The conventional a-Si:H thin film transistor (TFT) is a field effect transistor (FET), which has disadvantages of a low operation speed and a small current driving capability. To achieve a higher speed and larger current driving capability, a potential solution is to fabricate the a-Si:H-based bipolar thin film transistor (B-TFT). In this study, a-Si:H p-i-n junctions were prepared and studied in order to determine the proper layer thickness for good junction behaviors. B-TFTs composed of a stacked structure of n+/i/p/i/n+ were then fabricated. The complete B-TFT was made using plasma enhanced chemical vapor deposition (PECVD) to deposit all doped and undoped a-Si:H layers and SiNx dielectrics at 250°C. Reactive ion etching (RIE) and wet etching methods were used to define base and emitter regions and contacts. The I-V characteristics of the complete B-TFT were investigated. The common-emitter current gain is 3∼6, which is larger than the literature report of 2∼3. In addition, a collector current larger than the literature value was obtained. A significant current noise was observed, which may be contributed to the high series resistance of the base layer and defective junction interfaces. In this paper, process and structure influences on the a-Si:H junction and B-TFT performances are discussed.

We apply an experimental variation of the steady-state photocarrier grating technique by monitoring the photoresponse at higher electric fields and thus changing from diffusion to drift determined transport. Evaluation of the field-dependen experimental data is achieved with the analysis by Abel et al. [C.-D. Anel, G.H. Bauer and W. Bloss, Philos. Mag. B 72, 551 (1995)]. We study the photoelectronic properties of microcrystalline silicon samples and deduce the minority carrier diffusion length. From the variation in electric field the trapped charge density, responsible for charge neutrality under illumination, can also be determined. Thus additional information is gained which can be related to the density of states in the material.

We report on the properties of nanocrystalline Si:H solar cells. The solar cells were of the p+nn+ type, with the n+ layer deposited first on a stainless steel substrates. The solar cells were prepared under high hydrogen dilution conditions using either ECR plasma deposition, or VHF diode plasma deposition processes. The deposition pressures were kept low, 5 mTorr in the ECR reactor and 50 mTorr in the VHF reactor. All the solar cells reported showed a high Raman ratio of crystalline to amorphous peaks. Properties such as dark current, deep level defects and shallow doping densities, and hole diffusion lengths were measured in these cells. It was found that the base layer was always n type, but that its doping could be changed by adding ppm levels of B during growth. A sufficient B doping even type converted the base layer to p type. It was found that there was a good one-to-one correlation between the shallow doping and deep level defects, suggesting that the same element, probably oxygen, is responsible for generating both shallow dopants and deep levels. The diffusion length of holes was measured in these cells using quantum efficiency vs. voltage techniques, and it was found that the diffusion length data could be explained very well by invoking trap-controlled recombination statistics. The dark I(V) curves could be represented by a standard diode model for highly crystalline materials, but as the degree of crystallinity was reduced, the diode factor increased. Voltage could be improved by reducing the crystallinity of the layer, but doing so resulted in a decrease in quantum efficiency in the infrared regions of the solar spectrum.

We propose a short channel gate overlapped lightly doped drain (GOLDD) poly-Si TFT employing 45° tilt implant for source and drain (S/D) regions without any additional ion doping or mask. Oblique-incident ELA activation is performed to activate both n+ S/D and n- LDD regions as well as cure junction defects. The proposed poly-Si TFT can suppress the anomalous leakage current, and exhibit the better reliability against the hot-carrier stress.

Dilution by Ar of silane plasma has been reported to increase the stability of a-Si:H films. A critical question is whether Ar diluted i-layers offer higher stabilized solar cell efficiencies than the conventional hydrogen dilution method. We have fabricated a-Si:H p-i-n solar cells with RF-PECVD i-layers by Ar dilution of silane. Ar dilution ratio (ADR, Ar/SiH4), RF power,pressure, and i-layer thickness were varied. At low ADR < 20, such solar cells show comparable initial efficiencies and stability as those devices having H2-diluted i-layers of similar thickness. For cells made with ADR > 20, the initial efficiency decreases dramatically with further increase in Ar dilution, and light soaking causes only mild changes in efficiencies. The stabilized efficiencies of cells made with high ADR are inferior to the cells produced with low ADR or cells prepared by H2 dilution. Further, Voc of solar cells made with high ADR (> 50) decreases substantially in ambient, indicating a porous microstructure susceptible to oxidation. While thermal annealing improves the Voc, a full recovery of Voc is made by accelerated light soaking.

