Search

Three-diode lumped-parameter equivalent circuit model conforms to photovoltaic device physics and processes an exact analysis for ${I}$ – ${V}$ characteristics of industrial solar cells, because it explains the different leakage current components especially for recombination current resulting from defects or grain boundaries. However, three-diode model’s complete circuit topology leads to complicated transcendental ${I}$ – ${V}$ equation without analytical solution so that low computation efficiency limits three-diode model’s applications in photoelectric simulations. In this article, an effective-diode method is proposed to simplify three-diode model, derive accurate and efficient terminal current–voltage solution to three-diode model, and acquire electrostatic characteristics of the solar cells. The calculated values have good agreements with numerical iteration results and experimental data measured from solar cells, respectively. Finally, the effective-diode method performs an important role of solving three-diode model analytically, predicting ${I}$ – ${V}$ characteristics of industrial solar cells accurately, and providing three-diode model’s practicability and development in solar cells’ simulations.

Thermal annealing and effective surface passivation are significant to achieve high conversion efficiency (η) of passivated emitter and rear cells (PERC). Thus, we added an annealing process before front and rear surface passivation in the existing production line. We found that the front recombination current (J0,f) and the rear recombination current (J0,r) of annealed PERCs are both lower than baseline PERCs without annealing process, which means better defect repair and passivation effect were achieved. And the average η of annealed PERCs reaches 21.79%, which is 0.23% absolutely higher than that of the baseline PERCs. After that, we further explored the influence of gas composition (0%, 5% and 10% oxygen concentration in nitrogen atmosphere) during annealing process. We found that as the oxygen concentration increases, J0,fs and J0,rs of annealed PERCs (0%, 5%, 10%) are all lower than those of baseline PERCs. Meanwhile, the annealed PERCs (0%, 5%, 10%) get the average η of 21.77%, 21.82% and 21.84% respectively, possessing absolute efficiency increase over the baseline PERCs with the average η of 21.67%. The present work could help enhance the understanding of passivation property and improve the performance of industrial PERC cells.

In this work we have characterized screen-printed passivating contacts formed by different commercially available fire-through pastes on phosphorus doped (n+) polysilicon (poly-Si) layers at the rear side of monoPoly™ solar cells. Extremely low recombination current density under metal contacts (J01,metal) of 35–45 fA/cm2 and excellent specific contact resistivity (ρc) values of ~ 1.3 mΩ-cm2 are obtained for two different thicknesses of poly-Si (150 nm and 250 nm) used in this work. We demonstrate that, although the metal induced recombination increases with reducing thickness of the poly-Si layer, thinner poly-Si layers can lead to higher efficiencies on account of reduced parasitic absorption leading to higher cell current. A champion efficiency of 22.6% is reported for busbarless monoPoly™ cells with the best performing fire-through (FT) paste on large area (244.3 cm2) commercially available Czochralski grown Si wafers.

Analytical methods for investigating deep levels in GaN are reviewed. Quantitative accuracy of deep-level transient spectroscopy (DLTS) technology is discussed. The isothermal method allows a quick measurement and is useful to characterize trap states for the case that a temperature scan causes reaction of defects or the occupancy condition for each filling pulse needs to be kept constant. The DLTS measurement for hole traps in p-type GaN needs special considerations due to the relatively large ionization energy of magnesium acceptors, where the low-frequency DLTS system is utilized. For investigation of minority carrier traps, DLTS using injection bias pulses for p-n junctions is the most reliable method for the quantitative evaluation. DLTS methods using light filling pulses such as minority carrier transient spectroscopy and optical DLTS are applied for samples without p-n junctions, which needs considerations of light absorption, the minority carrier diffusion length, and the optical capture cross sections of deep levels. Investigation of deep levels far from both band edges requires the approaches without the use of thermal emission process. For this aim, this chapter overviews deep-level optical spectroscopy, photoluminescence involving a time-resolved method, positron annihilation spectroscopy, and an analysis of recombination current in a p-n junction.

We developed the algorithm for quantifying performance from PV module's I-V characteristics measured in exposure for estimating the lifetime of PV modules. The developed algorithm can accurately detect the PV module's shunt resistance from the I-V characteristic measured in exposure. Further, this algorithm can convert the detected shunt resistance to the current generated by recombination of minority carriers in the PV module. The rate of the recombination current change was compared with the rate of change in luminance of the EL image measured indoors, and a high correlation of 96.7% was obtained.

