related thermal donors (TDs) on the performance of silicon heterojunction (SHJ) solar cells was explored experimentally in this paper. It is found that a certain number of thermal donors could do much harm to the performance of SHJ solar cells. The efficiency of SHJ solar cells is usually reduced by a value of ~ 1% absolute in the case of thermal donors with a concentration of nearly 1015 cm−3. The microwave photoconductance decay and Hall Effect studies have proved that the TDs can significantly reduce the carrier lifetime of n-type silicon substrate, but have no influence on the carrier mobility. Deep level transient spectroscopy (DLTS) measurements have further demonstrated that the TDs cause an energy level at Ec-0.13 eV with carrier capture cross-section of 10−15 cm2, which is responsible for the reduction of carrier lifetime and solar cell efficiency.

Misfit dislocation–related deep levels in InGaAs/GaAs and GaAsSb/GaAs p–i–n heterostructures and the effect of these on the relaxation time of nonequilibrium carriers

A study of deep levels in InGaAs/GaAs and GaAsSb/GaAs p0–i–n0 heterostructures with misfit dislocations and identification of the effective defects responsible for the significant (by up to a factor of 100) decrease in the relaxation time of nonequilibrium carriers in the base layers (and in the related reverse recovery time) of InGaAs/GaAs and GaAsSb/GaAs high-voltage power p-i-n diodes is reported. Experimental capacitance–voltage characteristics and deep-level transient spectroscopy spectra of p+–p0–i–n0–n+ homostructures based on undoped GaAs layers without misfit dislocations and InGaAs/GaAs and GaAsSb/GaAs heterostructures with a homogeneous network of misfit dislocations, all grown by liquid-phase epitaxy, are analyzed. Acceptor defects with deep levels HL2 and HL5 are identified in GaAs epitaxial p0 and n0 layers. Dislocation-related electron and hole deep traps designated as ED1 and HD3 are detected in InGaAs/GaAs and GaAsSb/GaAs heterostructures. The effective recombination centers in the heterostructure layers, to which we attribute the substantial decrease in the relaxation time of nonequilibrium carriers in the base layers of p-i-n diodes, are dislocation-related hole traps that are similar to HD3 and have the following parameters: thermal activation energy Et = 845 meV, carrier capture cross-section σp = 1.33 × 10−12 cm2, concentration Nt = 3.80 × 1014 cm−3 for InGaAs/GaAs and Et = 848 meV, σp = 2.73 × 10−12 cm2, and Nt = 2.40 × 1014 cm−3 for the GaAsSb/GaAs heterostructure. The relaxation time of the concentration of nonequilibrium carriers in the presence of dislocation-related deep acceptor traps similar to HD3 was estimated to be 1.1 × 10−10 and 8.5 × 10−11 s for, respectively, the InGaAs/GaAs and GaAsSb/GaAs heterostructures and 8.9 × 10−7 s for the GaAs homostructure. These data correspond to the relaxation times of nonequilibrium carriers in the base layers of GaAs, InGaAs/GaAs, and GaAsSb/GaAs high-voltage power p-i-n diodes.

Large carrier-capture rate of PbI antisite in CH3NH3PbI3 induced by heavy atoms and soft phonon modes

