Al Si Contact Research Articles

Al-Si contact formation involving back surface field and voids of PERC

Local Al–Si contacts are a vital part of passivated emitter and rear cell (PERC) solar cells. They are generally fabricated using laser ablation to open the passivation dielectric layers, which is followed by Al paste printing and high-temperature alloying. However, the alloying process of Al–Si contacts and the formation mechanism of voids remain unclear in parts. In this study, based on the Al–Si binary phase diagram, we revealed the formation path associated with the back surface field (BSF) of the Al–Si alloying process and elucidated the formation mechanism of voids for dashed shape local contact openings. Also, a calculation model is proposed to determine and verify the Si concentration at the peak temperature. During peak-temperature firing, an equilibrium is established between the dissolution and diffusion of Si; subsequently, Si in the melt reaches its saturation concentration and laterally diffuses into the Al layer, thereby causing the net outflow of Si with the contact trench expansion and forming partial voids. During cooling, supersaturated Si precipitates to form BSF, while the degree of precipitated Si limits the thickness and forms the hypereutectic alloy. After the BSF formation, the melt was redistributed along the contact trench, resulting in partial voids turning into complete voids and fully filled contacts. This study also provides a new approach for regulating and optimizing Al–Si contacts. • The formation path of the back surface field of Al Si contact is revealed from the phase diagram. • Silicon saturated at peak temperature and BSF thickness is limited by the degree of silicon precipitation. • The contact of Hypereutectic component Al Si will affect the electrical properties. • The formation of voids is related to the lateral diffusion of silicon and the redistribution of melt. • The redistribution of melt transforms partial voids into complete voids.

Co-diffused bi-facial PERT solar cells

The authors present a fully solderable, co-diffused and bi-facial PERT solar cell where the rear BSG/SiN stack acting both as dopant and passivation source reaching a mean efficiency of 20.4 %. The conductivity induced by the presence of the p+ layer at the rear reduces series resistance losses significantly by simultaneously enabling the use of lowly doped Si wafers that reduceB-O degradation while maintaining a high fill factor. Parallel processed, mono-facial pendants reach the same efficiency level. A higher FF for the mono-facials is compensated by a higher Voc for the bi-facial PERT devices. SEM analysis on test structures show an overall superior, local Al-Si contact in terms of voiding and BSF thickness with reducing printed finger width. At reverse bias, the “conventional process” shows a typical behaviour for such architecture of elevated current flow around the edges. For the “adapted process” up to -12 V, the authors demonstrate a virtually shunt-free device exhibiting reverse currents far below 1 A with no additional process step.

General multilayer heat transfer model for optical-based thermal characterization techniques

Optical-based techniques have been used to characterize thermal energy transport in materials for a few decades. To implement these techniques, a modulated heat source (either pulse or continuous wave) is always employed to excite a periodic temperature variation, which subsequently causes variation of temperature within the material and various other parameters that are a function of temperature. Ambient pressure, surface infrared properties, and radiation are all affected by the temperature variation and serve as indicators to indirectly detect the temperature change. To extract the properties of interest, theoretical models and solutions are necessary. In this work, we propose a general heat transfer model in multilayer structures. We also derived general solutions in the frequency domain using recursive matrix relationships. The recursive matrix simplifies the heat transfer analysis by only considering key parameters within the adjacent layers. In addition, the general analytical solution is only composed of a group of equivalent resistances that yield the contribution to the overall phase shift from each interface and can be used to directly calculate the phase difference within similar configurations. We applied this model and the associated solutions to analyze the data from the phase-sensitive transient thermoreflectance (PSTTR) technique. In contrast with typical thermoreflectance techniques, in the PSTTR technique, the pump and probe beams are applied on the opposite surfaces of the sample. We conducted PSTTR measurements on different multilayer structures, and then determined the thermal/physical properties of interest by fitting the theoretical solutions to the experimental data. The thermal conductivity of thermal grease (TC-5022) was determined to be 3.5W/(mK). Where appropriate, the fitted results were in excellent agreement with results from the literature, which validates this general model and the solution methodology. As another example of a multilayer structure, a novel direct-bonded interface that contains four layers, was studied. Its overall thermal resistance was 0.46mm2K/W, including the Al–Al contact resistance of 0.33mm2K/W and Al–Si contact resistance of 0.06mm2K/W. Using this general model along with the PSTTR technique, an in-depth understanding of the interfacial resistance was achieved by investigating the contributions from each component in the interface.