The combination of high power and high ADR can lead to nanocrystalline silicon (nc-Si:H) growth, although nucleation is much more difficult to attain by the Ar dilution method compared to hydrogen dilution. We have succeeded in fabricating p-i-n solar cells with nc-Si:H i-layers prepared by the Ar dilution approach. The double dilution by Ar and hydrogen of silane (Ar+H2+SiH4) can result in nc-Si:H i-layers with enhanced long wavelength spectral response compared to devices incorporating nc-Si:H i-layers grown by H2 dilution only. The nc-Si:H solar cells with Ar+H2 diluted i-layers exhibit no light-induced degradation.

Using energetic Ar-rich plasma, in a process much simpler than the traditional nc-Si:H technique, doped a-Si:H thin layers can be prepared to form excellent tunnel junctions for multi-junction solar cells. We demonstrate such a novel, non-contaminating tunnel junction in tandem a-Si/a-Si and a-Si/nc-Si solar cells entirely fabricated in a single-chamber RF-PECVD system.

A systematic study has been made of the influence of the deposition conditions on the properties of SiO2 grown by liquid phase deposition (LPD), and a-Si:H manufactured by plasma enhanced chemical vapour deposition (PECVD) with the novel facility of source-gas heating. It is demonstrated that LPD-SiO2 can be grown at 50°C with good dielectric properties. Material has been produced with a resistivity of 1015 &cm and a dielectric strength of 9 MVcm-1.The oxide was found to have a negative fixed oxide charge of 4 × 1011 cm−2, with a dielectric constant of 3.08 and a refractive index of 1.44. In the case of a-Si:H, pre-heating the source gases has enabled material to be grown at 125°C with a hydrogen content of ∼ 10 at%, with a predominance of monohydride bonding and a photosensitivity of ∼ 104. Inverted-staggered thin film transistors have been fabricated incorporating these films with an On/Off ratio of five orders of magnitude, a sub-threshold slope of 1.3 Vdecade−1 and a field effect mobility of 0.20 cm2V−1s−1

The deposition of silicon films was investigated for the first time by the pulsed cathodic vacuum arc process. This method has been employed to take the advantages of its low deposition temperature, high deposition rate, and high-energy capabilities, coupled with its relatively low operational cost. Heavily doped silicon films were deposited on silicon and glass substrates at temperatures below 100°C with pulsed deposition rate of 0.2nm/A·s. Pulsed arc currents up to 400A in 1∼5ms pulse width with 20∼300 pulses per second were studied. Compared with the direct current (D.C.) cathodic vacuum arc, numerous possibilities exist for the pulsed arc deposition to suit specific targeted film growth. The characterization of the films was carried out by scanning electron microscopy (SEM) and X-ray diffraction (XRD) in terms of materials morphological and structural properties. The production of high quality silicon film materials at low temperature would further enable the integration of microsystems with microelectronics.

Growth dynamics and microstructure of thin-film silicon simulated by a 3D dynamical numerical model are investigated. The model, recently introduced, is characterized here with its phase diagram. It reproduces the main features of the growth and microstructure of thin film silicon: amorphous to crystalline phase transition, conical/columnar shape of the conglomerates of nanocrystals, surface roughness evolution of the layer. It is observed that preferential etching of the amorphous silicon is sufficient to reproduce qualitatively the surface evolution observed experimentally. In the presence of preferential etching, nucleation of the microcrystalline phase in the simulated layers always coincides with a surface roughness increase as observed experimentally. This model opens new perspectives for the simulation of thin-film microstructure and surface morphology.

We used Raman and photoluminescence (PL) spectroscopy to study the relationship between the material properties and the solar cell performance of hydrogenated nanocrystalline silicon (nc-Si:H). The crystalline volume fraction (fc) was deduced from the Raman spectrum. Generally, a high fc leads to a high short circuit current density and a low open circuit voltage. PL spectra were measured using 632.8-nm and 442-nm laser lines. There are two distinguished PL peaks at 80 K, one at ∼1.4 eV originating from the amorphous region, while the other at = 0.9 eV from the nanocrystalline grain boundary regions. Generally, the intensity fraction of this low energy PL peak, IPLc/(IPLa+IPLc), was larger for 442-nm than 632.8-nm excitation, indicating an increase in crystallinity along the growth direction. However, for the best initial performance cells obtained by H2 dilution profiling and the i/p buffer layer, the intensity fraction IPLc/(IPLa+IPLc) decreased from the bulk to the topi/p interface. The Raman and PL results give insight into the correlation between the microstructures and the cell performance, and verified that properly-controlled crystallinity in the intrinsic layer and buffer layer at the i/p interface layer are important for optimizing nc-Si:H solar cells.