The $\mathrm{P}{\mathrm{b}}_{\mathrm{I}}$ antisite was reported to be a possible deep-level recombination-center defect in $\mathrm{C}{\mathrm{H}}_{3}\mathrm{N}{\mathrm{H}}_{3}\mathrm{Pb}{\mathrm{I}}_{3}$ solar cells with a concentration higher than ${10}^{15}\phantom{\rule{0.16em}{0ex}}\mathrm{c}{\mathrm{m}}^{\ensuremath{-}3}$ under I-poor conditions. However, whether it is really an effective nonradiative recombination center and limits the photovoltaic efficiency depends also on its cross sections for capturing the electron or hole carriers, which is difficult to determine in both experiment and theory. Here we use a recently developed method to overcome the high computational cost of electron-phonon coupling calculation and implement it in the standard first-principles code quantum espresso so that we can calculate the carrier-capture cross sections effectively for the point defects in complicated semiconductors, such as $\mathrm{C}{\mathrm{H}}_{3}\mathrm{N}{\mathrm{H}}_{3}\mathrm{Pb}{\mathrm{I}}_{3}$. The calculation showed that both the hole- and the electron-capture cross sections and capture rates of $\mathrm{P}{\mathrm{b}}_{\mathrm{I}}$ are large relative to those of the point defects in conventional semiconductors, such as Si or GaP, which is attributed mainly to the heavy Pb-I atoms and the soft phonon modes in this Pb halide. Nonradiative recombination increases the thermal energy by exciting the phonons in the soft Pb-I lattice, whereas the contribution of the organic $\mathrm{C}{\mathrm{H}}_{3}\mathrm{N}{\mathrm{H}}_{3}$ group is negligible. Since $\mathrm{P}{\mathrm{b}}_{\mathrm{I}}$ has a higher concentration under I-poor conditions, especially when the semiconductor is $p$-type, our results suggest that the minority-carrier lifetime and thus the photovoltaic efficiency should be very limited in $p$-type $\mathrm{C}{\mathrm{H}}_{3}\mathrm{N}{\mathrm{H}}_{3}\mathrm{Pb}{\mathrm{I}}_{3}$ and $p$-type doping should be avoided for fabricating high-efficiency $\mathrm{C}{\mathrm{H}}_{3}\mathrm{N}{\mathrm{H}}_{3}\mathrm{Pb}{\mathrm{I}}_{3}$ solar cells under I-poor conditions. Similar calculations can be used for studying the influence of various defects on the photovoltaic performance of other organic-inorganic hybrid and inorganic halide perovskites, shedding light on the design of high-efficiency solar cells.

Application of the Newton–Raphson Method to Lifetime Spectroscopy for Extraction of Defect Parameters

Defect parameter solution surface is a widely accepted method to determine the values of the energy level and the carrier capture cross sections of a defect from injection-dependent lifetime measurements. Results of the defect parameter solution surface come from the recognition of intersection points, which can be difficult when the doping or the temperature range of the measurement is narrow. In this paper, we introduce a new method to extract the defect parameters from lifetime spectroscopy data. After linearization of the Shockley–Read–Hall (SRH) lifetime equation, the problem is transformed into a mathematical form, which can be solved by the Newton–Raphson method. This method requires a much shorter computing time and directly provides the possible values of the defect parameters without user interaction, even for measurements with narrow doping and temperature ranges.

Influence of Se beam pressure on deep-level defects in Cu(In,Ga)Se_2 thin films studied by photocapacitance and time-resolved photoluminescence measurements

The properties of deep-level defects in Cu(In,Ga)Se2 thin films grown under different Se beam pressure were investigated by photocapacitance (PC) and time-resolved photoluminescence (TRPL) measurements. A generally known deep-level defect located at around 0.8 eV above the valence band was identified for all samples by transient photocapacitance measurement, the position of which was unvaried with the Se beam pressure. The defect concentration and the capture cross section of the minority carrier were obtained from the steady-state photocapacitance and TRPL measurements, which were evaluated to be on the order of 1013 cm−3 and 10−14 cm2, respectively, in the Se beam pressure range from 1.3×10−3 to 4.4×10−3 Pa. The relatively higher defect concentration and larger minority carrier capture cross section were considered as possible causes of the degradation of the conversion efficiency of CIGS solar cells grown under low Se beam pressure.

(Invited) Analysis of Defect Levels at GaAs/GaAs Surface-Activated Bonding Interface for Multi-Junction Solar Cells