Large area p-type PERL cells featuring local p+ BSF formed by laser processing of ALD Al2O3 layers

In this work we present large area p-type PERL solar cells featuring local p+ Back-Surface-Field (BSF) obtained by pico-second (ps) laser processing of thin ALD Al2O3 layers capped either by PECVD SiNx or PECVD SiOx. In this specific approach, the laser processing is performed through the rear passivation stack, whereby the laser simultaneously patterns the dielectrics for the subsequent Al–Si contact formation, whilst incorporating the Al atoms from the Al2O3 film into the underlying Si bulk. With this technique, we register substantial dopant incorporation for 10nm of ALD Al2O3, without using additional doping sources. When this process is coupled to a front metallization scheme based on Cu plating, p-PERL cells can be fabricated adopting only low temperature process steps, which avoids thermal stability issues related to the Al2O3 layer, such as crystallization and blistering, and improves the properties of the rear reflector by avoiding Al–Si alloying. Efficiencies up to 20.7% are reported on 6-in. Cz–Si solar cells featuring local p+ BSF formed by laser doping through a rear stack composed of 10nm ALD Al2O3 and 120nm PECVD SiNx.

On the Refinement Mechanism of Silicon in Al-Si-Cu-Zn Alloy with Addition of Bismuth

Obtained results of micro and nano studies reveal that bismuth refines the silicon in which the flake silicon changed to lamellar structure with reduction in twin spacing from 160 to 75 nm. Bismuth segregates towards the inter-dendritic regions and decreases the Al-Si contact angle resulting in suppression of the silicon growth causing refinement of the eutectic structure. Increased recalescence temperature and time confirmed that the refinement effect is attributed to the growth stage.

Distribution of Silicon in the Aluminum Matrix for Rear Passivated Solar Cells

Abstract Solar cell concepts with passivated, locally contacted rear side have gained much interest over the last years due to the development of advanced passivation techniques. However, there are still open questions regarding the understanding of the local Al-Si contact formation and its influence on the local back-surface-field formation for industrial screen-printed aluminum pastes. In this work we show that the geometry of the rear pattern (contact size and contact spacing) strongly influences the Al-Si interaction in the local contact openings: Si diffuses in Al through narrow dielectric barrier openings and spreads laterally to a determined limit determined by the firing temperature. We found that the contact spacing can affect the tendency for formation of Kirkendall voids below the Al-Si contacts instead of an eutectic layer. When decreasing the contact spacing and therefore the overlap of Al on each side of the local opening, the melt saturates faster and the formation of a deep (high-quality) local back-surface-field is increased. This may minimize the presence of voids. This physical observation helps on the design of the rear pattern for rear passivated solar cells and gives further understanding of the Al-Si interaction during the firing process.

Microstructure development and metallurgical analysis of the Al–Si contact reaction

The mechanism of the contact reaction of aluminum–silicon was investigated in this paper. By analyzing the results through SEM and EDS, the behavior of the eutectic liquid of aluminum–silicon was studied. SEM photographs showed that the contact reaction took place not only on the Al surface but also over the cross-section. The eutectic liquid of Al–Si was first produced in the Al grain boundary. The results could be analyzed from thermodynamics and dynamics. In order to obtain an ideal brazed joint, a series of experiments was carried out to secure optimized parameters. If suitable conditions were chosen, sound contact reaction brazed joints of Al could be obtained using Si powder as an interlayer.

PC-Based Contact Resistance Measurement System using TLM Method

PC-based interfacial contact resistance measurement for a metal semiconductor contact is described which associates with the TLM (Transmission Line Model). The Al-Si contact samples using transmission line model are fabricated using the photolithography technique. The nature of alloyed contacts has been studied and the contact resistance Rc measured wiht the help of PC-baed Rc measurement system developed with four probe assembly. As the system is PC-based, it is able to acquire a number of data points for the same sample quickly and perform the required calculations in order to directly display the Rc.

How to fabricate low-resistance metal-diamond contacts

Three types of mesa etched contact resistance test structures with Al Si , TiAu, TiWN-Au contacts are compared. The contact resistivity was calculated using transmission-line model (TLM) theory which is fully applicable to diamond. For Al Si contacts the lowest contact resistivity was obtained after annealing at 500 °C in dry nitrogen. The lowest contact resistivity for as-deposited contacts was found for TiAu. In general, the contact resistivity drops with increasing levels of B dopant and depends on the metallization scheme. The dependence of the contact resistivity on the surface concentration of B for as-deposited Al Si contacts can be fitted by a power-law indicative of spatial inhomogenous areas in the contact region leading to an additional spreading resistance component. The contact resistivity depends exponentionally on the operating temperature for as-deposited Al Si contacts but the thermal coefficient is independent of the doping level for heavily B-doped diamond films, which is typical for tunneling.

