Showing papers on "Contact resistance published in 1984"

Solar cell contact resistance—A review

Abstract: An overview of ohmic contacts on solar cells is presented The fundamentals of metal-semiconductor contacts are reviewed, including the Schottky approach, Fermi level pinning by surface states, and the mechanisms of thermionic emission, thermionic/field emission, and tunneling for current transport The concept of contact resistance is developed and contact resistance data for several different contact materials on both silicon and gallium arsenide over a range of doping densities are summarized Finally, the requirements imposed by solar cells on contact resistance are detailed

Contact resistance: Its measurement and relative importance to power loss in a solar cell

Abstract: The concept of contact resistivity is discussed briefly and a technique for its measurement is presented. This technique allows for resistive contact material and for the possibility that the semiconductor sheet resistance beneath the contact differs from the semiconductor sheet resistance beside the contact. The test pattern is unique in that the effects of contact resistance are accumulated over the pattern, nearly unencumbered by voltage and current probes which might otherwise influence the current flow. Measurements of contact resistivities for typical solar cell metallizations using this technique are reported to be in the mid 10-6Ω-cm2range. The relative importance of contact resistance compared to other sources of power loss in a solar cell is determined for a typical contact system. Expressions derived in order to make this comparison are useful for evaluating and optimizing a solar cell contact system. Values of series resistance calculated using these expressions are compared with measured values.

Sliding contact material

Abstract: PURPOSE:To enhance the lubricity and wear resistance of an Ag-In alloy with superior sulfurization resistance as a sliding contact material and to reduce the contact resistance by adding a specified amount of at least >=1 kinds among Th, W, Ce, Mo, Nb, Sn, Se and As to the alloy or by further adding an Fe group metal. CONSTITUTION:A material for the contact chips 1a of the commutator 1 of a small-sized motor is obtd. by adding 0.5-5wt% in total of at least >=1 kinds among Th, W, Ce, Mo, Nb, Sr, Se and As to an Ag-In alloy with superior sulfurization resistance contg. 1-10wt% In or by further adding 0.01-0.5% Fe group metal. Oxide of said added metal formed during sliding is softer than In oxide and acts as a lubricant to enhance the wear resistance. By adding the Fe group metal, the grains are made fine to further enhance the wear resistance, and the contact resistance to the wire rods 2 of brushes is reduced, so ununiformity in the number of revolutions of the commutator 1 is considerably reduced.

Heat transfer to packed and stirred beds from the surface of immersed bodies

Abstract: The heat transfer between packed and stirred beds and immersed surfaces is controlled by the contact resistance at the surface followed by the heat penetration resistance of the bulk. Both resistances can be predicted from model equations with sufficient accuracy. The contact resistance and the bulk penetration resistance for packed beds follow from physical properties, while the prediction of the bulk penetration resistance for stirred beds requires the introduction of an empirical parameter, the so-called mixing number in order to describe the random particle motion. The mixing number was found to lie between 2 and 25, depending on the design of the stirrer.

Lateral current crowding effects on contact resistance measurements in four terminal resistor test patterns

Abstract: Lateral current crowding effects on contact resistance measurements in four terminal resistor patterns are discussed by using a computer model based on a three-dimensional resistor network. The model is then applied to extrapolate the contact resistivity in n+, p+ silicon/titanium silicide interfaces. Values in agreement with the ones predicted by the field and thermionic field emission theory are obtained.

Metalization process for headless contact using deposited smoothing material

Abstract: Contacting an underlying region (e.g., doped silicon) through an access hole in an overlying dielectric layer (e.g., p-glass) formerly required flowing the dielectric to smooth the edges of the hole, so that aluminum would deposit smoothly into the hole. The present technique smoothes the side of the hole by forming a smoothing region on the sidewall. Improved aluminum coverage results, as well as allowing a smaller contact head, if desired. Improved contact resistance can be optionally provided by depositing a more conductive layer on the underlying layer prior to forming the sidewall.

Determining specific contact resistivity from contact end resistance measurements

Abstract: A method is described to determine specific contact resistivity from contact end resistance measurements using a transmission line model. A test pattern is described which minimizes the effect of current fringing around contact corners and yields an accurate determination of contact width. With this pattern, the specific contact resistivities measured on 1.3 wt % Si/Al contacts to n+ silicon junctions with different dopings show very consistent values and are independent of contact geometries. The dependence of measured specific contact resistivities on doping concentration is also in good agreement with the predictions of tunneling theory. Surprisingly, the dependence on surface concentration extends well beyond the usual range of electrically active solid solubility.

