Thick Film Resistor Materials Research Articles

High–performance copper–substituted La0.5Sr0.5CoO3 ceramics as economical functional materials for thick film resistors

The demand for cost–effective materials for thick film resistors has driven the exploration of alternative functional materials to replace the traditionally used ruthenium oxides. This study investigates the synthesis and properties of copper–substituted La0.5Sr0.5CoO3 (La0.5Sr0.5Co1-xCuxO3, LSCCu, x = 0.02–0.10) ceramics as a potential economical substitute. The crystal structure, electrical performance, and sintering behavior of LSCCu with varying Cu2+ contents were characterized through solid–state reactions and first–principles simulations. The results indicate that the perovskite structure is maintained with increasing Cu2+, while the densification temperature, crystallinity, and conductivity decrease. Moreover, Cu2+ substitution reduced the DOS across the Fermi level and weakened the electron transport in LSCCu. The sample with x = 0.06 exhibited a promising conductivity of 2609 S/cm after sintering at 1150 °C, comparable to commercial thick film resistor materials. The temperature coefficient of resistance (TCR) of LSCCu showed a metallic behavior with a low value of 1148 ppm/°C for the x = 0.06 sample. Additionally, the LSCCu ceramics possess good long–term stability and frequency stability of resistance. These findings suggest that LSCCu ceramics are a promising economical material for thick film resistor applications.

The characterisation of lead-free thick-film resistors on different low temperature Co-fired ceramics substrates

Lead-free thick-film resistors were synthesised and investigated. The thick-film resistor materials with nominal sheet resistivities from 50ohm/sq. to 50kohm/sq. were prepared using combinations of two lead-free glasses with reflow temperatures at 940°C and 1240°C, respectively, and two RuO2 powders (fine-grained and coarse-grained RuO2). The thick-film resistors were printed and fired on alumina and on low temperature co-fired ceramics substrates and fired at 850°C and 950°C. The fired resistors were investigated by X-ray powder diffraction, by scanning electron microscopy and by energy dispersive X-ray analysis. The sheet resistivities, temperature coefficients of resistivity, gauge factors and noise indices were measured.

Fracture resistance evaluation of RuO 2-based thick film resistor material by in situ crack extension observation in a scanning electron microscope

Parameters for the fracture mechanics of thick film materials are scarce in the literature. One reason is that for many such materials it is very difficult to produce a bulk specimen as required for most standard tests. This paper describes an alternative method for measuring the fracture resistance of a ruthenium dioxide (RuO 2)-based thick film resistor material for electronic applications. The method is based on an in situ investigation of crack propagation in the loaded material. The investigated material is printed as a thick film on a substrate of low-temperature-co-fired-ceramic. An initial crack in the film is introduced with a Vickers indenter. The crack is subsequently loaded with a four-point bending equipment in a scanning electron microscope, which allows for in situ crack length measurement. The crack growth measurements reveal that once a certain crack length is achieved the load required to extend the crack becomes independent of the crack length. Beyond this length, the crack propagates in the so-called steady-state region, which is used in the present method to estimate the fracture resistance of the film. Both tensile stresses resulting from bending and tensile residual stresses are taken into account. Although a brittle substrate was used, the crack did not penetrate into the substrate. The measured fracture resistance of 0.69 ± 0.14 MPa√m is found to be realistic for the investigated thick film material with high silicate glass content.

The application of thick-film technology in C-MEMS

Laminated 3D structures made using low-temperature co-fired ceramic (LTCC) technology are practical for ceramic micro-electro-mechanical systems (C-MEMS). The sensors for mechanical quantities, and/or actuators, are fundamental parts of MEMS. Thick-film resistors can be used to sense the mechanical deformations, and thick-film piezoelectric materials can be used as electro-mechanical transducers in a C-MEMS structure. The integration of these thick-film materials on LTCC substrates is in some cases difficult to realise due to interactions with the rather glassy LTCC substrates. The subject of our work is an investigation of thick-film materials for electro-mechanical transducers (sensors and actuators) and their compatibility with LTCC substrates. Resistors made with commercial thick-film resistor materials for use as sensors on LTCC substrates have been investigated and evaluated. Ferroelectric ceramic materials based on solid solutions of lead zirconate titanate (PZT) with low firing temperatures around 850°C were developed for thick-film technology and evaluated on LTCC substrates.

