Application of power loss calculation to estimate the specific contact resistance of the screen-printed silver ohmic contacts of the large area silicon solar cells

The establishment of a suitable contact formation methodology is a critical part of the technological development of any metal-to-semiconductor contact structure. Many test structures and methodologies have been proposed to estimate the specific contact resistance (ρ_c) of the planar ohmic contacts formed on the heavily doped semiconductor surface. These test structures are usually processed on the same wafer to monitor a particular process. In this study, new experimental procedure has been evolved to assess the value of ρ_c of the screen-printed front silver (Ag) thick-film metal contact to the silicon surface. The essential feature of this methodology is that it is an iteration technique based on the calculation of power loss associated with various resistive components of the solar cell normalized to the unit cell area. Therefore, this method avoids the complexity of making the design of any lay out of a standard contact resistance test structure like transmission line model (TLM) or Kelvin resistor, etc. It was shown that value of specific contact resistance of the order of 1.0 × 10⁻⁵ Ω−cm ² is measured for the Ag metal contacts formed on the n⁺ silicon surface. This value is much lower than the ρ_c data previously reported for the screen-printed Ag contacts. The sintering process of the front metal contact structure at different furnace setting is carried out to understand the possible wet interaction and metal contact formation as a function of the firing. Therefore, the study is further extended to study the peak firing temperature dependence of the ρ_c of screen-printed Ag metal contacts. It will help to assess the specific contact resistance of the ohmic contacts as a function of firing temperature of sintering process.

---

SEM and specific contact resistance analysis of screen-printed Ag contacts formed by fire-through process on the shallow emitters of silicon solar cell

The SEM and specific contact resistance measurements of the Ag metal contact formed by applying a fire-through process on the shallow emitter region of the silicon solar cell have been investigated. The metal contact consists of screen-printed Ag paste patterned on the silicon nitride (Si₃N₄) deposited over the n⁺-Si emitter region of the solar cell. The sintering step consists of a rapid firing step at 800 °C or above in air ambient. This is followed by an annealing step at 450 °C in nitrogen ambient. It enables to drive the Ag metal paste onto the Si₃N₄ layer and facilitates the formation of an Ag metal/p-Si contact structure. It serves as the top metallization for the screen-printed silicon solar cell. The SEM measurement shows that sintering of the Ag metal paste at 800 °C or above causes the Ag metal to firmly coalesce with the underlying n⁺-Si surface. A thin layer of conductive glassy layer is also presents at the interface of the Ag metal and n⁺-Si surface. The electrical quality of the contact structure was characterized by measuring the specific contact resistance, ρ _c (in Ω-cm²) using the iteration technique based on the power loss calculation for the solar cell. It shows that best value of ρ _c  = 2.53 × 10⁻⁵ Ω-cm² is estimated for the Ag metal contact sintered at temperature above 800 °C. This value of ρ _c is two orders of magnitude lower than the typical value of ρ _c  = 3 × 10⁻³ Ω-cm² reported previously for the Ag contacts of the solar cell. Such low value of ρ _c for the Ag metal contacts indicates that fire-through process results in excellent ohmic properties. The plot of the ρ _c versus impurity doping level (N _s ) shows that measured value of the ρ _c follows a linear relationship with the N _s as predicted by the theory for the heavily doped semiconductor surface. Hence, carrier injection across the Schottky barrier height is quite appropriate to explain the observed ohmic properties of the Ag metal contacts on the n⁺-Si surface of the silicon solar cell.

