Search

The electronic properties of working nanocrystalline silicon (nc-Si:H) solar cell devices with conversion efficiencies up to 8.6% were studied using junction capacitance methods. The set of devices examined were deposited on both specular stainless steel substrates and Ag/ZnO textured back reflectors. These devices included nc-Si:H grown under constant H2 dilution, and also with profiled H2 dilution to control the crystallite sizes and volume fraction. Transient photocapacitance and transient photocurrent spectroscopies were used to obtain sub-band-gap optical spectra. A comparison of these two kinds of spectra also allowed us to deduce the minority carrier collection fractions as a function of temperature and light-induced degradation. Light-soaking was found to cause a distinct decrease in minority carrier collection, as well as a consistent decrease in defects responding to drive-level capacitance profiling. A tentative microscopic model is proposed that accounts for these degradation effects in nc-Si:H.

The electronic properties of hydrogenated nanocrystalline silicon (nc-Si:H) were studied using junction capacitance methods. Drive-level capacitance profiling (DLCP) measurements revealed significant differences for nc-Si:H layers deposited under constant hydrogen dilution compared to those deposited using hydrogen profiling, with lower DLCP densities in the latter case. Transient photocapacitance (TPC) measurements revealed the mixed-phase nature of these materials. It disclosed spectra that appeared quite microcrystalline-like at lower temperatures, but more similar to a-Si:H at higher temperatures where the minority carrier collection is higher in the nanocrystalline component of these samples. This then suppresses the TPC signal from this component compared to the a-Si:H component. In contrast, because transient photocurrent signals are enhanced by the additional minority carrier collection, those spectra appear microcrystalline like at all temperatures. We also investigated the effects of light-induced degradation in these devices. This caused a dramatic decrease in hole collection, similar to that caused by reducing the measurement temperature of the samples. However, the light exposure did not appear to increase the deep defect density (dangling bonds).

We report the results of junction capacitance measurements on thin tetrahedral amorphous carbon (ta-C) films to deduce their defect densities. We find defect densities in the range 3 - 7 × 1017 cm−3 in the undoped ta-C films, and roughly an order of magnitude larger in the nitrogen doped (n-type) films. In some cases fairly uniform defect profiles were obtained covering a thickness of a couple of hundreds angstroems. We also observed a thermal activation process of carriers from defect states at the ta-C/c-Si interface with an activation energy in the range of 0.4eV to 0.5eV.

We have applied the Drive-Level Capacitance Profiling (DLCP) method to n-i-p a-Si:H diodes to characterize the mid-gap defect densities in the i layer. Our results show that there are no significant changes in the drive-level densities, Ndl, in n-i-p diodes and p+-i-m as well as n-i-m Schottky diodes, which indicates that DLCP can directly provide reliable energy distribution and spatial distribution of the mid-gap defects in the n-i-p device. We have found that the ratio of Ndl to ND*, the ionized defect density determined by hole onset measurement, is changed with the deposition conditions; it is 3 for standard samples, 2 for helium diluted samples and 6 for hydrogen diluted samples. These results indicate that there may be different defect distributions in these materials, which suggest the ratio of charged (D-) density to the neutral defect (D0) density may be altered when growth conditions are varied.

Significant progress has been made in the area of multi-junction solar cells and stabilized efficiencies of 11.2% and 10.2% have been reported for small-area cells and large-area panels, respectively. The authors have performed voltage-pulse stimulated capacitance transient measurements along with other junction capacitance measurements on a-Si:H/a-Si,Ge:H heterostructures deposited on p-type crystalline silicon using the glow discharge technique. For a filling pulse that puts the c-Si/a-Si:H junction in forward bias, the capacitance transients consist of two components--a fast component corresponding to electron emission and a slow component corresponding to hole emission. For a fixed starting reverse bias, they have found that the density of trapped holes is proportional to the product of the filling pulse voltage and the filling pulse width and reaches a saturation value at a certain value of the product. These hole traps appear to reside at or very close to the a-Si:H/a-Si,Ge:H interface. The estimated trap concentration is around 10{sup 11}/cm{sup 2} and is independent of temperature and applied bias, ruling out the creation of the traps during the filling pulse. The authors also report results on samples in which a-Si:H/a-Si,Ge:H interface has been modified during growth to alter the concentration of the hole traps.