Reducing photoelectric response when functionally trimming with lasers

A laser processing technology has been developed that dramatically reduces, or virtually eliminates, laser-induced performance shift of a device during functional trimming. While functional laser trimming of monolithic integrated circuits and thick-film hybrid circuits has been used for over 20 years to improve yields and device performance, photoelectric response in the silicon itself to the conventional wavelengths created long delays in processing and even circuit latch-up. The answer to these problems turned out to be an alternative laser wavelength which is absorbed well by both thin-film and thick-film resistor materials but is invisible to silicon-based structures. This paper describes the results obtained by using the 1.3-/spl mu/m wavelength of a diode-pumped Nd:YLF laser for trimming deposited, thin-film-on-silicon integrated circuits and the 1.3-/spl mu/m wavelength of Nd:YAG laser for screen-printed, thick-film hybrid circuits. The trimming, or tuning of thin-film-on-silicon circuits with a low-power version of the alternative wavelength has permitted both the shrinking of resistor geometries and the expansion of the types of circuits that can be functionally tuned. Secondly, the advent of higher laser power output of the same wavelength has eliminated the photoelectric response problems associated with functionally laser trimming hybrid and multichip module circuits that incorporate silicon devices in their construction. Results are given from actual production devices.

Evaluating the compatibility of some thick-film resistors and tetragonal zirconia

The gauge factor (GF) of a resistor is defined as the ratio of the relative change in resistance ( R/R) and the strain ( l/ l). The GFs of thick-film resistors are generally between 3 and 15. Due to their stability, a temperature coefficient of resistivity (TCR) under 100 × 10−6/K and their relatively low cost, strain gauges using thick-film technology offer advantages, in some applications, over both metal films (low GF, low TCR, expensive) and semiconducting elements (high GF, high TCR, inexpensive) [1, 2]. Within the same resistor series, both the GFs and the current-noise indices of thick-film resistors increase with increasing sheet resistivity [3, 4]. Therefore, in most cases 10 kohm/sq. resistors are used for the strain sensors as a useful compromise between sensitivity and relatively low noise. Pressure or force sensors can be made with resistors on deformable membranes. Thick-film resistors are printed and fired on a ceramic substrate (diaphragm), which is usually based on alumina. However, sensing elements made on partially stabilized tetragonal zirconia would exhibit some improved characteristics, as tetragonal zirconia has a higher mechanical strength and a lower modulus of elasticity than alumina [5]. This means that for the same thickness of ceramic membrane the operating range will be extended. This is shown schematically in Fig. 1a and b for a diaphragm with free edges and with fixed edges, respectively. The deformation of Al2O3-based and tetragonal ZrO2-based membranes was calculated using the following parameters: membrane radius, 20 mm; thickness, 0.635 mm; and pressure, 1.5 × 104 Pa. Some mechanical characteristics for 96% Al2O3, which is normally used for the substrates of thick-film circuits, and for tetragonal ZrO2 (Y-TZP−Y2O3 stabilized tetragonal zirconia) are summarized in Table I. Thick-film resistor materials are a complex mixture of conductive particles, glass phase and oxide modifiers. During firing, all the constituents of the resistor material react with each other and also with the ceramic substrate [6–9]. As thick-film resistors were developed for firing on alumina substrates, their compatibility and possible interactions with ZrO2 have to be evaluated. TABLE I Some characteristics of 96% Al2O3 and Y-TZP (tetragonal ZrO2)

A contribution to the debate on the resistance–temperature characteristics of thick-film resistor materials

The almost parabolic resistance–temperature characteristic shown by a thick-film resistor material (Heraeus 8241), printed on an alumina substrate, is explained in terms of thermal expansion effects in the x, y and z directions between 77 and 535 K. The piezoresistive equation is used to strip the thermal expansion contribution from the almost parabolic resistance–temperature characteristic, leaving a near linear resistance–temperature characteristic. It is suggested that the point of inflection on the resistance–temperature characteristic is primarily due to the non-linear temperature coefficient of expansivity of alumina.