Time Distribution of Intermittents Versus Contact Resistance for Tin-Tin Connector Interfaces During Low Amplitude Motion

Abstract: An experimental study was made of the relationship between he static contact resistance increase in a separable connector degraded by interfaee motion and short duration dynamic events which could represent intermittent failures in high-speed digital circuits. These studies were a continuation of earlier work which demonstrated the occurrence of short duration "opens" in the range 5-10 ns duration for increases in contact resistance of as little as 5-6 m \Omega . No time discrimination for these events was obtained. The current studies had the primary objective of resolving the duration of such events and to show the relationship between event duration and static contact resistance. It was shown that as contact resistance increases, the probability increases for the development of longer duration events. From pulse widths >20 ns at \delta R/R=5-6, durations increased to more than 1 \mu s at \delta R/R = 40-45. These results imply that, in practice, a range of failure criterion could be applied for test purposes depending upon sensitivity of the specific application towards tort duration open events.

Contact resistance of LPCVD W/Al and PtSi/W/Al metallization

Abstract: The sensitivity of measured specific contact resistivity to surface doping concentration has been investigated for selectively deposited LPCVD W contacts to n+ and p +Si with surface concentrations from 1018to 1020cm-3. W contact resistance to n+ Si is about a factor of 20 lower than that of Al; W contact resistance to p +Si is comparable to that of Al. Ultralow resistance, stable contacts with self-aligned PtSi, and W to p +Si are demonstrated.

A new technique for characterization of the "End" resistance in modulation-doped FET's

Abstract: We propose a new extended "end" resistance measurement technique to determine the series source and drain resistance of a field-effect transistor. In this method the small-signal "end" resistance, defined as a derivative of the drain-source voltage, with respect to the gate current, is measured as a function of the drain current. The "end" resistance includes the contributions from the source series resistance and from the resistance related to the conducting channel under the gate. As the drain current increases, the drain side of the channel becomes more reverse biased with respect to the gate. This shifts the gate current distribution towards the source, decreasing the channel contribution to the "end" resistance. By extrapolation to infinite drain current the channel contribution can be eliminated and the series resistance of the undepleted source and drain regions can be accurately determined. The technique is applied to a long-gate modulation-doped field-effect transistor and is shown to Yield reasonable values of the series resistances.

A nonalloyed, low specific resistance Ohmic contact to n‐InP

Abstract: Nonalloyed Ohmic contacts to n‐InP have been made by sputter deposition of metals onto sputter‐etched InP surfaces. Contacts were Ohmic as deposited without intentional heating. For a 500 A sputter‐etch depth, the contact to 8×1018 cm−3 n‐InP was Ohmic up to 5×104 A cm−2 with a specific contact resistance of 4×10−7 Ω cm2 independent of sputter etch energy over the range 200 to 870 eV. To 6×1017 cm−3 n‐InP, the corresponding value was 1.8×10−6 Ω cm2. The In rich, degenerate n‐type sputter etch damaged surface is responsible for the quality of the contact, which is independent of the particular contacting metal, thus permitting the use of a barrier metal/noble metal combination. Nonalloyed contacts fabricated with Ti/Pt have low stress and excellent adhesion to InP. Even for the lowest etch energy, they are thermally stable, with no change caused by short periods up to 430 °C, or extended periods (∼100 h) at 250 °C. This contacting procedure and the resultant contact has many advantages over the usual Au:Sn...

Fretting Corrosion of Solder-Coated Electrical Contacts

Abstract: The contact resistance of fretting solder-plated contacts may rise quickly due to the accelerated formation of insulating oxides on their surfaces. This condition becomes worse with increase in length of wipe and with reduction in normal load. The solder-plated versus gold-plated system shows a more rapid rise in contact resistance which can be explained by differences in the mechanics of oxide film fracture compared to that of all-solder-coated contacts. Many fluid lubricants are able to maintain low, stable contact resistance when applied to surfaces in thick (about 20 000 A) layers, particularly on both contacts of a mated pair. The contact resistance of solder versus gold is more difficult to stabilize with lubricants than that of solder against itself, although lubricants may be useful provided that fretting distances are small. The effectiveness of coatings is due to their ability to greatly reduce oxygen arrival at the surface. When it is impractical to employ finishes other than solder on the contacts of separable connectors and when fretting can occur, it is recommended that coatings of viscous oils or greases be used to prolong contact life. An example of a useful coating is one which is obtained by immersion and withdrawal of the contacts from a mixture by weight of 20 percent polyphenylether fluid and 1 percent of microcrystalline wax in 1,1,1- trichloroethane.