Surface electronic states in the fundamental gap of the clean and group III–V doped Si(110) surface

The surface electronic states in the fundamental gap of the clean Si(110) surface and the effect on them of Al, Pb and Sb adsorption were studied with the help of field effect kinetics and photovoltage experiments. The surface geometry and chemical composition were characterized by LEED and AES. The obtained results showed that the electron spectra of the clean Si(110) surface with ‘16 × 2’ reconstruction have empty and filled bands. They were located near the midgap and separated from each other by an energy interval E gs ≈ 0.40 eV. Al, Pb and Sb adsorption on the Si(110)-‘16 × 2’ phase at room temperature was followed by the formation of (1 × 1) LEED patterns with strong background for coverages <2 ML. Drastic changes of the surface electron spectra under adsorption of these elements were revealed. Namely, the Si(110) Al system contained shallow states only with an amazingly large gap E gs ≈ 0.80 eV (!) between them. The surface barrier height corresponded to the AlSi(110) contact potential difference. On the contrary, the Si(110) Pb phase demonstrated one partially filled surface band with the top edge at 0.7 eV above the bulk valence band maximum. Finally, rather small changes were introduced in the surface electron spectra by Sb atoms. Some qualitative explanations of these effects were proposed concerning the links between atomic and electronic structures of the surface. In particular (1 × 1)Al RR and (1 × 1)Pb primitive unit cells could be suggested to explain the observed electron spectra changes.

Very low resistivity Al [sbnd]Si ohmic contacts to boron-doped polycrystalline diamond films

The effects on the contact resistivity of annealing Al Si (99:1) contacts on diamond with different B concentrations have been investigated. The change of the current-voltage characteristic from rectifying to ohmic on lightly doped samples, and the drop of the contact resistivity by orders of magnitude for more heavily B-doped samples after annealing at 450 °C in N 2, are attributed to the formation of SiC at the metal-diamond interface. The existence of the SiC interface has been verified by X-ray-induced photoelectron spectroscopy. In addition, the contact resistivity is a very sensitive function of doping at high doping concentrations. Contact resistivities as low as about 10 −7 θ cm 2 have been achieved.

Theoretical Evaluation of Compositional Contrast of Scanning Electron Microscope Images

The compositional contrast in the scanning electron microscope image is calculated for Al-Cu, Si-Cu and Al-Si contacts. An electron scattering phenomenon in the specimen is simulated in a direct manner. Electron refraction at the boundary, because of the agreement of each Fermi energy at the boundary, is precisely taken into account. The backscattered electron image shows better resolution than the secondary electron image in terms of the boundary contrast when the primary electron energy is 1 keV. The signal intensity varies depending on materials adjacent to the location observed. The ultimate resolution of the compositional contrast of the scanning electron microscope can be below 1 nm.

A study of the interfacial layer of Al and Al(1% Si)Si contacts using a zero-layer ellipsometry model

In this work, the interfacial layer of the AlSi and Al(1% Si)Si contacting system annealed at different temperatures have been studied using ellipsometry incorporated with a zero-layer model. The results show that the thickness of the interfacial P + layer between Al or Al(1% Si) and the Si-substrate increases with the annealing temperature and the thickness of the P +-layer of the Al(1% Si)Si system is generally lower than that of the AlSi system. Also, the fraction of Al in the interfacial layer decreases with the annealing temperature for the AlSi system, but increases for the Al(1% Si)Si system. The electrical measurements show that the interface of the Al(1% Si)Si contact is more stable than that of the AlSi contact.

Characterization of TiSi2/TiN thin films on Si by HREM

Titanium Nitride is commonly used as a diffusion barrier layer in Al-Si contacts. There are different techniques employed to deposit TiN layers on silicon, out of them thermally grown and reactively sputtered are most common. Reactive sputtering is a one step process and is favoured for manufacturing. The orientation of TiN thin films controls the orientation of Al films deposited. As sputtered films has random orientation hence the overlying Al films is randomly oriented as well. In order to achieve higher electromagnetic resistance, Al films should be well oriented. Thermally grown TiN thin films are highly oriented but their thickness is limited to about 20nm. We have used HREM and micro-analytical techniques to analyze the thickness, structure and compositions of various layers formed during thermal growth of TiN layers deposited under different conditions.Thermally grown Ti films on Si &lt;100&gt; wafers of varying thicknesses were deposited and annealed in NH3 atmosphere at different temperatures and times. Second layer of titanium was deposited and processed on some of the samples (double process) in order to increase the thickness of TiN layer.

I-V Characteristics and Interface Properties of Al-Si(P) Contacts by the Krf Excimer Laser Induced Recrystallization

ABSTRACTAl was deposited on the both sides of p-type Si by the sputtering, and the pulsed KrF excimer laser was irradiated on the one side and both sides of Al dot contacts. Due to the thin-high doping recrystallized layer, the transitions from the Schottky diodes to the ohmic contacts via backward diodes were observed in I-V measurement. Good N-shape negative resistances and notches were observed in the diodes, and ohmic contacts. Possible interpretations of the orgin of these phenomena are discussed. And, the simple model of the effective barrier height reduction under the thermionic field emission was developed using WKB approximation, and tunneling theory. By the depth Auger analysis, Al, and Si profiles near Al-Si interface after the laser irradiation were observed.