Temperature sensors made by combinations of some standard thick film materials

Thick film resistors consist basically of a conductor phase, a glass phase and an organic vehicle, which burns out during high temperature processing. In most contemporary resistor compositions the conductive phase is either RuO2 or ruthenates. In addition, some other oxides are added either as temperature coefficient of resistivity (TCR) modifiers or modifiers of the temperature coefficient of expansion (TEC) of glass phase [1–3]. Most thick film resistor materials are designed for printing and firing on alumina substrates. The main requirements for their characteristics are stability, relatively narrow tolerances of sheet resistivities, and a low temperature coefficient of resistivity (TCR). During firing the resistors are at the highest temperature (typically 850◦C) a relatively short time (typically 10 min). All the constituents of the resistor material react with each other, with the conductor termination and also with the substrate [4–6]. Due to the short time at firing temperature the reactions do not reach equilibrium so that the characteristics of fired materials are, to an extent, a compromise as the consequence of this frozen unequilibrium. If thick film resistors, intended for firing on alumina substrates, interact with other materials, for example with multilayer dielectrics or NTC materials (thick film resistors with high negative and nonlinear TCRs) this could influence the resistors’ characteristics, in most cases for the worse. This is probably due to the interaction between different materials although some authors ascribe this to a different TEC [6–9]. As already mentioned the TCR of thick film resistors is aimed to be as low as possible. However, some combinations of thick film materials have been reported to result in characteristics, which allow them to be used for some temperature sensor applications (7, 10, 11). To make a useful sensor the dependence of resistivity vs. temperature ought to be high enough (high TCR) and reasonably linear. Some combinations of resistor materials and multilayer dielectrics or NTC resistors resulted in these desired characteristics. This indicates a possible use of standard thick film pastes for making fairly useful and inexpensive temperature sensors. Note, however, that data reported in this letter are not the result of a systematic search for such combinations and that the mechanisms responsible for such “behavior” were not investigated. The thick film materials, which were used in different combinations, are listed in Table I. The nominal sheet resistivities of the resistors and NTC thermistors are 1 and 10 kohm/sq. For comparison, the PTC resistor (high positive linear dependence of resistivity vs. temperature) and the platinum based conductor are also included, but note that these two materials were not fired in combination with other materials. Thick film resistors (sheet resistivities 1 kohm/sq. and 10 kohm/sq.) are part of the Du Pont HS-80 resistor series. However, the materials denoted 80× 1 and 80× 9 (for example, 1 kohm/sq. resistors 8031 and 8029) differ in the conductive phase used as well as in the fired resistor microstructure. In order to observe the microstructure the fired resistors were mounted in epoxy in a cross-sectional orientation and then polished using standard metallographic techniques. Prior to analysis in the scanning electron microscope (SEM), the sample was coated with carbon to provide electrical conductivity and to avoid charging effects. A JEOL JSM 5800 SEM equipped with an energy dispersive X-ray analyser (EDS) was used for overall microstructural and compositional analysis.

The electrical response of thick-film resistors to hydrostatic pressure and uniaxial stress between 77 and 535 K

The longitudinal gauge factor for a thick-film resistor material (Heraeus 8241) printed on an alumina substrate is found to be 12.6 at 295 K. The piezoresistive coefficient G, the unit change in resistivity per unit change in strain is 19.5, with a negative temperature coefficient of −0.00335 K −1, measured between 77 and 535 K. Thick-film resistors of different geometry were subject to hydrostatic pressure, and by the use of the piezoresistive equation based on elastic theory, and elastic modulus data for Heraeus 8241, G was calculated and found to be 19.7 at 295 K, thus validating the piezoresistive equations and the elastic modulus value for Heraeus 8241. Hydrostatic pressure tests at elevated temperature were believed to be subject to some error, 10% at 500 K, due to adiabatic heating effects. However, it is apparent that resistance change can be predicted with a knowledge of strain in the x, y and z axes. This will prove useful for thick-film strain sensor design, where the thick-film resistor is simultaneously stressed in more than one direction.

A novel method for the measurement of Young's modulus for thick-film resistor material by flexural testing of coated beams

A method for measuring the Young's modulus of thick-film resistor material is presented. 96% alumina beams were coated with Heraeus 8241 resistor material and flexural tested in the three-point bend mode using a DMA 2980 Thermal Mechanical Analyser. Young's modulus was found to decrease linearly over the 173 K to 573 K temperature range. There was no evidence of non-elastic behaviour. Modulus results for the alumina substrate relate closely to published data. This method could be used to measure the modulus of a wide variety of thick-film coatings on substrates.