A study of how electron irradiation affects the current-voltage (I-V) and electroluminescence (EL) characteristics of two types of 4H-SiC n+p structures with p-base and base doping to ~5∙1015 cm-3 is presented. The characteristics were measured prior to irradiation and after each of five stages of irradiation with 0.9 MeV electrons at doses in the range from 1∙1015 to 1.1∙1016 cm-2. The irradiation leads to an increase in the recombination current, decrease in the intensity of the edge EL (hνmax≈3.18 eV), and increase in the intensity of the infra-red (IR) EL (hνmax≈1.35 eV), which starts to predominate. Presumably, this indicates that the nonequilibrium carrier lifetime decreases and the concentration of acceptor type defects grows as a result of the irradiation. The IR EL, attributed to a complex defect containing a silicon vacancy, is of interest for development of single-photon sources of light.

Low-cost aluminum (Al) metallization process on p+ side of n-type silicon (Si) substrate with low recombination current density (J0met) is presented. Narrow-line Al screen-printed fingers on symmetrically passivated p+ -n-p+ Si precursors with local contact opening (LCO) are investigated. The Si content of the Al pastes, LCO patterns and Al finger widths are varied to evaluate the formation quality of the Al-doped p+ region. By increasing the pitch distance of the dot-shaped LCOs, the inverted pyramid cavities are found after removing the Al-Si eutectic part by HCl. The tendency of the reaction during the sintering process is controlled by fritting the Al paste with Si, which allows a faster Si saturation in the paste matrix and facilitates faster return of Si to the LCO region. With a combination of narrow-line fingers of 90 μm and a dot-shaped LCO pitch of 140 μm, J0met value of 420 fA/cm2 is realized.

The efficiency record of industrial type PERC solar cells exceeded 22% at the turn of the year 2015 to 2016. Our best screen-printed PERC solar cell reached 22.04% efficiency while the best cell batch showed a very narrow efficiency distribution. A detailed electrical and optical loss analysis of those industrial type high efficiency PERC solar cells is carried out which enables further optimization and strategic improvements. A variety of characterization data allows for a recombination current density, resistance and optical loss analysis based on numerical device simulation, analytical calculations and raytracing, respectively. The main recombination losses at maximum power point (MPP) occur in the homogenous and selective diffused regions of the emitter. A series resistance loss analysis is analytically performed. The emitter contribution to the lumped series resistance dominates the series resistance losses. The optical loss analysis performed with raytracing shows main reflection and absorption losses in the rear metal layer which is partly due to the light trapping capability of the PERC cells.

Neutron irradiation (~1 MeV, dose 1014-5.6∙1015 neutron/cm2) of packaged diodes based on 6H-SiC pn structures (with the base n-layer doped to ~5∙1016 cm-3) has been studied. In addition to the well-known rise in the series resistance of the diodes, the effect of a partial suppression of the excess current in both forward-and reverse-biased diodes and that of an increase in the recombination current, probably associated with the decrease in the nonequilibrium carrier lifetime, were discovered and discussed. These effects are common to 6H-and 4H-SiC pn structures.

This paper presents a new physics-based compact model implementation for interface state creation due to hot-carrier degradation in advanced SiGe HBTs. This model accounts for dynamic stress bias conditions through a combination of the solution of reaction-diffusion theory and Fick’s law of diffusion. The model reflects transistor degradation in terms of base recombination current parameters of HiCuM compact model and its accuracy has been validated against results from long-term DC and dynamic aging tests performed close to the safe-operating-areas of various HBT technologies.

In this paper, the degradation characteristics of the series diodes under DC high voltage operation are investigated. Five series connected diodes are operated at a reverse bias voltage of 1500V. The influence of the DC high voltage on the diode parameters is analyzed after the I-V characteristic test. Finally, an annealing experiment is performed on a diode that stressed under DC high voltage. The results show that the high voltage will introduce defects into the diodes, which leads to the increases of the recombination current and the ideality factor. The degradation degree of five diodes is basically the same. For the annealing experiment, the recoverability of defects can be observed.

The damage mechanism of proton irradiation in InP/InGaAs heterostructures was studied. The deep level traps were investigated in detail by deep level transient spectroscopy (DLTS), capacitance&ndash, voltage (C&ndash, V) measurements and SRIM (the stopping and range of ions in matter, Monte Carlo code) simulation for non-irradiated and 3 MeV proton-irradiated samples at a fluence of 5 &times, 1012 p/cm2. Compared with non-irradiated samples, a new electron trap at EC-0.37 eV was measured by DLTS in post-irradiated samples and was found to be closer to the center of the forbidden band. The trap concentration in bulk, the interface trap charge density and the electron capture cross-section were 4 &times, 1015 cm&minus, 3, 1.8 &times, 1012 cm&minus, 2, and 9.61 &times, 10&minus, 15 cm2, respectively. The deep level trap, acting as a recombination center, resulted in a large recombination current at a lower forward bias and made the forward current increase in InP/InGaAs heterostructures for post-irradiated samples. When the deep level trap parameters were added into the technology computer-aided design (TCAD) simulation tool, the simulation results matched the current&ndash, voltage measurements data well, which verifies the validity of the damage mechanism of proton irradiation.