Room-temperature surface-activated wafer bonding between bare III-V semiconductor surfaces has become a key technology for high-efficiency multi-junction solar cells, where the reduction of interfacial electrical resistance is of crucial importance for achieving highest efficiency. In the bonding process, surface cleaning using fast atom beam (FAB) of noble gas elements is vital for successful bonding but it damages the surface, resulting in numerous crystal defects at the bonded interface and increases electrical resistance. We here developed quantitative evaluation of such defects introduced by FAB treatment. The surface of n-GaAs was treated with the FAB using Ne, Ar and Kr, and Au Schottky electrodes were formed on the surfaces. Capacitance of a Schottky diode as a function of both probe frequency and DC bias allowed us to characterize both energy depth of the defects and their density profile along the physical depth from the GaAs surface. The results indicated that atoms with the smaller diameter generate high-density defects to the deeper region from the surface. When the defect density exceeding the doping level of GaAs spreads to wider than 5 nm, significant Schottky characteristics appears in the interfacial current-voltage characteristics, as suggested by simulations. Such a tendency was semi-quantitatively in good agreement with the measured current-voltage characteristics of the n-GaAs/n-GaAs bonded interfaces treated with the FAB of Ne, Ar and Kr, suggesting that the capacitance analysis of the FAB-treated surface provides us a direction for optimizing the surface-activated bonding process using FAB.As a simple structure for clarifying the impact of SAB process conditions on the electrical conduction property of the bonded interface, two n-GaAs wafers were bonded by SAB process. Prior to the bonding process, AuGe/Ni ohmic electrode was formed on the backside of n-GaAs wafers. The wafers were then diced into chips and their surfaces were treated with the FAB of Ne, Ar and Kr in a high-vacuum chamber, with an acceleration voltage of 1.4kV and a duration of 3 min. For the admittance spectroscopy to analyze the defects introduced by the FAB treatment, the n-GaAs diced substrates were taken out of the vacuum chamber and Au Schottky electrode was evaporated on the surface. For the conductivity measurement of the boded interface, two GaAs dices after the FAB treatment were put together and pressurized to complete bonding. Subsequently, the bonded dices were further cut into 2×2 mm2 chips and current-voltage characteristics were measured by 4 terminal method.From the dependence of the junction capacitance on both AC frequency and DC bias, with an assumption of a value of carrier-capture cross section by a defect, it is possible to estimate both the energy position and the density profile along the depth direction of the defects existing in the surface vicinity of the substrate, or in the depletion region of the Schottky junction. Remarkably high defect density (around 1019 cm-3) was obtained exceeding the doping level of the substrate and the depth profile indicates that the FAB of the smaller atom introduces defects into the deeper region from the surface. The energy level of the defects were estimated to be 0.8 eV below the conduction-band edge.Taking such distribution of interfacial defects into account, the current-voltage characteristics at the junction was simulated. It was predicted that the defect region thicker than 4 nm results in nonlinear, Schottky-like conduction behavior, which we should avoid in high-efficiency cells. The measured current-voltage characteristics at the n-GaAs/n-GaAs junctions by the SAB using the FAB of Ne, Ar and Kr were compered with the prediction. The tendency of Schottky-like conduction becomes more apparent in the sequence of Ne, Ar and Kr. This is the sequence of smaller atomic radius. In the defect density profile, if we focus on a certain defect density level such as 0.5×1019 cm-3, the penetration depth of the defects is larger also in the sequence of Ne, Ar and Kr.It is strongly suggested that the high-density defects at the interface by wafer bonding distorts band edges and makes potential barrier for electrical conduction. In SAB process, FAB treatment introduces the defects in 1019 cm-3 level and it has significant impact on the interfacial electrical conduction. A suggestion for reducing such introduction of defects is the use of larger atoms in FAB treatment, which reduces penetration depth of defects. It is clear that the condition of FAB such as acceleration voltage of primary ions and exposure time impacts the density and distribution of defects. The optimization of such process parameters is quite possible using the admittance spectroscopy proposed in this work. Figure 1

Piezoelectric effects on carrier capturing time in a hybrid p–n junction system: a numerical study using finite element method

The piezoelectric influence on carrier (electrons and holes) trapping time in a hybrid p---n junction system is investigated in this paper. The hybrid nature of the junction is the conceptual combination of p---n junction solar cell and piezoelectricity. The mechanism is that the piezoelectric field induced on the p---n interface can affect the transport of free carriers such as electrons and holes present inside the junction system. These free carriers included are dissociated from excitons which are generated by the light or by the piezoelectric field on the interface via Langevin recombination. The numerical analysis focuses on carrier density (n, p), carrier velocity $$(v_{n},v_{p})$$(vn,vp), carrier capture cross section $$(\sigma _{n},\sigma _{p})$$(?n,?p), and carrier capturing time $$(\tau _{n},\tau _{p})$$(?n,?p) using the finite element method. These parameters are affected by the piezoelectric potential induced by different vertical stresses $$(T_{z})$$(Tz) on the p---n interface. Based on these features, the simulation of the distribution of the carrier (electrons or holes) capturing time over the junction system can be used to analyze the piezophototronic devices where traps are present.