Phase equilibria in the RuO2-PbO-TiO2 and RuO2-PbO-NiO systems

Contemporary thick film resistor materials are based on RuO 2 or ruthenates, mostly BieRuzO 7 [1, 2]. However, to obtain high sheet resistivity values, for example over 100 kf2 E3 -z, lead ruthenate is sometimes used because of its higher specific resistivity [3-5]. The specific resistivities of RuO2, BizRueO7 and Pb2Ru206.5 are around 40 X 10 -6 ~'~ cm, 150 X 10 -6 f2 cm and 270 x 10 -6 Q cm, respectively (the specific resistivity of metallic ruthenium is 4 x 10 -6 ~ cm) [3, 6, 7]. Around 65% of the approximately 5500 kg of ruthenium produced annually is used for the preparation of thick film resistor pastes [81. Thick film resistors, prepared only from a mixture of the conductive and glass phases, possess a relatively high positive temperature coefficient of resistance (TCR). To decrease the TCR, TCR modifiers, i.e. semiconducting oxides with negative TCR, are added. TiO2 and NiO are two of the numerous TCR modifiers [3]. While most TCR modifiers decrease TCR and at the same time increase the specific resistivity [6], TiO2, when added in small quantities, does not influence the resistivity of thick film resistor materials significantly [9]. The aim of this work was to investigate the phase equilibria in the binary RuOz-NiO and PbO-NiO systems and in the ternary R u O z P b O T i O 2 and RuOz-PbO-NiO2 systems. Both ternary systems imply possible interactions between the conductive phase (lead ruthenate) in the high sheet resistivity thick film resistors and TiO2 or NiO, when used as the TCR modifiers. It should be mentioned, however, that during firing the thick film resistors are exposed for only a relatively short time (5 to 10 rain) to the highest temperature (850 °C), and therefore the reactions probably do not reach equilibria. Also, the resistor material is a very complex system consisting of many components which interact during firing. Still, the phase equilibria indicate the direction of the reactions. Phase equilibria in the RuOz-PbO and RuO2TiO2 systems were studied by Hrovat et al. [10, 11]. The eutectic composition in the RuOz-PbO system is around 95% PbO and the eutectic temperature is 875 °C. In the RuOz-TiO2 system there is no binary compound and no liquid phase (eutectic) at temperatures below 1400 °C, i.e. the temperature at which R u O 2 decomposes to metallic ruthenium and oxygen. The solid solubility is around 16% TiO 2 in RuO2 and 13% R H O 2 in TiOz at 1350 °C and decreases with decreasing firing temperature to less than 2% below 1100 °C [12]. In the TiOz-PbO system, studied by Eisa et al. [13], the binary compound PbTiO3 melts congruently at 1250 °C. The melting point of the eutectic with a lower melting temperature on the PbO rich side (85% PbO) is at 840 °C [13, 14]. For experimental work, TiO2 (Fluka, Anatas, +99%), RuO2 (Ventron, 99.9%), PbO (Merck, 99.9%) and NiO (Riedel de Haen, +99.9%) were used. The anatase form of T i O 2 was transformed into rutile by firing at 1200 °C. The samples were mixed in ethyl alcohol, pressed into pellets and fired with intermediate grinding. During firing pellets were placed on platinum foils. The compositions of the relevant samples are shown in Figs 3 and 4. The results were evaluated by X-ray powder analysis, differential thermal analysis (DTA), scanning microscopy and energy dispersive X-ray microanalysis. The proposed nickel oxide/lead oxide phase diagram is shown in Fig. 1. There is no binary compound. The eutectic temperature and composition were determined using a standard DTA procedure. The area of the peaks on the DTA curves was measured for samples with different compositions and the eutectic composition was extrapolated from these data. The eutectic composition is around 93% PbO and the eutectic temperature is 875 °C. The ruthenium oxide/nickel oxide "phase diagram" is shown in Fig. 2. The RuO2-NiO system is "empty"; there is no binary compound and no liquid phase (eutectic) at temperatures below 1400 °C, at which R u O 2 decomposes to Ru and O2. The ternary phase diagram (subsolidus) of the RuO2-PbO-NiO system is presented in Fig. 3. A tie line exists between PbzRu206. 5 and NiO. The ternary phase diagram (subsolidus) of the RuO2-PbO-TiO2 system is shown in Fig. 4. The tie

An evaluation of some commercial thick film resistor materials for strain gauges

An evaluation of some commercial thick film resistor materials for strain gauges

Abrasion resistance of various thin film coatings on thick film resistor materials

Abstract The lifetime of resistive heaters in thermal printing heads is mainly determined by the continuous abrasion due to the mechanical contact with the sliding paper. The thick film resistor materials used are not specifically developed for abrasion resistance; therefore protective coatings were applied to the resistor surface in order to enhance the long-term performance of the printing head. The optimum abrasion resistance was achieved with a diamond-like carbon coating but, in this paper, standard coatings are also investigated, since they are interesting from an economical point of view. On two different thick film resistor materials (Tanaka and DuPont), sputtered thin films of Al2O3, SiOx and Si3N4 as well as coatings of SiO2 (substrate temperatures 250 and 350 °C) obtained by plasma-enhanced chemical vapour deposition were produced (d = 2 μm) and subjected to the polishing test described elsewhere. In comparison with the uncoated samples the abrasion of the resistors is decelerated by all coatings but to different extents. The criterion of minimum thickness decrease after 20 min of polishing revealed the chemically vapour-deposited coatings of SiO2 as the best in the present test.