---

Specific contact resistance measurements of the screen-printed Ag thick film contacts in the silicon solar cells by three-point probe methodology and TLM method

The specific contact resistance of the screen-printed Ag contacts in the silicon solar cells has been investigated by applying two independent test methodologies such as three-point probe (TPP) and well-known transfer length model (TLM) test structure respectively. This paper presents some comparative results obtained with these two measurement techniques for the screen-printed Ag contacts formed on the porous silicon antireflection coating (ARC) in the crystalline silicon solar cells. The contact structure consists of thick-film Ag metal contacts patterned on the top of the etched porous silicon surface. Five different contact formation temperatures ranging from 725 to 825 °C for few minutes in air ambient followed by a short time annealing step at about 450 °C in nitrogen ambient was applied to the test samples in order to study the specific contact resistance of the screen-printed Ag metal contact structure. The specific contact resistance of the Ag metal contacts extracted based on the TPP as well as TLM test methodologies has been compared and verified. It shows that the extraction procedure based on the TPP method results in specific contact resistance, ρ _c  = 2.15 × 10⁻⁶ Ω-cm² indicating that screen-printed Ag contacts has excellent ohmic properties whereas in the case of TLM method, the best value of the specific contact resistance was found to be about ρ _c  = 8.34 × 10⁻⁵ Ω-cm². These results indicate that the ρ _c value extracted for the screen-printed Ag contacts by TPP method is one order of magnitude lower than that of the corresponding value of the ρ _c extracted by TLM method. The advantages and limitations of each of these techniques for quantitatively evaluating the specific contact resistance of the screen-printed Ag contacts are also discussed.     

Specific contact resistance and metallurgical process of the silver-based paste for making ohmic contact structure on the porous silicon/p-Si surface of the silicon solar cell

Porous silicon has been considered as a promising optoelectronic material for developing a variety of optoelectronic devices and sensors. In the present study, the electrical properties and metallurgical process of the screen-printed Ag metallization formed on the porous silicon surface of the silicon solar cell have been investigated. The contact structure consists of thick-film Ag metal contact patterned on the top of the porous silicon surface. The sintering process consists of a rapid firing step at 750–825 °C in air ambient. It results in the formation of a nearly perfect contact structure between the Ag metal and porous silicon/p-Si structure that forms the top metalization for the screen-printed silicon solar cells. The SEM picture shows that Ag metal firmly coalesces with the silicon surface with a relatively smooth interfacial morphology. This implies that high temperature fire-through step has not introduced any signs of adverse effect of junction puncture or excessive Ag indiffusion, etc. The three-point probe (TPP) method was applied to estimate the specific contact resistance, ρ _c (Ω-cm²) of the contact structure. The TPP measurement shows that contact structure has excellent ohmic properties with ρ _c = 1.2 × 10⁻⁶ Ω-cm² when the metal contact sintered at 825 °C. This value of the specific contact resistance is almost three orders of magnitude lower than the corresponding value of the ρ _c = 7.35 × 10⁻³ Ω-cm² obtained for the contact structure sintered at 750 °C. This improvement in the specific contact resistance indicates that with increase in the sintering temperature, the barrier properties of the contact structure at the interface of the Ag metal and porous silicon structure improved which in turn results a lower specific contact resistance of the contact structure.     

Low-resistance and highly transparent Ag/IZO ohmic contact to p-type GaN

The electrical, structural, and optical characteristics of Ag/ZnO-doped In₂O₃ (IZO) ohmic contacts to p-type GaN:Mg (2.5 × 10¹⁷ cm^− 3) were investigated. The Ag and IZO (10 nm/50 nm) layers were prepared by thermal evaporation and linear facing target sputtering, respectively. Although the as-deposited and 400 °C annealed samples showed rectifying behavior, the 500 and 600 °C annealed samples showed linear I-V characteristics indicative of the formation of an ohmic contact. The annealing of the contact at 600 °C for 3 min in a vacuum (~ 10^− 3 Torr) resulted in the lowest specific contact resistivity of 1.8 × 10^− 4 Ω·cm² and high transparency of 78% at a wavelength of 470 nm. Using Auger electron spectroscopy, depth profiling and synchrotron X-ray scattering analysis, we suggested a possible mechanism to explain the annealing dependence of the electrical properties of the Ag/IZO contacts.     