Improving the internal quantum efficiency of green InGaN quantum dots through coupled InGaN/GaN quantum well and quantum dot structure

The InGaN quantum dot (QD) is promising for use in green light-emitting diodes and laser diodes owing to its small strain and weak quantum-confined Stark effect. However, its small carrier capture cross section still sets a limit to its internal quantum efficiency (IQE). Tunneling-enhanced carrier transfer in a coupled InGaN/GaN quantum well (QW) and quantum dot structure has been studied on the basis of temperature-dependent and time-resolved photoluminescence. It is found that carriers can tunnel from a shallow QW to deep QDs at room temperature. Compared with the conventional single-QD layer, the IQE of the QDs can be enhanced more than two times to about 45%.

Scaling effects in ohmic contacts on semiconductors

This article reports on scaling effects calculated with ohmic contacts on various types of high and low resistivity semiconductors, from near-intrinsic to heavily compensated. The investigation was performed using a finite element computation. It is shown that small ohmic contacts exhibit Schottky-like energy band diagrams on compensated and uncompensated semiconductors. It is shown that the currents of ideal ohmic contacts scale with the contact's perimeter below critical contact size and with area for larger contacts. Bulk velocity saturation effect is shown to introduce non-linearity and asymmetry in the current-voltage characteristics even for ideal ohmic contacts. The effect of finite contact recombination velocity is analyzed. In compensated semiconductors, the deep level carrier capture cross-sections have no effect on resistivity, but may play an important role in the device dc characteristics.

Electronic properties of manganese impurities in germanium

The electronic properties of manganese in crystalline germanium have been investigated by means of deep level transient spectroscopy (DLTS). Mn was diffused in the material by a thermal treatment at 700 °C. Next to the deep levels of nickel and copper, which are known contaminants in Ge treated at high temperature, three not previously reported levels were observed. These two hole and one electron traps, with apparent energy level positions at EV +0.136 eV, EV +0.342 eV and EC − 0.363 eV, were assigned to substitutional Mn. The analysis of the carrier capture cross-sections, the absence of field-assisted emission and the observation of the Mn2− electron paramagnetic resonance spectrum in n-type Ge:Mn at low temperature are all compatible with Mn introducing two acceptor and one donor levels in the band gap of Ge.

Direct detection of carrier traps in Si solar cells after light-induced degradation

In solar cells fabricated from boron-doped Cz-Si wafers minority and majority carrier traps were detected by deep level transient spectroscopy (DLTS) after so-called “light-induced degradation” (LID). The DLTS signals were detected from mesa-diodes with the full structure of the solar cells preserved. Preliminary results indicate metastable traps with energy levels positioned at EV + 0.37 eV and EC – 0.41 eV and apparent carrier capture cross-sections in the 10–17–10–18 cm2 range. The concentration of the traps was in the range of 1012–1013 cm–3. The traps were eliminated by annealing of the mesa-diodes at 200 °C. No traps were detected in Ga-doped solar cells after the LID procedure or below the light protected bus bar locations in B-doped cells. (© 2015 WILEY-VCH Verlag GmbH &Co. KGaA, Weinheim)

Photoelectric activity of defects in La-doped layered TlInS2 crystals

Photoelectric relaxation spectroscopy (photoinduced current relaxation spectroscopy, PICTS) is used to study pure and La-doped single crystals of TlInS2. The characteristics of electrically active defects are determined including the capture cross section for charge carriers, the thermal activation energy, and the temperature interval within which the charge state of a defect varies. The pyroelectric properties of TlInS2:La are studied. The defect responsible for the anomalies in the pyrocurrent is identified.