Abrasion resistance of diamond-like carbon coatings on thick film resistor material

Thick film resistors used in thermal printing heads suffer from abrasion due to kilometres of paper sliding over the surface of the resistive material. In the present work we try to improve the long-term performance by the introduction of an abrasion-resistant coating in the contact area between resistor and paper. Since diamond-like carbon (i-C) is known to exhibit high hardness (3500–6000 HV) and low friction as well as good thermal conductivity and electrical insulation, we deposited 2–4 μm of carbon by ion-assisted chemical vapour deposition and tested it with respect to adhesion, crystal structure and abrasion resistance. After variation of the deposition parameters power, time and bias voltage, we discovered that the i-C layers are only adhesive to one of the two resistor pastes of different suppliers. For the abrasion resistance test, a special mechanical fixture was constructed comprising an excentrically rotating disc on a polishing wheel. The results in terms of time dependence of stylus profile depths are compared with those of uncoated samples. From this it is shown that for i-C-coated samples, the time scale for a fixed amount of thickness decrease is extended by more than one order of magnitude, representing an accordingly enlarged lifetime of the printing head.

Phase equilibria in the RuO2−Bi2O3−CdO−Nb2O5 system

Interactions between the conductive phase in thick-film resistor materials (RuO2 and Bi2Ru2O7) and TCR modifiers (CdO and Nb2O5) were studied. Phase equilibria in the RuO2−Bi2O3−CdO, RuO2−Bi2O3−Nb2O5 and RuO2−CdO−Nb2O5 systems were examined. The lines in the systems were established. The existence of a solid solution of composition Bi 2−x Cd x Ru2O 7−x/2 was confirmed.

An Investigation of Thick Film Resistor Materials' Properties During the Firing Process

Thick film resistor materials' microstructure, sheet resistivity and temperature coefficient of resistivity during the firing process in the temperature range between 350°C and 950°C were investigated. During firing the sheet resistivities at first decrease because of evaporation and/or oxidation of the organic phase and later, at higher firing temperatures, increase due to sintering of the glass phase and the rearrangement of conductive particles. The TCR is negative for firing temperatures below 450°C, due to the presence of partly pyrolysed organic vehicle, and positive for higher firing temperatures.

The Interaction Between Thick Film Conductors and Low Ohmic Resistors

For thick film resistor materials with low sheet resistivity (less than 10 ohm/□) sheet resistivity increases with decreasing resistor length, when terminated with Pd alloy based conductors. This is due to the reaction of palladium from the thick film conductor and bismuth ruthenate (Bi2Ru2O7), which is the active phase in thick film resistor material. The experimental results indicate the formation of Pd/Bi solid solution on the interface between the conductor and resistor. Because of its higher specific resistivity, this layer functions as an additional source of resistivity.

Strain Sensivity of Thick-Film Resistors

The strain sensitivity measurements of Dupont 1400Biroxr thick-film resistor materials in untrimmed and laser trimmed conditions are described. This was studied to better understand the post-laser trim drift phenomenon and evaluate the effect of packaging/assembly induced stresses on the resistors. The strain coefficient of resistance or the strain sensitivity of the untrimmed 102_ 106 \Omega / \Box$^b$ resistor materials is small, reversible, and independent of resistor geometry or fired thickness (t f ). This strain sensitivity (?i) is an intrinsic material property which is determined by the conduction mechanism in the material. The strain sensitivity of the laser trimmed · resistors is a sum of the intrinsic sensitivity (?i) and an additional term called the extrinsic strain sensitivity (\gamma ex ). The additional extrinsic contribution can be large and irreversible and depends strongly on the fired thickness (t f of the resistor. A mechanism is proposed to explain these observations. The mismatch of the thermal expansion coefficients of the resistor material and the substrate leads to high internal stresses in the thick resistor. Hence crack formation and propagation is easier which leads to increased strain sensitivity (extrinsic) and post-trim drift. It was experimentally determined that Al 2 O 3 from the substrate dissolved 6-8 µm into the thick-film resistor during resistor firing. For thin resistors the relatively higher amount of dissolved Al 2 O 3 from the substrate in the resistor glass reduces the thermal expansion mismatch between the fired resistor and substrate. Thus its strain sensitivity and drift are much smaller than a thicker fired resistor.