Electrical switching due to metallic filament formation in hydrogenated amorphous silicon (a-Si:H) is studied in metal/a-Si:H/metal structures. We examine the effects of a-Si:H switch layer thickness, applied voltage and polarity, metal contact material, and contact interface properties. For switching, the voltage applied to the contacts must be large enough to establish: 1) a minimum threshold voltage of about 2V at the contacts and 2) a bias field of about 1 MV/cm in the bulk. Changing contact material and polarity strongly affects the switching behavior.

Large area hydrogenated amorphous silicon single and stacked p-i-n structures with low conductivity doped layers are proposed as monochrome and color image sensors. All have the same intrinsic layer and in the doped layers the resistivity and optical gap was controlled through the addition of methane to the doping gas. The current-voltage characteristics and the spectral sensitivity under different illumination conditions are analyzed. The sensor output characteristics are evaluated under different read-out voltages and scanner wavelengths. To extract information on image range, intensity and color, a modulated light beam scans the sensor active area at three appropriate bias voltages and the photoresponse in each scanning position (“sub-pixel”) is recorded. The investigation of the sensor output under different scanner wavelengths and varying electrical bias reveals that the response can be tuned, thus enabling color separation. The operation of the sensor is exemplified and supported by a numerical simulation.

Glow discharge amorphous hydrogenated silicon (a-Si:H) prepared at near room temperature typically results in an inhomogeneous morphology that is undesirable for a number of thin film applications. The most commonly observed features of this include columnar morphology and surface roughness. This usually results from anodic deposition, where substrates are placed on the grounded electrode. We have discovered that placing substrates on the RF-powered electrode (referred to as cathodic deposition) offers a much wider processing range for homogenous growth than anodic growth. We have also found that the magnitude of the surface roughness and the bulk void fraction of both anodic and cathodic a-Si:H thin films processed at low-temperatures is proportional to ∼D/F, where D is the surface diffusivity and F, the adatom flux, though anodic and cathodic deposition affect these global parameters differently. Surface processes unique to cathodic deposition can enhance adatom surface diffusion, while diffusion during anodic deposition is fixed and cannot attain homogeneous growth at high adatom fluxes. Processing a-Si:H on the cathode, associated with enhanced adatom surface diffusion, allows for homogeneous growth even at high deposition rates that has benefits for a number of applications.

We have applied a high resolution scanning Kelvin probe to perform dark surface potential topographies of multicrystalline silicon solar cells having thin coatings of Si3N4 and SiO2. We clearly observe the electrical characteristics of the screen printed bus-bar and associated fingers, grain boundaries, together with characteristic structures on the oxide and nitride, coupled to significant surface potential variations across larger sections of the wafer. Associated surface photovoltage measurements can be unambiguously decoded to show coating and bulk contributions. The nitride coating exhibits carrier trapping lifetimes in excess of 13 minutes at 300K.

Amorphous silicon (a-Si) Thin Film Transistors (TFT) will continue to play an important role in large-area liquid crystal displays (LCD), however, there is also a strong demand for ultra-high performance TFTs aimed at intelligent small (or medium) size displays. For this reason, “Crystal Si on glass” has become an increasingly attractive solution. This paper reviews recent work performed at ALTEDEC (Advanced LCD Technologies Development Center Co., Ltd, with regard to growing an array of large Si grains at low temperature. The results obtained so far indicate that satisfactory progress has been made towards the achievement of this goal.

We have presented a model of microcrystalline silicon (μc-Si) growth based on the Van der Drift model. The model needs growth velocities of the facets (100) and (111), an amorphous silicon growth velocity and a grain nucleation rate. The growth velocity ratio of the facets, (100) and (111), determines the preferred orientation and the morphology of the μc-Si film, especially oriented to (110). As the grain nucleation rate increases, the ratio of the living grain number to the total grain number decreases and the crystallinity increases, so the grain nucleation rate governs the trade-off relation of the μc-Si cells between decreasing the open circuit voltage and increasing the short circuit current as the crystallinity increases.