Formation of Ag/Si and Al/Si films by ion assisted deposition

The Al and Ag films made by IVD were significantly modified in both mechanical and electrical properties. The cause lies in the deposition parameters: bombardment energy and ion/atom arrival rate ratio. It was found that the Al film hardness, the AlSi contact resistivity and the AlSi adhesion were improved in the deposition range: energy 250–500 eV and arrival ratio 0.06–0.10.

TiW AND Al CONTACTS TO SHALLOW p+ JUNCTIONS - A COMPARISON BETWEEN FURNACE AND RAPID THERMAL ANNEALING (RTA)

The use of sputter deposited TiW (15/85 w.t.%) as a direct contact material to p+ silicon is evaluated. The effect on the contact resistance of different post-implant activation conditions, e.g. rapid thermal processing and conventional furnace annealing in dry or wet ambients, has been studied. For comparison parallell measurements were done on identically processed wafers metallized with aluminium. The resulting dopant profiLes were determined by spreading resistance profiling (SRP). Schottky barrier height measurements were performed on n-type wafers metallized with TiW or Al. The respective values were ? Bn = 0.55 eV and ? Bn = 0.8 eV. The results from contact resistance and SRP measurements show that a high surface carrier concentration is necessary in order to achieve a low contact resistance. This effect was more pronounced for the material with the higher barrier height to p+ silicon i.e. TiW was found to be sore sensitive to a small variation in the boron surface carrier concentration. Despite of this it was observed that given a high surface carrier concentration, after RTP the performance of TiW contacts to p+ silicon surpassed that of Al. in case of a direct Al-Si contact, silicon precipitation occurs in the contact hole thereby reducing the effective contact area. This will cause an increase in the resistance of contacts which becomes more pronounced for small dimensions. With TiW, silicon precipitation is avoided.

Effect of MS contact on the electrical behaviour of solar cells

The electrical behaviour of the AlSi contact and its effect on the performance of solar cells has been studied. When the sheet resistance ϱ s of the n-layer is greater than about 200 Ω/□ the contact will function as a rectifier and the dark I– V characteristics of the solar cell is mainly limited by the reverse biased MS contact. This accounts for the large ideality factor (> 2) observed experimentally. In addition the illuminated I– V curve of the solar cell in the fourth quadrant has an S-shape with an inflection around zero cell current, and the conversion efficiency is considerably reduced. In fact this inflection is a diagnostic tool for a rectifying MS contact. A circuit model has been developed to describe the behaviour of a solar cell with any internal construction and under all operating conditions. This model is confirmed by the electrical behaviour. Finally fabrication methods are proposed to overcome the contact problems associated with low doping concentration at the metal-semiconductor interface.

Simultaneous formation of TiN and TiSi2 by lamp annealing in NH3 ambient and its application to diffusion barriers

The dependence of TiN/TiSi2 bilayer formation on Si by lamp annealing of Ti upon annealing temperature, ambients, and impurity in Si was investigated. In NH3 ambient, nitridation ratio on undoped Si at 680 and 800 °C is 40% and 25%, respectively. When annealing is performed in N2, the nitridation ratio of Ti is only about 25% at both 680 and 800 °C. The above dependence of nitridation ratio on annealing ambient is due to the difference in decomposition energy of gas molecules. The nitridation ratio increases on arsenic-implanted Si because of a retardation effect of the arsenic on the silicidation reaction. The bilayer formation process was applied to the AlSi contact metallization on the n+ diffused layer. From measurements on electrical characteristics of the AlSi/TiN/TiSi2/n+ Si contact system, the following results were obtained: (1) Contact resisitivity is 3.5–5.3×10−7 Ω cm2. (2) The TiN/TiSi2 contact electrode is thermally stable against Al. In particular, when the bilayer is formed in NH3 at comparatively lower temperature, its thermal stability is still further improved. This is due to the formation of the thicker surface TiN layer.

A Highly Stable Al–Si Contact to Mo-Silicided Shallow Junctions

Highly stable (up to 550°C) contacts have been realized in an Al–2%Si contact to n+–p and p+–n shallow junctions (Xj∼0.16 µm) covered by a 0.1 µm thick uniform MoSi2 layer, which was formed by the ion implantation through metal ITM technique. Low contact resistance was maintained, at submicron (0.5 µm square) Al–Si/MoSi2 contacts, after 550°C sintering for 30 minutes.