Requirements imposed on the electrical properties of the absorber layer in CdTe-based solar cells

The requirements for minimizing the electric losses in the CdTe layer in CdS/CdTe thin-film solar cells are discussed. It is shown that for achieving the total absorption of the radiation and to avoid electrical losses, the separation between Fermi level and the valance band should not be more than 0.3 eV. In order to fix the Fermi level near the valence band, it is necessary to dope with acceptor impurities which can introduce shallow level in the band gap with a concentration considerably exceeding the concentration of native impurities and defects (10¹⁵–10¹⁶ cm⁻³). Taking into account the fact that the location of the Fermi level in the bandgap of a semiconductor depends on the degree of compensation, the energy of ionization of the impurity should not be greater than 0.05–0.15 eV.

---

Performance of Pd–Ge based ohmic contacts to n-type GaAs

The performance of Pd–Ge based ohmic contacts, with and without Ti–Pt or Ti–Pt–Au capping layers, has been investigated. The contacts were deposited by electron beam evaporation, then characterized electrically using a modified transmission line method (TLM) and structurally using both cross-section and plan-view transmission electron microscopy (TEM). Although both capped and non-capped contact structures underwent the same phase transformations during annealing, capped contacts had significantly better contact resistances (a minimum value of 4×10-7 Ω cm² was achieved) – almost three orders of magnitude better. The superior performance is attributed to the capping layers providing protection for the Pd–Ge layers during contact processing, where the metallization was exposed to a CF₄–O₂ plasma, oxyen descumming, organic solvents and deionized water. Non-capped contacts exhibited PdGe decomposition and oxidation of exposed Ge. Long-term reliability testing of capped contacts showed virtually no change in contact resistance at 235°C (1350 h) and a sevenfold increase after ageing at 290°C for 370 h. There were no phase changes during ageing; the increase in contact resistance was attributed to interdiffusion between Ge and GaAs. This revised version was published online in July 2006 with corrections to the Cover Date.     

Thermally stable and low-resistance W/Ti/Au contacts to n-type GaN

We report on the formation of thermally stable and low-resistance Ti/Au-based ohmic contacts to n-type GaN (4.0 × 10¹⁸ cm⁻³) by using a W barrier layer. It is shown that the electrical characteristic of the sample is considerably improved upon annealing at 900 °C for 1 min in a N₂ ambient. The contacts produce the specific contact resistance as low as 6.7 × 10⁻⁶ Ω cm² after annealing. The Norde and current–voltage methods are used to determine the effective Schottky barrier heights (SBHs). It is shown that annealing results in a reduction in the SBHs as compared to that of the as-deposited sample. Auger electron spectroscopy (AES), scanning transmission electron microscopy (STEM) and X-ray diffraction (XRD) examinations show that nitride and gallide phases are formed at the contact/GaN interface. Based on the AES, STEM and XRD results, a possible ohmic formation mechanism is described and discussed.     

Spark Plasma synthesis and diffusion of Cu and Ag in vanadium mixed valence oxides

Spark Plasma sintering (SPS) technique allows powders to be compacted at low temperature with a very short holding time. The powder loaded into a carbon die is heated via direct current pulses and simultaneously submitted to an uni-axial pressure of several MPa. Full density of the sample is achieved within minutes. This process is used to study Cu and Ag metals interactions with V₂O₅ oxide. Syntheses of M _x V₂O₅ phases (M = Cu, Ag) have been achieved within minutes. Thus Cu and Ag atoms penetrate microcrystals of V₂O₅ oxide at a high speed, shearing its crystal network and simultaneously rebuilding the crystal structures of the prototype networks β, β′, ε or δ M _x V₂O₅. To account for the formation of these phases identified by X-ray diffraction, structural mechanisms are proposed. Cu and Ag atomic diffusion parameters have been determined from energy dispersive X-ray spectroscopy (EDX) and electron micropobe analysis (EPMA) line scans. High values of diffusion coefficients have been determined. Cu atoms diffuse faster than Ag, D _Cu ≈ 4 × 10⁻⁸ m²/s and D _Ag ≈ 0.5–1 × 10⁻⁹ m²/s in ε and δ M _x V₂O₅ phases, respectively. Their formation may also be used as a model for further investigations into the diffusion mechanisms of atoms in solids and for a better understanding of the SPS process.