Although a-Si,Ge:H alloys have been increasingly studied in recent years because of their desired tunable band gaps for application in tandem solar cells, the relation between the mid gap defects with solar cell performance is still not completely clear. The authors have characterized the defect state structure in a series of device quality glow discharge produced a-Si,Ge:H alloys with Ge content ranging from 30 at.% to 50 at.% using capacitance profiling, modulated photocurrent, transient junction photocurrent and photocapacitance measurements. As previously reported, these methods allows one to identify two types of thermally induced defect transitions plus two types of optical transitions from deep defects. In the current study the authors have examined the changes in these defects, along with the changes in the hole mobility-time products, that result from prolonged light exposure. By comparing these changes in the annealed state and light soaked state of the same sample, they attempt to correlate the changes in defects with the hole mobility-time product. In general, although all of the defect bands are found to increase after light soaking, the relative factor is found to be different for the various defect transitions within the same sample. The authors also try to identify amore » defect bands may be acting as a safe electron trap, enhancing the lifetime of the minority carriers. They propose that the observed decrease of this defect band relative to the midgap defect band with light soaking could be a significant factor in determining the degradation of these a-Si,Ge:H alloys in the device performance.« less

We have carried out a detailed study of the energy distribution of deep defects for high quality glow discharge a-Si,Ge:H alloys using both thermal emission and optical Methods: drive-level capacitance profiling, transient photocapacitance and photocurrent plus modulation photocurrent spectroscopy. Four distinct bands of transitions involving defect states have been identified: two associated with thermal transitions, and the other two related to optical transitions. We have, for the first time, observed a negative signal in the photocapacitance spectra at photon energies near 1.2eV. This striking aspect verifies the presence of a distinct defect band above Ep from which electron thermal emission is greatly suppressed. Our Measurements also disclose a fairly narrow defect band below the Fermi level which contrasts with the defect properties observed in a-Si:H. Time resolved photocapacitance spectra indicate that this defect band exhibits configuration relaxation.

Bandtail defect carrier trapping dynamics in high quality photo-CVD grown a-Si1-xGex:H (x=0.3 and 0.62) and PECVD deposited a-Si:H films were studied using photo-modulated electron spin resonance (PMESR). Comparison of the dark spin densities with junction capacitance (drive-level profiling) methods indicate that a majority of the defects are neutral in the a-Si:H and 30% Ge samples, but the majority of the defects in the 62% Ge sample are charged. Lineshape deconvolution of the PMESR signals in the in-phase and quadrature phase as a function of photo-modulation frequency, excitation energy and excitation intensity for both intrinsic (state A) and light soaked (state B) states revealed three significant insights: in state A all samples behave similarly, state B behavior of the a-Si:H is dramatically different from the state B alloyed samples, and strong evidence for 'safe hole traps' (SHT) is observed in the state B a-Si:H material.

The effect of C impurities in a-Si:H in levels of 0.4 to 2.6 at. % were studied over a wide range of metastable defect densities. Three complimentary experimental techniques [electron spin resonance (ESR), drive-level capacitance (DLC) and photo-current] were employed to track the material's defect density with light soaking and annealing, as well as Urbach energies, midgap absorption and mobility gaps energies as a function of the C content. Our results show C impurities have a definite effect on the initial and saturated defect densities, as well as the midgap absorption and Urbach energies at levels 1 at. % and above. The results indicate that C acts mainly as a center for increased disorder in the material which results in an increase in the bandtail widths, and consequently an increase in intrinsic defects. Comparison to the ESR and drive-level data show an excellent agreement between these two techniques in determining the bulk defect densities in a-Si:H.

Depletion-width-modulated (DWM) ESR spectroscopy and junction capacitance techniques were employed to investigate deep-defect states in various metastable states of 10 and 80 Vppm PH3-doped a-Si:H. From the joint application of these methods, we enumerate the change in total spin and charge associated with emission of carriers from states within the mobility gap resulting from a modulated bias voltage. We conclude that the magnitude of the modulated spin-to-charge ratio is nearly unity, consistent with earlier similar investigations of n-type a-Si:H, where the D-/D0 transition was shown to dominate. Furthermore, we present results of a previously unreported “quadrature” DWM-ESR signal of large magnitude under a wide range of experimental conditions. Similar related effects are also reported from spin- and capacitance-transient studies. These results clearly demonstrate that the changes in spin significandy lag corresponding changes in charge.

Capacitance vs. temperature, deep-level transient spectroscopy (DLTS), and transient photocapacitance spectroscopy have been used to investigate the amorphous-crystalline silicon interface region of a device made of hydrogenated amorphous silicon deposited on a lightly doped n-type crystalline silicon.By comparing our results between substrates with and without oxide contamination with those in a earlier study, we have been able to correlate the effects of substrate preparation on the density of interface states.