This work explores the possibility of using ultrathin silicon nitride (SiN x ) films with high positive fixed charge in a SiN x /poly-Si passivating contact. The factors including (i) film thickness, (ii) annealing condition (time, temp and ambient) and (iii) surface pre-treatment were optimized to boost the passivation performance of ultrathin LPCVD SiN x films. Our preliminary experiments reveal excellent surface passivation and low recombination current density, J o (45 fA.cm-2) by ~1.5 nm thick LPCVD SiN x films when subjected to an air ambient anneal at 800°C for 30 mins. This is due to the formation of high positive fixed charge density (1.5 ×10 12 cm-2). Air ambient annealed (465 μs) samples also have a higher lifetime when compared to the forming gas annealed (208 μs) samples. These passivating SiN x films were further integrated into SiN x /n+poly-Si contacts and characterized for J o,contact and tunneling resistance, ρ contact . The best SiN x /n+poly-Si passivated contact in this study has J o, contact = 5.9 fA.cm-2, ρ contact = 0.525 Ω.cm2 and an efficiency potential > 22.75%. According to our knowledge, it is the first report confirming the formation of passivated contacts with SiN x as the dielectric tunnel layer.

The conditions under which Schottky contacts become ohmic are analyzed. Proceeding from classical concepts of the mechanisms of current flow, a generalized Schottky contact model is investigated. This model takes into account the thermionic current of majority charge carriers and the recombination current of minority carriers in Schottky contacts with a dielectric gap. Based on an analysis of the predictions of this model, ohmicity criteria are obtained for Schottky contacts and the conditions for a low injection level and the ohmicity of Si-based Schottky contacts are compared. It is shown that the conditions for Schottky-contact ohmicity do not coincide with those for p–n junctions.

It is well known that the quality of contact passivation can greatly influence the efficiency of Si solar cells. In this work, the impact of annealing a wet-chemically formed silicon oxide (SiOx) tunnel layer (prior to poly-Si capping) to form SiOx/n+ poly-Si passivated contacts for high efficiency solar cells is investigated. The effect of different annealing temperatures (ranging from 425 to 900 °C) and ambients (air and forming gas), on the overall performance of the resulting SiOx/n+ poly-Si passivated contacts are studied. The efficiency potential of these SiOx/n+ poly-Si contacts is calculated from the measured values of the recombination current density (J0,contact), and the effective contact resistivity (ρcontact), according to the generalized Brendel's model. The results are compared to the reference case, i.e. non-annealed SiOx tunnel layer with poly-Si deposition. The lifetime, the fixed charge and also the SiOx tunnel layer thickness increase upon annealing at higher temperatures. The sample annealed in forming gas at 900 °C shows an increase in field effect passivation (Qtot = 1 × 1012 cm−2) as well as in chemical passivation (Dit = 3 × 1011 eV−1 cm−2), along with reduced ρcontact, indicating pinhole formation upon annealing at 900 °C. The best SiOx/n+ poly-Si passivated contact achieved J0,contact = 4.2 fA cm−2, ρcontact = 0.27 Ω cm−2 and a corresponding maximum ideal efficiency potential of 27.9%. This is 1.2% absolute higher as compared to the reference case, when no annealing of the tunnel layer is performed.

Dark I–V curves of silicon solar cells with various Shockley–Reed–Hall lifetimes have been studied. The lifetimes are determined from the short-circuit-current internal quantum yield. The recombination currents in the space charge region (SCR) are found to be formed within time intervals that are at least an order of magnitude shorter than the charge-carrier bulk lifetime. This effect can be associated with a high defect concentration (and, therefore, a high deep-level concentration) in the SCR of examined Si structures. The parameters of deep centers that are responsible for the recombination in the SCR have been evaluated., Дослiджено темновi ВАХ кремнiєвих сонячних елементiв з рiзними часами життя Шоклi–Рiда–Холла, якi визначались з спектральних залежностей внутрiшнього квантового виходу струму короткого замикання. Встановлено, що рекомбiнацiйнi струми в областi просторового заряду (ОПЗ) формуються на основi часiв життя, менших, принаймнi на порядок, за об’ємнi часи життя. Це пояснено великою концентрацiєю дефектiв, якi приводять до появи глибоких рiвнiв, в ОПЗ дослiджуваних структур кремнiю. Оцiнено параметри глибоких рiвнiв, вiдповiдальних за рекомбiнацiю в ОПЗ.