---

The effect of molybdenum on the microstructure and creep behavior of Ti–24Al–17Nb–<Emphasis Type="Italic">x</Emphasis>Mo alloys and Ti–24Al–17Nb–<Emphasis Type="Italic">x</Emphasis>Mo SiC-fiber composites

The effect of molybdenum (Mo) on the microstructure and creep behavior of nominally Ti–24Al–17Nb (at.%) alloys and their continuously reinforced SiC-fiber composites (fiber volume fraction = 0.35) was investigated. Constant-load, tensile-creep experiments were performed in the stress range of 10–275 MPa at 650 °C in air. A Ti–24Al–17Nb–2.3Mo (at.%) alloy exhibited significantly greater creep resistance than a Ti–24Al–17Nb–0.66Mo (at.%) alloy, and correspondingly a 90°-oriented Ultra SCS-6/Ti–24Al–17Nb–2.3Mo metal matrix composite (MMC) exhibited significantly greater creep resistance than an Ultra SCS-6/Ti–24Al–17Nb–0.66Mo MMC. Thus, the addition of 2.3 at.% Mo significantly improved the creep resistance of both the alloy and the MMC. An Ultra SCS-6 Ti–25Al–17Nb–1.1Mo (at.%) MMC exhibited creep resistance similar to that of the Ultra SCS-6/Ti–25Al–17Nb–2.3Mo (at.%). Using a modified Crossman model, the MMC secondary creep rates were predicted from the monolithic matrix alloys’ secondary creep rates. For identical creep temperatures and applied stresses, the 90°-oriented MMCs exhibited greater creep rates than their monolithic matrix alloy counterparts. This was explained to be a result of the low interfacial bond strength between the matrix and the fiber, measured using a cruciform test methodology, and was in agreement with the modified Crossman model. Scanning electron microscopy observations indicated that debonding occurred within the carbon layers of the fiber-matrix interface.

---

A quenching apparatus for the gaseous products of the solar thermal dissociation of ZnO

Rapid cooling for avoiding the recombination of Zn vapor and O₂ derived from the solar thermal dissociation of ZnO is investigated using a thermogravimeter coupled to a quenching apparatus. The ZnO sample, which is placed in a cavity receiver and directly exposed to concentrated solar irradiation, underwent dissociation in the temperature range 1,820–2,050 K at a rate monitored by on-line thermogravimetry. The product gases were quenched by water-cooled surfaces and by injection of cold Ar at cooling rates from 20,000 to 120,000 K/s, suppressing the formation of ZnO in the gas phase and at the walls. Zinc content of the collected particles downstream varied in the range 40–94% for Ar/Zn(g) dilutions of 170 to 1,500.

---

Morphology and constitution of the compound layer formed on nitrided Fe–4wt.%V alloy

The microstructure of the compound (“white”) layer formed on the surface of Fe–4wt.%V alloy, by nitriding in a gas mixture of ammonia and hydrogen at 580 °C, has been investigated by employing light and scanning electron microscopy, X-ray diffraction and electron probe microanalysis. The compound layer is dominantly composed of γ^|-Fe₄N nitride. Quantitative analysis of the composition data demonstrated that V is present in the compound layer as VN precipitates, i.e. V is not taken up significantly in (Fe, V) nitrides. A mechanism for compound-layer formation has been proposed.

---

Peculiarities in the mechanism of current flow through an ohmic contact to gallium phosphide

The temperature dependence of the electric resistance of the In-GaP ohmic contact has been studied in the range from 77 to 420 K. The resistance was measured in GaP plates of various thickness with two In ohmic contacts. The measured ohmic contact resistance increases with temperature in the interval from 230–420 K. It is suggested that the In-GaP ohmic contact is formed by metallic shunts appearing upon deposition of In atoms on dislocations and other imperfections present (with a density evaluated at (4.5–8)×107 cm⁻²) in the subsurface region of the semiconductor.     