This contribution reviews the compatibility of Staebler-Wronski models with experimental data and observations. The review will show that neither the “bond-breaking models” (originally proposed by Dersch and Stutzmann) nor the “defect conversion” models (originally proposed by Adler) can explain all observations on films and/or solar cells. It has been well accepted for some time that experimental stress and recovery phenomena, both on films and devices, always identify both “slow” and “fast” degradation and recovery mechanisms. It is argued that the quintessential understanding of the Staebler-Wronski mechanisms will come from identifying a fundamental physical process that provides a quantitative understanding of the “coupling” between the slow and fast mechanisms.

Doping of polysilicon (poly-Si) films was performed at a low temperature (<150°C), by using three different dopant incorporation methods: ion shower, dopant layer deposition and plasma immersion. All three techniques were shown to be capable of obtaining sheet resistance values that were smaller than 104 Ω/sq., which were considered to be sufficient to form good source-drain contacts. Also, a sheet resistance value that is as low as 300 Ω/sq. was demonstrated. It was found that the laser energy used for dopant activation was the major parameter to control the sheet resistance of the poly-Si films. The lowest attainable sheet resistance was not affected much by the ion dose, as long as the initial dose is higher than 1015 cm−2. The plasma immersion method was shown to be a good alternative to the ion shower, as the doping could be performed in a relatively short time without causing a structural damage to the poly-Si film.

We have applied real-time spectroscopic ellipsometry (RTSE) as both an in-situ diagnostic and post-growth analysis tool for hydrogenated amorphous silicon (a-Si:H)/crystalline silicon (c-Si) heterojunction with intrinsic thin-layer (HIT) solar cells grown by hot-wire chemical vapor deposition. RTSE enables precise thickness control of the 5 to 25 nm layers used in these devices, as well as monitoring crystallinity and surface roughness in real time. Utilizing RTSE feedback, but without extensive optimization, we have achieved a photovoltaic energy conversion efficiency of 14.1% on an Al-backed p-type Czochralski c-Si wafer coated with thin i and n layers on the front. Open-circuit voltages above 620 mV indicate effective passivation of the c-Si surface by the a-Si:H intrinsic layer. Lifetime measurements using resonant coupled photoconductive decay indicate that surface recombination velocities can approach 1 cm/s. RTSE and transmission electron microscopy show that the intrinsic a-Si:H i-layers grow as a mixture of amorphous and nano-crystalline silicon.

A new ultrahigh vacuum setup is presented which is designed for studying the surface science aspects of a-Si:H film growth using various advanced optical diagnostic techniques. The setup is equipped with plasma and radical sources which produce well-defined radicals beams such that the a-Si:H deposition process can be mimicked. In this paper the initial experiments with respect to deposition of a-Si:H using a hot wire source and etching of a-Si:H by atomic hydrogen are presented. These processes are monitored by real time spectroscopic ellipsometry and the etch yield of Si by atomic hydrogen is quantified to be 0.005±0.002 Si atoms per incoming H atom.

The effect of substrate temperature and interface oxide layer on aluminum induced crystallization (AIC) of amorphous silicon (a-Si) is investigated. The effect of substrate temperature on the AIC process was studied by changing the deposition temperate of a-Si from 200 to 300°C in a Al/a-Si/glass configuration. To study the effect of interface oxide on AIC, samples with a-Si/Al/glass, a-Si/Al-oxide/Al/glass, and Al/Si-oxide/a-Si/glass configurations were prepared at a fixed substrate temperature. The samples were annealed in the temperature range from 300°C to 525°C for different periods of time. The X-ray diffraction (XRD) patterns confirmed the crystallization of the a-Si films in the various configurations. From the analysis, we report that crystallization of a-Si happen at 350°C annealing temperature in the Al/a-Si/glass configuration. However, with or without the presence of Si-oxide at the interface, crystallization saturated after annealing for 20 minutes at 400°C. On the other hand, when Al-oxide is present at the interface, higher annealing temperatures and longer annealing times are required to saturate the crystallization of a-Si. Environmental Scanning Electron Microscope (ESEM) and Energy Dispersive X-Ray (EDX) mapping were used to study the surface morphology as well as the layer sequence after crystallization. This analysis revealed that Si-Al layer-exchange happens regardless of the deposited film configuration.