Thermal conductance of pressed copper contacts at liquid helium temperatures

Abstract: Etude experimentale de la resistance thermique de contact de surfaces de cuivre avec un fini de surface de 0,4 micron entre 1,6 et 4,2°K pour des forces de pression allant jusqu'a 670 N. Etablissement d'une correlation

Silicide for contacts and interconnects

Abstract: The use of silicide materials in silicon VLSI technology is reviewed in three most important areas. The application of silicides to device contacts for lowering the contact resistance and/or controlling Schottky barrier heights. The application of silicides on top of polysilicon (Polycide) as gate materials in MOS devices for enhancing the interconnections. And, the application of silicides to all diffusion and polysilicon areas simultaneously by using self-aligned scheme (Salicide) for reducing device series resistance, and enhancing interconnects. In each case, the choice of materials, the critical processing parameters, and its technical constraints are discussed.

Basic parameter measurement and channel broadening effect in the submicrometer MOSFET

Abstract: An improved measurement technique of the basic MOSFET parameters is presented. This method is based on the measured data of two identical devices with different channel lengths. The effect of series resistance and channel length can be separated from the mobility measurement. This is the only method in which it has been proved that mobility can be measured on a short channel device. A new technique of channel length and series resistance is done by the measurement of newly defined quantity, R T K, which is the equivalent channel length that includes the effect of series resistance. Because of the success of this measurement method, a new phenomena, which we call channel broadening effect, was investigated. Its effects on the submicrometer device characteristics were investigated.

Investigation of Fretting Corrosion at Dissimilar Metal Interfaces in Socketed IC Device Applications

Abstract: Gold has traditionally been the material of choice for separable connector contact surfaces in high reliability applications With the rising cost of gold, major efforts have been focused on finding ways of reducing or altogether eliminating its use while still maintaining low and stable values of contact resistance Currently in commercial semiconductor devices, gold-plated device leads are often not available Tin and solder finishes are common Recently early intermittent contact resistance failures were found when dissimilar metal (Au/Sn and Au/solder) interfaces were used in conjunction with socketed integrated circuit (IC) devices Analysis of these failures indicated that fretting corrosion of tin had occurred To better understand the phenomena of fretting corrosion in such applications, a laboratory study was conducted on a variety of dissimilar metal (socketed device) interfaces under conditions of mild vibration The static resistance behavior of the samples was studied for a ten month period The samples were then disassembled and the contact surfaces were examined using scanning electron microscope (SEM) and scanning Auger microprobe (SAM) techniques The resistance data obtained indicated that, without lubrication, all metal interfaces except Au/ Au failed repeatedly Subsequent SEM and SAM investigations showed that all interfaces of Sn and SnPb (solder) mated to themselves or to Au failed by tin fretting corrosion, resulting in high contact resistances (>100 \Omega

Electrical contact material

Abstract: PURPOSE:To obtain a contact material equal to a conventional contact material in contact resistance and the extent of consumption and superior to it in welding resistance by sintering a powdered mixture prepared by adding In2O3 and ZnO to Ag, Ni and Cu2O. CONSTITUTION:This contact material is a sintered material obtd. by sintering a powdered mixture of Ag with Ni, Cu2O, In2O3 and ZnO, and it consists of, by weight, 10-20% Ni, 2.5-20% Cu2O, 2.5-15% In2O3, 2.5-15% ZnO and the balance Ag. The material has considerably improved welding resistance as compared to an Ag-Ni-Cu2O type material, and it is relatively superior to an Ag- CdO type material in welding resistance. It is equal to the Ag-Ni-Cu2O type material and slightly inferior to the Ag-CdO type material in contact resistance and the extent of consumption. The difference is, however, negligibly small in practical use.

Effect of Fretting on the Contact Resistance of Palladium Electroplate Having a Gold Flash, Cobalt-Gold Electroplate, and DG R156

Abstract: Contact resistance stability is a function of the contact materials, design, and conditions of connector use. The purpose of the present work was to identify certain potentially unstable materials systems when fretting occurs at the contact interface. In previous studies of fretting, it was found that variable resistance can be caused by wearout of the finish to base substrate or underplate, or to the formation of frictional polymers when platinum group contact metals are used. An evaluation of the fretting of cobalt-gold electroplate, palladium electroplate having a pure gold flash, and DG R156 was made with these materials mated to themselves or to each other. A bench test apparatus was employed with wire contacts joined in crossed rod configuration. It was found that mated to themselves and for the conditions of this test, unlubricated cobalt-gold plate performed well, although DG R156 was nearly as good and gold flashed palladium was poor. Unlubricated cobalt-gold plate mated to DG R156 was good, followed by cobalt-gold plate joined to gold flashed palladium, which was marginal. Again, for the conditions of this test, gold flashed palladium versus DG R156 was poor. Lubrication with a thin film of a polyphenylether attenuated the contact resistance changes in all cases by lowering the rate of wear of the material or, with palladium-containing systems, by causing the frictional polymers that formed to be nonadherent. The tendency of a connector to degrade by fretting depends on its design. If material systems are to be conducted which perform poorly in the bench test, it is recommended that connector tests be conducted which tend to cause micromotion at the contact interface. Procedures involving shock, vibration, and cycling temperature may be employed for this purpose.