Development of refractory ohmic contact materials for gallium arsenide compound semiconductors

Recent strong demands for optoelectronic communication and portable telephones have encouraged engineers to develop optoelectronic devices, microwave devices, and high-speed devices using hetero-structural GaAs-based compound semiconductors. Although the GaAs crystal growth techniques had reached a level to control the compositional stoichiometry and crystal defects on a nearly atomic scale by the advanced techniques such as molecular beam epitaxy and metal organic chemical vapor deposition techniques, development of ohmic contact materials (which play a key role to inject external electric current from the metals to the semiconductors) was still on a trial-and-error basis.Our research efforts have been focused to develop low resistance, refractory ohmic contact materials to n-type GaAs using the deposition and annealing techniques, and it was found the growth of homo-or hetero-epitaxial intermediate semiconductor layers (ISL) on the GaAs surface was essential for the low resistance ohmic contact formation. In this paper, two typical examples of ohmic contact materials developed by forming ISL were given. The one was refractory NiGe-based ohmic contact material, which was developed by forming the homo-epitaxial ISL doped heavily with donors. This heavily doped ISL was discovered to be formed through the regrowth mechanism of GaAs layers at the NiGe/GaAs interfaces during annealing at elevated temperatures. To reduce the contact resistance further down to a value required by the device designers, an addition of small amounts of third elements to NiGe, which have strong binding energy with Ga, was found to be essential. These third elements contributed to increase the carrier concentration in ISL. The low resistance ohmic contact materials developed by forming homo-epitaxial ISL were Ni/M/Ge where a slash ‘/’ denotes the deposition sequence and M is an extremely thin (∼5 nm) layer of Au, Ag, Pd, Pt or In. The other was refractory In_xGa_1−xAs-based ohmic contact materials which were developed by forming the hetero-epitaxial ISL with low Schottky barrier to the contacting metals by growing the In_xGa_1−xAs layers on the GaAs substrate by sputter-depositing In_xGa_1−xAs targets and subsequently annealing at elevated temperatures. To reduce the contact resistance, it was found that this In_xGa_1−xAs (ISL) layer had to have In compositional gradient normal to the GaAs surface: the In concentration being rich at the metal/In_xGa_1−xAs interface and poor close to the In_xGa_1−xAs/GaAs interface. This concentration graded ISL reduced both the barrier heights at the metal/ISL and ISL/GaAs interfaces and reduced the contact resistance. The ohmic contact materials developed by forming hetero-epitaxial ISL was In_0.7Ga_0.3As/Ni/WN₂/W. These contact materials formed refractory compounds at the interfaces, which was also found to be essential to improve thermal stability of ohmic contacts used in the GaAs devices.     

Hybrid organic–inorganic materials based on polypyrrole and 1,3-dithiole-2-thione-4,5-dithiolate (DMIT) containing dianions

The synthesis of hybrid materials by electropolymerization of pyrrole and inorganic complexes based on the DMIT ligand (1,3-dithiole-2-thione-4,5-dithiolate), e.g. [NEt₄]₂[M(DMIT) _n ] (M = Ni, Pd or Pd, n = 2; M = Sn, n = 3], in acetonitrile solution is reported. Spectroscopic data showed that DMIT-containing anions, [M(DMIT) _n ]²⁻, were inserted into the polypyrrole framework without chemical modification during the electropolymerization process. Cyclic voltammetry showed that materials obtained were electroactive, undergoing redox processes related to both the conducting polymer and the counteranions. The electrochemical results also suggest that, in the case of the transition metal containing films, the counteranions are not trapped in the PPy matrix but undergo anion exchange during the redox cycle of PPy. However, an opposite behaviour was observed with the film with [M(DMIT) _n ]²⁻. The films exhibit good thermal stabilities and have conductivity values expected for semiconductors. This study of these hybrid materials highlights the importance of targeting specific materials for specific applications.