Nongeminate recombination in organic solar cells is one of the main causes of reduced device performances. The quantification, and the subsequent reduction, of nongeminate recombination losses requires the development of methods, which allow an accurate and reliable determination of the charge-carrier density, mobility, and lifetime. Here, we present a method, based on the charge-density dependence of the recombination current, to determine the recombination parameters from impedance measurements of organic solar cells. We show that the use of the difference between the capacitance under illumination and the capacitance in the dark for the calculation of the charge density present in the active layer, allows the identification of the dominant recombination mechanism and the subsequent determination of either the charge-carrier lifetime or the bimolecular recombination coefficient.

The laser isolation of p-n junctions is the origin of well known electrical damage in solar cells. Although several models based in complex recombination phenomena have been used to explain these effects, this work uses a simple model of a single 2kT diode connected to the cell by a large variable value resistance. A previous work has shown the coherence of the model in terms of I-V behavior at different temperatures but without any evidence of their physical nature. This work uses electroluminescence techniques to validate this model finding the recombination current paths from the cell volume to the high recombination defects.

In this paper, the impact of deep level traps on the I-V and C-V characteristics of InP/InGaAs heterojunction have been studied by introducing deep level traps in pre and post-irradiation samples. In this case, forward recombination current and junction capacitance in dependence of the deep level density and energy have been analyzed by TCAD simulation and experiments. The simulation results are in good agreement with the experimental results. The junction capacitance and recombination current both increase with the increasing of deep level density. Apart from this, the radiation induces deeper level traps, which contributes more to these increased trends.

We discuss the effect of out-of-plane excess carrier density variations on contact recombination currents extracted from quasi steady state photoconductance (QSSPC) measurements on a dedicated test structure. The test structure features lattices of point contacts. The contacts are thick and non-transparent. Therefore, asymmetric test structures are generally used to ensure a homogeneous in-plane generation rate. As a result, the QSSPC measurements are relatively prone to out-of-plane excess carrier density variations. Their effect on the extracted contact recombination characteristic is discussed theoretically and the theory is confirmed by experiment. The resulting framework contributes to the set of design rules and best practices for contact recombination current characterization using QSSPC measurements on point contact lattice based test structures.

We recently proposed a novel test structure for the extraction of recombination characteristics of metal-contacted areas on silicon wafers using quasi-steady state photoconductance measurements. In this test structure, photoconductance measurements are performed on several wafer areas with different contact fractions. Each area consists of a lattice of point contacts on an otherwise passivated wafer. The requirement of constant excess carrier densities throughout the quasi-neutral wafer bulk is absolutely essential for the simple extraction of figures of merit for contact recombination. In this study, we first discuss the requirements for constant injection levels during photoconductance measurements, without reference to a particular geometry. Then, we use a simple 1-D model to elucidate how injection levels vary in the test structure plane. We use the model to investigate limiting cases for which the requirement of constant injection levels is satisfied. In addition, we investigate the breakdown of the assumption of constant injection levels. We discuss the influence of nonconstant excess carrier concentrations on photoconductance measurements in the context of our test structure. Finally, we validate our model by comparison with experiments. Our analysis is particularly useful in the context of understanding the design rules for contact size and pitch in our test structure.

In this work, we evaluate the impact of implementing a selective back surface field (BSF) on the performance of bifacial n-type Passivated Emitter and Rear Totally diffused (PERT) cells with Ni/Ag co-plated contacts. For this, we compare bifacial nPERT cells featuring a homogeneous BSF etched-back to ~100 Ω/sq and patterned by laser ablation to cells featuring an identical front side design and a selective BSF obtained by means of laser doping. In the first part of this paper, we demonstrate using test structures that selective BSF cells benefit from drastically lower area weighted recombination current density (J0) contributions at the BSF side resulting in estimated Voc improvement from ~657 mV to ~ 681 mV. In the second part of this paper, we confirm at cell level that implementing a selective BSF results in gains in Voc ~ 30 mV, jsc~ 0.2 mA/cm2, and Fill Factors FF ~ 0.9%abs. resulting in average cell efficiencies improving from 20.7±0.4 % to 22.0±0.2 %. This clearly confirms that a selective BSF is required in nPERT cells with Ni/Ag co-plated contacts to maximize cell efficiencies. Finally, we fabricated a 9-cells bifacial glass-glass laminate for proof-of-concept and discuss several possible improvements to bring 60-cells module front maximum power values under standard test conditions (STC) above 340 Wp.