The study of thermoluminescence using the contact method of sample heating

Abstract: A model of the thermoluminescence (TL) phenomenon using the contact method of sample heating, taking into account contact thermal resistance and temperature distribution across a sample, has been developed. It has been shown theoretically and confirmed experimentally that glow curve shift depends on sample height and contact resistance R0 in the case of imperfect contact between the sample and its heating element (HE). Temperature non-uniformity in the usual TL detectors with H

Synchronously operable electrical current switching apparatus having multiple circuit switching capability and/or reduced contact resistance

Abstract: A synchronously operable electrical current switching apparatus which may be used to switch multiple circuits, or to lower the contact resistance in one or more circuits, or both, includes a plurality of piezoelectric benders of the type having at least two electrically conductive layers separated by a piezoelectric material, and which exhibit bending motion in a direction substantially perpendicular to the plane in which the benders lie in response to an applied electrical signal. The apparatus also includes at least one pair of switching contacts, with one contact being movable and mechanically linked to at least one bender so that the bending motion causes corresponding movement of the contact and produces a change in the relative position of the contacts between open and closed positions. One of the electrically conductive layers of each bender is connected to the electrical signal by a common electrical conductor. For predetermined ones of the benders, a second one of the electrically conductive layers is connected to the electrical signal by a separate conductor, so that those benders exhibit bending motion in response to the signal. One pair of contacts may be provided for each bender, or a number of benders may be mechanically linked to one movable contact, so that the force exerted on the contact by the benders is increased.

Electrical characteristics of amorphous molybdenum-nickel contacts to silicon

Abstract: The electrical characteristics of sputtered, amorphous Mo-Ni contacts have been measured on both p- and n-type Si, as functions of composition (30, 54, and 58 at. % Mo). The contact resistivity on both p+ and n+ Si is in the 10^−6 Ω cm^2 range. The barrier height for as-deposited samples varies between φbp=0.47–0.42 V on p-type Si and between φbn=0.63–0.68 V on n-type Si, as the composition of the amorphous layer goes from Ni-rich to Mo-rich. The sum φbp+φbn always equals 1.12 V, within experimental error. After thermal treatment at 500 °C for 1/2 h, the contact resistivity changes by a factor of two or less, while the barrier height changes by at most ~0.05 V. In light of these results, the amorphous Mo-Ni film makes good ohmic contacts to silicon.

Direct W–Ti contacts to silicon

Abstract: We have studied the contact resistance in a metallization system that employs a direct contact between a tungsten–titanium alloy and shallow junctions in silicon. The values obtained in the present study are all within acceptable limits (<100 Ω μm2) for very large scale integration applications. The metal–silicon system has been subjected to moderate heat treatments, similar to those required in processing two‐level metallization schemes. No detrimental effects on the electrical properties of these contacts have been observed.

Au–Ge Ohmic Contact to n-GaAs by IR Lamp Alloying

Abstract: Utilisation d'un four a lampe I.R. pour obtenir des contacts ohmiques a resistance relativement basse. La resistance decroit quand la temperature d'alliage croit, quand les autres parametres sont fixes. Cependant la morphologie de surface tend a se deteriorer a haute temperature

Characterization of non-ohmic behavior of emitter contacts of bipolar transistors

Abstract: Emitter contacts of bipolar transistors, with silicide or polysilicon contacts, are electrically characterized by analyzing the deviation of the base current at high currents from its ideal exponential behavior. A simple theory is presented that explains the deviation of the series voltage drop from ohmic behavior, observed in some of the devices with polysilicon emitter contact, in terms of an interface with tunneling properties.

Contact resistivity of IR lamp alloyed Au-Ge metallisation on GaAs

Abstract: A rapid infra-red (IR) lamp system used to alloy Au-Ge ohmic contacts to n-GaAs has yielded specific contact resistivity a factor of 3 lower than achieved by conventional graphite strip heater. In this initial study temperatures between 430 and 445°C and alloying times of 1 to 5 s have produced the best results.