---

Microstructure and stresses in a keatite solid–solution glass–ceramic

The microstructure of a translucent keatite solid–solution glass–ceramic (keatite s.s.) of the LAS-system (Li₂O–Al₂O₃–SiO₂) has been analyzed with SEM, AFM, XRF, XRD, and TEM. The glass–ceramic consists mainly of keatite s.s. with minor secondary phases such as zirconium titanate, gahnite and probably rutile. Furthermore the resistance to temperature differences (RTD) of this glass–ceramic was investigated. It is shown that, in spite of the relatively high coefficient of thermal expansion (CTE) of about 1 × 10⁻⁶ K⁻¹, an improved RTD can be achieved by special ceramization treatment. With this, compressive stresses in the first 100 μm to 150 μm are induced. These stresses can presumably be contributed to a difference in CTE between the surface-near zone and the bulk. Said CTE difference is caused by chemical gradients of CTE-relevant elements, such as Zn, K, and supposedly additional alkali elements such as Li. These stresses are useful to increase the strength and application range of glass–ceramics based on keatite s.s.

---

Melt synthesis of oxide phosphors with K2NiF4 structures: CaLa1? x Eu x GaO4

In order to synthesize compounds of various Perovskite-related structures, we have utilized a novel “melt synthesis technique” for phosphors rather than the conventional solid state reaction techniques. The solid state reactions require multi-step processes of heating/cooling with intermediate grindings to make homogeneous samples. However, for the melt synthesis, it is possible to make a homogeneous sample in a single step within a short period of time (1–60 s) due to the liquid phase reaction in the molten samples, which were melted by strong light radiation in an imaging furnace. In this study, we have prepared a red-phosphor CaLaGaO₄:Eu³⁺ which has a perovskite—related layered K₂NiF₄ structure. Well-crystallized CaLa_1−x Eu _x GaO₄ samples with the K₂NiF₄ structure have been obtained up to x = 0.25, but there was the formation of an olivine phase when x = 0.5–1.0. The red emission at 618 nm increased with the increasing value of x up to x = 0.25.

---

Time-lapse photography of the β-Sn/α-Sn allotropic transformation

When tin transforms from the β to α phase it undergoes a dramatic process. The crystal structure changes from tetragonal to diamond cubic; the material properties transform from a ductile metal to a brittle semiconductor; however the most notable change is the decrease in density, which goes from 7.31 g/cm³ to 5.77 g/cm³ [1]. It can be calculated that this decrease in density is equivalent to an increase in volume of about 26% [2]. Due to this volume increase and the brittleness, the transformed material progressively cracks and eventually falls apart. This could potentially be a threat for tin-rich alloys used in electronics in low temperature applications. Due to the optimal transformation temperature of approximately 240 K and the long time required for the transformation, a direct observation of the phenomenon has not been possible. This study proposes a new method for observing the β/α transformation in situ using a time-lapse photographic technique. This study concentrates on pure tin, but the applicability of the method opens new possibilities for studying the phenomenon for other tin alloys, such as the two commonly encountered eutectics of SnCu and SnAgCu.

---

A comparison of impression and compression creep behavior of polycrystalline Sn

This paper reports on the application of a miniaturized impression creep test to measure the creep behavior of pure polycrystalline Sn, and compares the results to compression creep data on the same sample, in order to experimentally determine a scaling constant to formulate the equivalent uniaxial creep constitutive law from the impression creep data. The creep parameters determined via impression and compression creep are found to be identical, with n ∼ 5 and Q ∼ 42 kJ/mol, indicating that over the tested stress–temperature range, the mechanism is core diffusion controlled dislocation creep. In conjunction with results from previous modeling work, a single conversion factor, κⁿ/C, which depends on material properties, is shown to be usable for converting the impression creep relation to the equivalent uniaxial creep relation, and the experimentally determined value of κⁿ/C for polycrystalline Sn is very close to that obtained via modeling.

---