We have carried out a detailed study of the competition between optical and thermal transitions from mobility gap states in hydrogenated amorphous silicon films. By employing transient photocapacitance and capacitance DLTS techniques on n-type doped films, we have observed the dangling bond optical transitions:D− → D0 + e and D0 → D+ + e, and the thermal transitions: D− → D0 + e and D0 → D0 + h. From the transition energies we have determined that the degree of lattice relaxation for dangling bonds does not exceed 0.1 eV and that the value Ueff of the D−/D0 splitting is 0.3 ± 0.1 eV. Voltage pulse photocapacitance studies on undoped films revealed the existence of the optical transition D−/D0 + e. In addition, in all films we have observed a photocapacitance signal due to optical transitions from the valence band tail states (over the energy range 1.4 to 1.9 eV) at temperatures as high as 440 K, in spite of the expectation that such shallow photoexcited holes would immediately be thermally emitted yielding no photocapacitance signal. We discuss possible explanations for this result by considering two different trapping models.

A novel a-Si:H/c-Si Schottky diode heterostructure device premits the study of the capacitive response to relatively low temperatures (25 K) and also allows fast pulse filling capture measurements of electrons from the c-Si substrate into a-Si:H defect states. These latter measurements, as well as capacitance vs. temperature measurements on these diodes, indicate a nearly zero conduction band offset (50±50 meV). We also have observed trapping of holes at the a-Si:H/c-Si valence band discontinuity ΔEv. A clear threshold for the subsequent optical release of these trapped holes by sub-bandgap light yields a value of ΔEv = 0.58±0.02 eV. Finally, photocapacitance spectra along with thermally stimulated capacitance (TSCAP) measurements indicate an anomalously large (l×1018/cm3) Gaussian-shaped defect band located a Ec - 0.88 eV with a FWHM of 0.46 eV. Model calculations of the high temperature capacitance-voltage dependence indicate these defects lie predominately within 350 Å of the a-Si:H/c-Si interface.

The effects of light saturation on the properties of undoped a-Si:H films were studied by a new capacitance profiling technique which can be used to directly determine changes in the dangling bond density of states near midgap. Coplanar conductivity and capacitance vs. temperature measurements save the changes in activation energies for electrical conductivity. These studies indicate that, while substantial increases in the dangling bond densities are observed for most samples, the detailed behavior of the light induced changes in these films are inconsistent with the creation of such defects by breaking weak valence band tail states.

Amorphous-crystalline heterostructures composed of sub-micron thick undoped amorphous hydrogenated silicon deposited on lightly doped n-type crystalline silicon substrates have been studied by capacitance techniques. Capacitance vs. temperature scans along with model calculations on these samples are used to investigate the amorphous-crystalline silicon interface region. We deduce the existence of an anomalously defective a-Si:H region extending roughly 1000 Angstroms from the growth interface. We also deduce considerable fluctuations in the interface potential which indicate lateral variations of ±20% in the defect distritribution over a 1000 Angstrom scale.

We have studied the light induced instability problem in hydrogenated amorphous silicon using junction capacitance techniques. These techniques are used to examine specific changes in the density of gap states, and occupation of gap states, for undoped a-Si:H samples after light saturation and for a series of partial anneal “states” which culminate in the original dark annealed state (state A). We find that the observed changes in the metastable occupied and unoccupied defects contradict the Si-Si bond breaking model and indicate at least two defect creation processes. In several samples we also find clear evidence that the metastable defect distribution near midgap has a slightly different energy distribution than the stable deep state (dangling bond) distribution. At the same time, these results seem to be qualitatively consistent with many aspects of recent ESR and optical absorption studies of metastable defect creation. We discuss these findings in terms of alternative possible microscopic models for metastable effects in a-Si:H.

The density of gap states in the bulk of intrinsic a-SiH has been probed using depletion width modulation electron spin resonance spectroscopy (DWM ESR) as well as complementary capacitance techniques. Our sample consists of a 2.85 micron thick layer of undoped a-Si:H sandwiched between a large area Pd Schottky barrier contact and a p+ c-Si substrate, allowing charge-counting capacitance and spin-counting ESR measurements to be performed on the same sample. DWM ESR measures the spin change that occurs when electrons are removed from gap states by increasing reverse bias on a Schottky barrier. The phase of the signal indicates whether unpaired spins are created or destroyed with increasing bias, thus revealing the defect charge state at the Fermi level The DWM ESR signal from our sample indicates that removing electrons from gap states leads to a reduction of the spin signal. This is qualitatively consistent with a band of Do states near EFo. However, for all metastable states and at all temperatures studied, the magnitude of the DWM ESR signal account for only about 14% of the responding charge as determined by junction capacitance. This would occur only if the majority of the filled bulk gap states were doubly occupied with a small correlation energy Ueff. In that case the majority of transitions caused by depletion width modulation would be D−/D+ which would produce no spin change. Model calculations show that the small observed Do/D+ DWM ESR signal would indeed result from the occupation statistics for a near zero value of Ueff at the moderate temperatures of the experiment. Comparison with our data sets the overall limits If −20meV < Ueff < 50 meV. Implication for recent models of the density of states in intrinsic a-Si:H are discussed.